The integration of artificial intelligence and tourism culture industry requires that it is consumer-centered, and everything is based on the fundamental starting point of improving service quality and providing better tourism products. The article explores the impact of AI application on the cultural cognition level of tourists based on the role mechanism of AI and innovative inheritance methods in tourism culture inheritance. The level of tourists’ cultural cognition is quantified through the degree of understanding of tourism culture, the willingness to accept and disseminate tourism culture, the degree of preference and internalization of tourism culture, and the willingness to practice tourism culture, and the relevant research data are obtained through questionnaires. Then the benchmark regression model was constructed by combining the multiple linear regression model with the level of cultural cognition of tourists and the level of AI application as the explanatory variables and core explanatory variables. For every 1 percentage point increase in the level of AI application in tourism cultural heritage, the level of cultural cognition of tourists will increase by 0.419 percentage points. The application of artificial intelligence in tourism culture inheritance can expand the way of tourism culture inheritance and enhance the cultural cognition level of tourists through intelligent transmittable knowledge base.

The rapid development of the information age has prompted the exchange and sharing of information resources more and more frequently. Aiming at the problem of propagating information data in the center of data network, which is easy to cause congestion and delay, this paper uses deep neural network to research on the optimal path selection method for propagating information. A network traffic prediction model is designed based on multi-task learning and LSTM, and a dynamic multipath load balancing algorithm (FNN-LB) based on feed-forward neural network is proposed to solve the problem of scheduling and allocation of network traffic. The traffic prediction accuracy and generalization ability of the MT-LSTM model are verified, and the prediction mean square error is only 0.573%. Analyzed from several performance metrics, the FNN-LB algorithm improves the network throughput by 2.34% to 10.35% relative to other algorithms, effectively reduces the number of idle and overloaded links, as well as the average network delay and packet loss rate of the rat flow, while the first packet round-trip delay of the rat flow is reduced by more than 12.58%. Therefore, the proposed method in this paper can ensure the transmission quality of communication information data and improve the efficiency of data flow of communication information.

Under the support of education digitalization strategy, in order to adapt to the development needs of education modernization, it is necessary to strengthen the research on the application of artificial intelligence technology in the main education of Marxism. Based on this, this paper closely follows the background of the artificial intelligence era, takes Marxist theory as a guide, and builds an intelligent communication platform for Marxist education based on the deep reinforcement learning model and the new media platform, which serves as a key link in the precise communication path of Marxist education. Relying on the state representation model and decision-making model in the deep reinforcement learning algorithm, the platform realizes the intelligent recommendation and dissemination of Marxist education content. The results of the precise communication path show that the intelligent communication platform has good application recognition and perceived satisfaction, and the audience students have a strong sense of belonging and responsibility for Marxist education in the communication, and the average score of the survey on the cultivation of values such as life ideals and political attitudes is above 4.50 points. The precise communication path of Marxist education proposed in this study, as an implementable countermeasure in the new media environment, can help the audience students to establish a correct worldview, life view and values.

This paper attempts to conduct a systematic study on the constructions of quantity phrases in modern Chinese on the basis of relevant research results, drawing on the theory of constructive grammar, in order to demonstrate the mechanism of constructions of quantity phrases in modern Chinese. The study firstly researches and analyzes the matching and distribution of quantity phrases as well as the Chinese construct grammar. Then, the study is based on random forests to investigate the constructions of quantifiers. By extracting and labeling six modern Chinese corpora, the analysis is carried out using random forests. On this basis, in order to further analyze the role of the relationship between the constructions of quantifiers, this paper also invokes a multinomial logit regression model for the study. It is found that the construct variant, regional variant, verb immediately following at the end of the sentence, structure initiation, and verb prototypicality are important factors affecting the number word constructions. In addition, the probability of quantifiers was higher when the construction variants were A and D, and sentence initiation while more inclined to co-occur with quantifiers. These findings reveal constraints on quantifier constructions and demonstrate the advantages of combining machine learning methods to analyze Chinese constructions.

With the deterioration of the global economic situation and the stagnation or regression of the development of enterprises, the problem of college students’ employment and entrepreneurship has been particularly prominent in recent years, and it is also one of the key points that can not be ignored in carrying out economic construction. The article realizes the prediction of college students’ entrepreneurship and employment market trends based on ARIMA-LSTM by designing the ARIMA algorithm model and combining it with the LSTM model architecture, taking the college students’ entrepreneurship and employment data from 2010 to 2022 as the research data, and using two evaluation indexes, namely, the mean absolute percentage error (MAPE) and the root mean square error (RMSE), to predict the results. Evaluation. From the analysis results ARIMA model prediction fit is high. Comparing the prediction results of the combined model with those of the LSTM model and the ARIMA model, the comparison results show that the combined model constructed in this paper can effectively fit the linear and nonlinear intertwined and superimposed trends of the time series compared with a single model, and the relative error of prediction is smaller at 33.78, which makes the results more accurate. The combined model can help the management department related to college students’ employment and entrepreneurship make reasonable decisions and improve efficiency.

Based on the scheme of multi-objective planning, this paper conducts an in-depth investigation on the design path of interdisciplinary teaching aids for STEAM project-based learning in the context of science education. A multi-objective planning model is constructed, which includes the integration of subject knowledge, the cultivation of students’ ability and cost control, and a multi-objective genetic algorithm is introduced to solve the model. The feasibility of the design path of this paper and the enhancement of students in project-based learning are verified through real cases. Compared with the other three schemes, the interdisciplinary teaching aids production using the mathematics and electricity fusion scheme can maximize the Pareto optimality, i.e., the integration of disciplinary knowledge and the cultivation of students’ abilities are maximized, as well as the goal of minimizing the production cost. The use of this paper’s scheme to produce teaching aids and apply them in course practice can effectively enhance students’ interest in learning and course performance.

The international development of the railroad industry puts forward higher requirements for the English application ability of senior railroad students, and reinforcement learning provides new ideas for the optimization of their teaching strategies. Based on reinforcement learning, the article constructs an adaptive learning path recommendation model (RL4ALPR). The model achieves application learning of multi-scenario knowledge of English in the railroad industry through railroad English knowledge level modeling, candidate learning item screening, recommender modeling, and reward calculation. The recommended effective value of the model in this paper is 0.581 at a learning path length of 60, which is 7.79% to 13.70% higher than the control model. The model realizes accurate recommendation of English exercises for the railroad industry based on the answers to the exercises. The evaluation scores of the students in the experimental class under the intervention of the model in this paper are improved to 24.26, 17.50, and 19.64 for speaking, reading comprehension, and translation of English in the railroad, respectively. Under the model of this paper, English teaching in the higher vocational railroad industry is highly recognized by students in terms of “content setting”, “teaching quality” and “teaching effect”. And the experimental class is better than the control class in terms of the level of knowledge about English for the railroad industry, the application of English for the railroad industry in multiple scenarios, and the comprehensive ability evaluation scores of 4-5 points more than the control class.

Although China’s research on English is not as early as that of the western countries, researchers, combining the basic national conditions of China and the actual situation of the nationals’ learning of English, have been making continuous efforts in the research on the construction and application of English corpus, and have already achieved satisfactory results. In this paper, we first analyze the related contents of English corpus, and construct English corpus corpus from phonological and semantic aspects by analyzing the correlation characteristics between English corpus and semantics, according to the basic principles of corpus selection. Combining two word vector similarity measures, Jaccard similarity and edit distance, finally constitutes the final similarity calculation algorithm for English sentences. The MECNC model is constructed by integrating the joint representation and co-representation learning methods, and using edge probability to abstract the connection between two nodes. Experimentally analyze the word vector similarity of English corpus with the results of English corpus recommendation based on multilayer network representation. The correlation scores of Jaccard similarity metric in WS-SIM, WS-REL, MEN, Mtruk-771, and Simverb-3500 are 0.8069, 0.6668, 0.7389, 0.7125, respectively, 0.2769, which achieves the best results, so Jaccard captures more of the correlation between words. Experiments on link prediction task were conducted on five corpora using 3, 5, 8, and 10-fold cross-validation methods, and on the corpus CKM [245,1550], MECNC model OM3 has a maximum AUC value close to 0.94 at a cross-validation number of 8, which shows that MECNC, which is used as a guiding information for intra-layer wandering, shows a better performance.

This paper combines the development situation of blue carbon industry to formulate the multi-dimensional optimization model construction of blue carbon industry cluster path. First set the model decision variables and objective function, and divide the constraints. Select the genetic algorithm to solve the optimization model. Determine the research data sources and genetic algorithm parameters, and analyze the multidimensional optimization model. The sensitivity coefficients of each decision variable to the optimization model are 0.2~0.1, and its sensitivity level is III, which means that the selected decision variables meet the research requirements. Compared with the other three algorithms, this paper’s genetic algorithm has superiority in four performance indicators, indicating that the genetic algorithm is more suitable for optimization model solving, and finally, the optimization model of this paper is put into the actual blue carbon industry, and it is found that there is a significant difference in the effect of carbon reduction, economic gain, green environmental protection, and satisfaction before and after the optimization (P<0.05), which verifies the effectiveness of this paper's optimization for practical application, and finally, according to the optimization results, the Finally, according to the optimization results, the corresponding optimization path is proposed.

Aiming at the traditional pavement construction, there are problems such as poor construction conditions, limited quality inspection methods, backward control mode and incomplete management means. In this environment, the project in this paper (Gansu Road and Bridge Anlin Pavement Second Standard Project) uses multi-objective particle swarm optimization algorithm to establish a multi-objective machine group optimization configuration model based on quality constraints under the schedule – cost, and the first time to quote asphalt pavement to carry out the intelligent construction of unmanned machine group in Gansu Province. Analyze the intelligent unmanned machine group composed of auto-pilot paving technology and roller auto-pilot technology. Design the optimal configuration model of highway construction machine group, and use multi-objective particle swarm algorithm to design the cooperative operation of unmanned machine group. Combined with the optimal configuration of highway construction fleet problem itself, the standard particle swarm algorithm and fleet configuration model are also modified and improved. Simulate the highway pavement construction process, emphasizing the preparation of construction personnel, machinery, and management platform. The parameters of particle swarm algorithm are designed to solve the optimal construction machine fleet optimization configuration under quality constraints of duration-cost. The machine utilization and duration of scheme 2 are 15.23% and 10.96%, respectively. With the priority of duration, scheme 2 is selected as the machine fleet configuration scheme. Option 4 has the lowest machinery cost of 9.41%. With the priority to ensure the maximum profit, option 4 can be chosen as the machine swarm configuration scheme.

The development of Artificial Intelligence has renewed the direction of art history, making the relationship between technology and art a matter of great interest once again. The application of artificial intelligence in the field of fashion design brings new tools to the designers’ way of designing and displaying. This paper researches artificial intelligence technology and analyzes the application of artificial intelligence as an auxiliary means in the field of art and design, and deeply researches the way of applying artificial intelligence in fashion design as well as its advantages. It also researches the intelligent image generation problem under the fashion big data environment, adopts the method of fusing the external features of fashion images and decoupling the internal features, and provides theoretical methods and bases for the controllable generation of fashion images based on the architecture of generative adversarial network. A multiconditional information fusion generative adversarial network architecture (MCF-GAN) is proposed, and the experimental results show that the image generation performance of the model in this paper is excellent, and better performance is obtained compared with other comparative methods. And it is applied to the actual fashion design for evaluation, the designer’s evaluation in all dimensions are more than 10 points, indicating that the method in this paper has a better application value in fashion design, and provides an effective path for fashion design optimization.

Data empowers educational evaluation, and blockchain technology aids in the governance of educational evaluation data. The union of big data and blockchain technology has prompted the development of educational evaluation toward digitalization and precision of educational evaluation. This paper combines the multifaceted governance utility of blockchain technology for educational evaluation data and proposes to improve the consensus mechanism in educational evaluation information sharing. The PBFT consensus algorithm is updated with node contribution reward and punishment mechanism, the consensus nodes are selected by Fibonacci function characteristics, and the consistency protocol is optimized, so as to design a practical Byzantine fault-tolerant algorithm NCG-PBFT based on node contribution grouping, and analyze the credit value, throughput, normal block out delay, and the number of communications of NCG-PBFT consensus algorithm. Build a comprehensive education quality evaluation platform and bring in the improved PBFT consensus algorithm to test the operation performance of the comprehensive education quality evaluation platform. When the request frequency tends to be stable, the education comprehensive evaluation system of NCG-PBFT consensus algorithm is able to improve the system throughput by 74.54% compared with the PBFT algorithm, which is able to meet the performance and stability requirements of the education comprehensive quality evaluation system.

The construction of harmonious labor relations is of great significance in improving the quality of public services and promoting social harmony and stability. The study uses multi-period DID algorithm to construct a mathematical model of artificial intelligence application and labor dispute resolution, and conducts research on the influence relationship between the two. Aiming at the lack of preventive mechanisms for labor dispute resolution at present, principal component analysis and artificial neural network are used to establish a labor relations early warning model. The results show that artificial intelligence application has a significant positive impact on labor dispute resolution at the 5% level, and there is regional heterogeneity.The prediction accuracy of PCA-ANN model on labor relations in the training set and test set is 81.25% and 85.71%, respectively, which presents a good effect of early warning of labor relations, and it can be used to improve the mechanism of labor dispute resolution. Finally, based on artificial intelligence technology, the online labor dispute resolution mechanism is proposed to prevent the escalation of labor disputes and improve the effectiveness of labor dispute resolution by focusing on prevention, secondary control and subsequent resolution.

County economic development directly affects the national economy, and the county economy of Henan Province has become the economic pillar of the province. The purpose of this paper is to analyze the county-level economic development of Henan Province and its economic influencing factors by using the quantitative evaluation method. From the time series, the level of economic development of 105 county units in Henan Province from 2000-2023 is analyzed from two perspectives, absolute difference and relative difference, using the indicator of GDP per capita. Screening of factors affecting the level of economic development of counties in Henan Province is carried out from the aspects of population, resources, policies, etc., and a four-aspect indicator system is constructed, namely, human capital, government regulation, industrial level, and economic vitality. A multiple linear regression model is established, and the regression model is fitted by the regression coefficients of each influencing factor, and the fit of the regression model is examined. Each county in Henan Province is divided into three development gradients: developed, generally developed and less developed counties. Panel data regression analyses were conducted on the overall county economy of Henan Province and the influencing factors of developed, generally developed and less developed counties respectively. In the overall economic development of counties in Henan Province, the degree of influence of physical capital investment and the structure of secondary and tertiary industries on the overall differences in county economies is particularly significant. It is manifested in the fact that for every 1% increase in the investment in fixed capital of the whole society, the output of GDP per capita increases by 0.09112% accordingly. Therefore, in order to improve the differences in the economic development of counties in Henan Province, local governments and enterprises should make efforts to improve the market and investment environment and adjust the structure of secondary and tertiary industries.

The rapid growth in the scale of cross-border data flow has pushed the protection of personal information to become an important issue of global concern. This paper drafts a legal adjustment mechanism for the protection of personal information under cross-border data, and builds a data sovereignty practice system from the aspects of comprehensive strength construction and cross-border flow pilot. It utilizes civil law, criminal law and administrative law to protect personal information in cross-border data flow. Based on the numerical analysis method, the legal protection mechanism of personal information in cross-border data flow is discussed in depth. The numerical analysis results show that the probability of personal information exposure increases to about 0.35 when the ratio of malicious nodes under the legal mechanism of this paper is 0.5. The estimated accuracy of personal information protection effect increases by 65.16% to 80.52% when the enforcement strength of this paper’s mechanism is 0.7 and the sample size of companies is 300. Fixing the initial ratio of cross-border data information disclosure, the smaller the initial ratio of personal information protection, the faster the speed of personal information leakage under the legal mechanism. The investigators’ scores on the personal information risk indicators of a cross-border e-commerce platform are uniformly distributed between 1 and 2, and the sum of the overall scores is less than 10, demonstrating the effectiveness of the legal mechanism constructed in this paper on the protection of personal information.

In recent years, the scale of the electric vehicle industry and social ownership are gradually growing, in the case that the charging facilities are not yet able to meet the demand for electric vehicle charging. Aiming at the situation described above, the research of charging station siting supported by variable neighborhood genetic algorithm is proposed. Based on the principle of charging station siting, the objective function and constraints are set, and the design of charging station siting model is realized. It is found that the traditional genetic algorithm, which has the problem of poor search ability, adopts the variable neighborhood genetic algorithm to solve the model. Calculated, this paper’s algorithm in the charging demand peak period scenario, to determine the optimal charging station site selection there are four, the two objective function value of 0.94, 0.98, both in the charging peak period or the low peak period, this paper’s method compared to the traditional genetic algorithm has a higher superiority.

In recent years, the construction of education informatization has been comprehensively promoted, and the personalized learning recommendation model has brought a new direction for the development of intelligent learning platform for college English vocabulary. This study constructs the KCPE-SR model based on collaborative filtering algorithm and knowledge graph, generates and optimizes the suitable personalized learning paths for learners through the interaction between learners of college English vocabulary and resources, and develops a personalized college English vocabulary learning system based on this model. The analysis of the application effect of the system reveals that the experimental class students’ English vocabulary learning performance has been significantly improved with the help of the personalized learning system, and the students’ English vocabulary knowledge mastery (20.00 points) and vocabulary comprehensive application ability (20.49 points) have also increased. The personalized college English vocabulary learning path generation and optimization system proposed in this paper is able to achieve accurate personalized recommendation of learning resources and can meet the needs of college English vocabulary learning.

With the rapid development of virtual reality technology, its application in the field of art and design is attracting increasing attention. Based on the perspective of user demand, the article combines the Kano model to analyze user satisfaction with virtual reality technology used in modern Chinese image culture design, and finds that its landing point is the desired attribute in the first quadrant, with the Better and Worse coefficients of 0.531 and -0.141, respectively, which indicates that users expect the application of virtual reality technology in the design of image art and culture. Then the evaluation index system of VR image art and culture design is constructed, and the principal component analysis method is used to assign weights and establish the quantitative model of VR image art and culture design. The analysis shows that the weight of the sensory level is the largest 0.3780, and users attach great importance to the aesthetic experience (0.3780) and emotional experience (0.2710) of VR image art and culture design. The application of virtual reality can draw on the results of the quantitative model to design optimization strategies, combine traditional and modern elements, use the interactivity of VR to enhance artistic expression, create an immersive experience, and create more in-depth and original works of video art and culture.

This study utilizes the Apriori algorithm for association rule mining, aiming to deeply explore the intrinsic connection between college students’ physical health and sports performance. The relevant definitions of association rule mining and the application process of Apriori algorithm in this study are elaborated in detail, including data preprocessing, frequent item set generation, and association rule extraction. Through empirical analysis, various combinations of physical fitness factors affecting college students’ athletic performance and the corresponding association rules are revealed. For example, under the condition of support degree of 0.598 and confidence level of 0.709, when male students’ “stiffness upward grade” is passing, their athletic performance is also passing. By mining the correlation rules between college students’ sports performance and physical health, it provides scientific basis and targeted suggestions for physical education and students’ health management in colleges and universities.

New biofuels, as a sustainable energy alternative to traditional fossil fuels, are attracting global attention. With the increasing awareness of environmental protection and the continuous growth of energy demand, biofuels offer the possibility of reducing greenhouse gas emissions and decreasing dependence on fossil fuels. In this paper, by introducing the Wasserstein distance, which is used to describe the objective function of the GAN model, the self-attention mechanism is applied to improve the discriminator structure of the traditional WGAN-GP to achieve more efficient generation of high-quality data samples. The WGAN-GP model is used to design a new biofuel combustion scenario, and based on the combustion data, the new biofuel is prepared in the scenario. The final data generation results of the model are evaluated based on relevant evaluation indexes. It can be seen that the trend of the generated data set is consistent with the trend of the actual output value of the power station, and the interval range formed by the generated 50 sets of data can include the real data in a more complete way, with a high data coverage, and the error between the generated value and the real value is in the range of ±250-±300. The new biofuel output scenarios generated by the WGAN-GP model were utilized for EMF synthesis experiments. PTFE@ACMS-SO3H samples showed strong absorption peaks at 759cm-1 and 54cm-1 , indicating that the acidic groups-SO3H were successfully loaded on the surface of the material and the preparation of the novel biofuel was successful.

Due to differences in lifestyle, cultural capital and social support, foreign immigrants often have difficulty integrating into the ecology of their native communities and are limited in their space for development. To solve this difficulty, this paper applies the principle of regularization to obtain a logistic regression model by categorizing the factors affecting the social integration of foreign immigrants. The algorithms of log-likelihood function and negative Hessian matrix are used to optimize the parameters of the model, construct the multivariate logistic regression model based on the social integration of foreign immigrants, and analyze the regression results among various factors. The success rate of foreign immigrants’ local integration is higher when the immigration-related system is more perfect, the foreign immigrants’ cultural identification with the local area is higher, the cognitive deviation between foreign immigrants and locals is smaller, and the community integration structure is more appropriate. The highest correlation between the factors affecting the social integration of foreign immigrants is the formation of ethnic networks that are not embedded in the community by foreign immigrants who “embrace the group”, and the cognitive bias of local residents towards foreign immigrants, with a correlation coefficient of 0.9214, and the correlation coefficients of the rest of the indicators are less than 0.9. This paper classifies the migrants into “migrants of work nature” and “migrants of employment nature” in accordance with the purpose of their migratory activities. In this paper, according to the purpose of migration, migrants are classified into four categories: “work migration, study migration, investment migration and shelter migration”, and the results of the multivariate logistic regression analysis are credible.

This paper combines the project response theory to dynamically adjust and update the resources according to the learning effect and learning feedback in the process of Civic Education, so as to achieve the goal of matching the learners with the learning resources and realize efficient learning. The differential artificial raindrop algorithm based on perturbation mechanism is designed to realize the solution of multi-objective combinatorial optimization of learning resource allocation. Performance experiments show that the convergence curve of the resource allocation algorithm in this paper is gradually flattened, and the algorithm still has the evolutionary ability, the convergence curve is still decreasing, and the final characteristic difference value is also better than other BPSOR and GAR algorithms. In the case of the number of learning resources of 10, 20, 30, 50, 100, the time consumed is 207ms, 1602ms, 20506ms, 68430ms, 354687, all of which are the lowest, and the success rate is also the highest in the model. The optimal learning path is applied to an experimental class in a university for a 6-week teaching experiment, and the experimental class scores 87.2 points in the Civics test, which is much higher than the control class. This paper realizes the accurate capture of students’ Civics learning problems and the recommendation of targeted teaching resources, which can improve the quality and effect of Civics teaching.

This paper puts forward countermeasures to maximize the ecological benefits of agroforestry resources from the perspective of sustainable development of urban agroforestry resources. Taking the maximization of ecological benefits as the goal, the optimal allocation of agricultural and forestry resources is carried out. Based on the results of the optimal allocation of water resources, the planting structure of crops in the irrigation area is adjusted with the water allocation of irrigated crops as the constraint. The optimization model under the constraint of eco-efficiency objective was constructed based on the variational method and optimal control model, and the model was solved by the method of Pontryagin’s great value. After the model adjustment in this paper, the planting structure of crops in the irrigation area of city A was obviously optimized, and the planting area of potatoes accounted for the largest share of the planting area of all the crops in the irrigation area, which was about 40.61%, and the ecological benefits of potato crops were higher, which got the priority of the model, and at the same time, the model also reduced the planting area of the crops with low ecological benefits, and this reasonable allocation adjustment method satisfied the goal of maximizing ecological benefits.

Medical image segmentation is the basis for realizing intelligent medical treatment, and plays a very important clinical significance in the localization and identification of lesion areas and the formulation of surgical plans. In this paper, we investigate the image segmentation techniques based on algebraic topology methods in computer vision, and propose an image segmentation network model based on asymmetric topology preservation (ATSNet), with a view to applying it to clinical practice.The ATSNet model adopts the parallel branching structure of CNN and Transformer in the coding part, and proposes a hybrid feature aggregation strategy (HFAS) to achieve image segmentation with high efficiency. Comparison experiments on three benchmark datasets and one clinical dataset prove that the ATSNet model proposed in this paper achieves better results on different datasets, and the statistical analysis results obtained by the model are consistent with those of clinical experts (P>0.05). Meanwhile, ablation experiments demonstrate the effectiveness of the hybrid feature aggregation strategy used in this paper in improving the image segmentation performance of the model. In addition, the proposed method in the Transformer branch when the number of network layers is 3 when the overall accuracy of the largest, and the use of bilateral filtering can be better edge retention, improve the effect of image segmentation. This paper provides a technical path for the practical application of image segmentation technology.

This paper designs a multimodal data mining and learning behavior analysis model for civic education, uses improved clustering and association rule algorithms to analyze the multimodal data obtained from students, mines the basic consumption, learning and life behavior characteristics, and carries out analysis of the students’ civic situation in order to take targeted civic education measures. Aiming at the problem that traditional clustering results are greatly affected by the selection of initial clustering centers, Gaussian density function is used to determine the initial clustering centers, and Euclidean distance is replaced by density-sensitive distance to avoid sensitivity to noise and anomalies, which improves the accuracy of the clustering results of students’ behaviors. Then we use the FP-Growth association rule algorithm to improve the Apriori construction, recursively and iteratively construct the frequent pattern tree and get the final frequent item set, which improves the efficiency of student behavior data mining. After analyzing the processed student data of a university, it is found that most of the students have low interest in borrowing books, 38.22% of the students borrowed only 2.19 books on average, and the total number of times of book borrowing is only 5.4 times, and the average number of days of single borrowing is 62.3 days, and the school library needs to increase the promotion of students’ reading, which can be done through the way of offline book fairs and e-recommendations to improve students’ interest in reading books. Reading interest. The study makes a useful exploration for the informatization and intelligentization of ideological education in colleges and universities.

This paper realizes the detection of changes in physical fitness of track and field athletes in different training cycles by monitoring their sports training functions. The method used is the time series model ARMA. The athletic training function time series data were preprocessed to fit the ARMA (p,q) model, and the optimal time series fitting model was selected by examining the coefficient of determination, AIC criterion, and SC criterion. Four biochemical indexes, hemoglobin, urea, creatine kinase, and testosterone, were selected as the content of training monitoring for track and field athletes, and the ARMA(1,1) model was selected to analyze the changes in physical fitness of track and field athletes in different training cycles. Taking the hemoglobin index (HB) as an example, through the numerical simulation of the time series of HB levels of 16 track and field athletes preparing for the 15th National Games in Guangdong Province, it can be learned that the change trends of male and female track and field athletes are basically the same throughout the whole year training cycle. From the first cycle, the athletes’ Hb levels began to decrease, fell to the lowest level in the third cycle, and rebounded in the fourth cycle, reaching the highest Hb level in the winter training period.

The article firstly outlines the concept of parametric design and modeling techniques and processes, then expresses the relationship between customer needs and functional design parameters of smart home products with discrete sensitivity matrix, and introduces the fuzzy pairwise comparison method to calculate the importance of customer needs. The correlations in the dataset are mined on the Rough Set (RS) tool. AGO and IAGO are used to predict the customer demand importance and design parameter importance in the future cycle, and the parametric product family optimization model is solved by combining the non-occupancy sorting genetic algorithm with congestion distance. In this paper, the optimization ranking and core parts of the functional modules of the smart flowerpot are obtained through the parametric smart home design method, and the functional rankings of the modules are automatic irrigation function, intelligent light replenishment function, monitoring function, and human-computer interaction function; the core parts include temperature and humidity sensors, light sensors, water tanks, and single-chip microcomputer parts, and so on. In the intelligent flowerpot product family design, this paper finds that the efficiency of this paper’s optimization method increases significantly (4.23%-9.12%) and the weight of the product decreases significantly (0.1141kg-0.617kg), both in the known platform mode and in the unknown platform mode. The results of this paper are extremely important for the development and design of parametric product families based on platforms.

At the present stage, the staff of mental health center in colleges and universities have a heavy workload, fatigue work and low work efficiency, and it is urgent to explore new paths to alleviate the severe situation of mental health work in colleges and universities. In this paper, we first start from the students’ mental health assessment data and use data mining technology to analyze the students’ mental health status. Then, students’ behavioral characteristics are digitally represented to construct a prediction model of students’ mental health status based on PDNN neural network. Finally, the design method of psychological intervention system in colleges and universities is proposed. In the collected mental health assessment data, the age distribution is skewed toward the younger population, and nearly 55% of these students show a tendency toward psychological abnormality. And the average accuracy and high group recall of the prediction model of students’ mental health status established using PDNN neural network were 88.95% and 87.44%, respectively, which verified the feasibility of the modeling method in this paper. Using the psychological intervention system designed based on the method of this paper for the intervention experiments, there is no significant difference between the experimental group using the system and the control group not using the system in the factors before the intervention (p>0.05), while after the intervention the experimental group scored significantly lower than the control group in the total mental health score, interpersonal relationship sensitivity, depression and anxiety factor items. This proves the validity of the intervention system design method in this paper, which can be applied in psychological intervention methods in universities.

This paper takes 2020-2022 Shanghai main board listed companies as the research object, and empirically examines the relationship between factors such as the establishment of internal audit department and the quality of internal audit empowered by new quality productivity, with the effectiveness of the quality of internal control as the explanatory variable, the degree of separation of two powers and so on as the explanatory variable, and the corporate governance structure as the control variable to carry out a via gradient descent Logistic regression analysis optimized by gradient descent algorithm. On this basis, to address the problem that internal audit is prone to bias or falsehood due to management’s self-interest, the fsQCA method is combined to analyze the influencing factors of the choice of auditing policy (capital item or expense item) for general R&D expenditures. It is found that there is a significant positive relationship between companies with an internal audit department and a higher hierarchical level of affiliation and obtaining a standard audit opinion, and the regression relationship holds at the 0.05 level of significance, with a positive correlation with a regression coefficient of 3.745, and an OR value of 40.099. However, the effect of the company’s twofold separation of powers governance structure on the quality of the audit fails the significance test. Firms with lower profitability levels, higher R&D intensity, higher debt levels, lower tax benefits for R&D additions and deductions and lower external audit quality are more likely to capitalize R&D expenditures. The study uses cutting-edge algorithms to accurately analyze new quality productivity-enabling internal audit quality factors and innovate corporate compliance internal control paths.

This paper first describes the basic theoretical knowledge of supply chain inventory control and analyzes the existing supply chain inventory control strategies. For the relationship between safety stock and customer service level and inventory cost, the safety stock factor is used as a decision variable, and a supply chain multilevel inventory control model is established under (t,s,S) inventory replenishment strategy. Secondly, the selection operator, crossover operator and mutation operator of the traditional genetic algorithm are adaptively improved, and an improved multi-objective adaptive genetic algorithm is proposed, and this algorithm is used to solve the inventory optimization with the two objectives of supply chain inventory cost and customer service level. The simulation results of the algorithm show that the improved genetic algorithm has better convergence and the obtained Pareto optimal solution set is closer to the real optimal frontier. When the IGD value is minimized and kept constant, the convergence speed of this paper’s algorithm (34 times) is 38.18% lower than that of the traditional genetic algorithm (55 times), and the model converges faster while its Pareto solution set is more uniformly distributed. Example results also show that using the model in this paper can reduce the inventory of each node in the supply chain system and reduce the transportation cost.

In order to improve the planning efficiency of urban landscape, this paper proposes a combination design method of urban landscape construction based on grid division and a spatial optimization model of urban landscape based on particle swarm algorithm to optimize the spatial and pathway layout of urban landscape that takes both economy and ecology into account. The original landscape image was mapped with 3D remote sensing image to generate a 3D image model, and the gradient decomposition method was used for image sampling. Then the multi-dimensional dynamic feature distribution model of urban landscape was constructed, on which the urban landscape area grid was divided to realize the landscape construction combination design. Using particle position to simulate the meta-space layout results of landscape type raster images, the optimization of landscape pattern space and path is completed. The experiment proves that the algorithm in this paper reduces the influence of multiple types of perturbations on the landscape layout results, and the spatial optimization model of urban landscape pattern based on particle swarm algorithm realizes the organic coupling of quantitative and spatial optimization, which not only improves the utilization rate of the urban land, but also substantially reduces the risk index of the urban landscape, and meets the design expectations.

Aiming at the learning path recommendation problem, which is the key in personalized teaching, this paper takes the personalized learning path recommendation model as a guide, and researches and gives a method that combines the learning path recommendation model with the NFSBPSO algorithm. The learning path recommendation model based on the two-dimensional features of learners and learning resources is constructed, the population is initialized using the chaos strategy, and the optimal and worst particles in the iteration are optimized using the particle optimization strategy to obtain the optimal solution of the learning path. In order to verify the effectiveness of the personalized learning path recommendation optimization model in this paper, simulation experiments are carried out, and the teaching prototype system of a higher education institution in F city is seen as the experimental platform, and the model in this paper is applied to carry out personalized learning path recommendation practice. The first group of experimental subjects who learn according to the recommended path of this paper have an average test score of 83.6 and an average learning time of 371.7 minutes, which is better than the second group of experimental subjects who learn according to the default path. Most of the values of the recommended matching degree of personalized learning paths are between 0.64-0.9, and most of the adaptation degrees are between 0.11-0.21, which proves that the learning paths recommended by this paper’s model to the users have a high degree of accuracy and adaptability.

In the joint electrical drive system of industrial robots, the optimization and improvement of robot motion control is one of the hotspots of current research, and this paper proposes a method of optimizing the joint electrical drive control of robots using multilevel genetic algorithm. An improved PID control method is used to fuzzify the robot motion, and the robot trajectory fuzzy PID controller is optimized according to the idea of multilevel genetic algorithm. The rise time of each joint of the robot is about 5ms, 55ms, and 75ms, respectively, and the overshooting amount is smaller, and the optimized joint electrical drive system of the industrial robot is more stable in speed control in both the acceleration and deceleration phases, and shows a good dynamic control capability of the motion. It can be seen that the work in this study effectively optimizes the control performance of the industrial robot drive system using multilevel genetic algorithm.

With the continuous development of the network environment, the traffic data in the network increasingly presents high-dimensional, huge and complex characteristics, and the network threat is also increasing, the network information security threat prediction and defense mechanism plays an irreplaceable position in network security. Based on the general process of network anomaly detection, combined with deep learning algorithms, the article proposes a network anomaly detection method based on data enhancement to improve the detection accuracy of network anomaly detection model. Self-attention mechanism is embedded in the neural network framework to accomplish the improved SA-GRU network information security threat prediction method. In the performance index comparison experiments of network security posture values predicted using different prediction models, the average absolute error of the training data of the results predicted by this paper’s model is 0.00266, and the average absolute error of the test data is 0.00369, and the prediction accuracy of this paper’s model prediction is significantly higher than that of other deep learning methods. This verifies the effectiveness of the method proposed in this paper. Finally, based on the experimental results, the network information security defense mechanism is proposed from the three levels of data encryption, the use of secret keys and intrusion detection.

Pre-school education, as a key stage on the path of children’s growth, plays a vital role in their overall development. Based on the independent sample t-test method, this paper explores the gender differences in preschool education. It also takes digital media education methods as an example, and utilizes Pearson correlation coefficient, linear regression model, and systematic clustering algorithm comprehensively to quantitatively assess the impact of education methods. The results of the study showed that there were extremely significant differences (P<0.01) in the five dimensions of language ability, creativity, social interaction ability, critical thinking ability, and independent learning ability between male and female toddlers, indicating that there are significant gender differences in preschool education effectiveness. The correlation coefficients between the frequency and duration of use of digital media education methods and language skills, creativity, social interaction skills, critical thinking skills, and independent learning skills ranged from 0.47 to 0.75, with significant positive correlations, and were associated with higher scores on each of the competencies as well as higher levels of satisfaction. This paper reveals in depth the gender differences in preschool education and the important role of digital media in preschool education, which is of great value for the optimization of teaching methods in preschool education.

With the development of the Internet, public safety public opinion events have gradually become an important part of social public opinion and an important content of government response. In order to establish a standard system for evaluating the response effectiveness of the public safety public opinion incident response system, this paper, in accordance with the current status of the government’s public safety public opinion incident response system and the literature, selects four indicators, namely, serviceability, dynamics, timeliness and legitimacy, as the criterion layer of the evaluation system. Hierarchical analysis method and TOPSIS method are used to evaluate the public security public opinion incident response system. Finally, in order to verify the reasonableness of the AHP-TOPSIS method for evaluating the response effect of the public security public opinion event response system to public opinion events, 80 cases were selected, which were clustered and analyzed and the proximity scores between the samples and the positive ideal solutions were calculated, and the clustered samples were ranked to obtain the response effect ratings of the system to different events. The analysis of the data shows that timeliness has the most significant effect on the evaluation of the public security public opinion event response system, and the public security public opinion event response system responds best to government-led policy-oriented major public opinion events. The legal compliance framework can be constructed from three aspects: improving the existing laws and regulations on public security public opinion events, strictly enforcing the existing laws and regulations, and carrying out in-depth legal publicity, so as to lay the foundation for the implementation of the legal review work.

This paper first outlines the theoretical method of parametric modeling of BIM technology in building structural design, and introduces Revit and Dynamo software to ensure the interactivity and sharing of data while parameterizing the influencing factors of the building structure and automating the extraction of data. Multiple linear regression analysis and the least squares method are used to quantitatively analyze the building energy consumption and the enclosure structure, and to construct a calculation model for the overall structural energy consumption of the building. In order to maximize the comfort of the users and minimize the source consumption of Huizhou architecture, NSGA-III algorithm is introduced to design the multi-objective optimization model of Huizhou architecture. Finally, the optimization effect of the model is verified through simulation and emulation tests. The results show that: the proportion of time that the internal temperature of the antechamber of the building is in the thermal comfort zone is the highest throughout the year (38.29%), and the thermal insulation performance of the building is insufficient; the average illuminance of the compartment space does not meet the lighting requirements (52.07 Lux), and there is a lack of diversity in the lighting design; and it is necessary to optimize the thermal insulation performance of the building enclosure structure to ensure the comfort and livability of the indoor environment. In addition, between the optimal solution and the worst solution interval of the annual energy consumption value and the absolute comfort value of Huizhou architecture, the maximum difference between the energy consumption and comfort indexes is 1.051×107kwh and 0.807, respectively, which can be used for the intuitive analysis of the BIM model and the comparison of the solutions.

Scientific and efficient curriculum design and teaching activity plan is the key to the quality of teaching in higher vocational colleges and universities. Based on the principle of SPOC segmented teaching, this paper proposes a “two-line hybrid” language teaching model. Combined with the implementation process of blended teaching, a blended teaching quality evaluation index system for higher vocational colleges is constructed, which includes the dimensions of rule of law and ethics, professionalism, learning ability, skills and technology. Using the standardization principle of hierarchical analysis, the judgment matrix was constructed by comparing two by two to achieve the empowerment of the indicator system. Introducing cloud model comprehensive evaluation, combining the weights of indicators from the forward cloud generator to get the cloud diagram, and derive the evaluation results. The initial matrix is constructed according to the scores of experts, and all the items passed the consistency test, which verifies that the index system has high reliability and validity. The obtained cloud diagram shows that the cloud model parameter Ex = 5.462, in which the A rule of law ethical dimension Ex is about 5.58, closest to the medium level. This paper makes a useful exploration for actively promoting the teaching reform of higher vocational discipline courses.

This paper focuses on the characteristics of multilevel information extraction, based on the convolutional neural network model (CNN), introduces the multi-scale feature fusion and multilevel feature fusion strategy to study the multilevel information extraction method, and proposes the full convolutional neural network based on the attention mechanism and residual connection to form the multilevel information extraction model. Aiming at the gradient disappearance and saddle point problem of convolutional neural network, an activation gradient (AG) algorithm is proposed to optimize its training, which is improved to a class of activation gradient convolutional neural network (AG-CNN). The practical application effect of the multilevel information extraction model in this paper is verified by the information extraction work of net-pen culture in river-type reservoirs. Compared with the classical models such as UNet and ResUNet, the intersection and integration ratio (IoU), recall rate, precision rate, and F1 score of this paper’s model reach the highest 80.28%, 91.02%, 87.18%, and 89.03% among all the models, which possesses a stronger extraction capability. And in the multilevel information extraction experiments on Cifar100 and Caltech256 datasets, when the number of batch training data is greater than 100, the accuracy rate and performance of the experimental group basically remain stable.

In the digital campus network security construction, the existence of potential security vulnerabilities can easily cause serious threats to campus information security, resulting in significant losses. In order to prevent and mitigate the risk, the article designs a security vulnerability identification system. Firstly, the URL similarity is compared by machine learning in order to scan the vulnerability information. The SeCF embedding layer is utilized to improve the input speed and the discard layer is designed to improve the overfitting problem during the training process. Finally, TextACBL security vulnerability identification model is proposed by combining CA, 1D-CNN and BiLSTM techniques and analyzed numerically. The average recognition rate of this paper’s method is as high as 80% for 10 common security vulnerabilities, which achieves better security vulnerability recognition results compared with existing methods such as cppcheck, deepbugs, flawfinder and vuldeepecker. The experimental results verify the effectiveness and feasibility of the method in this paper, which provides ideas for safeguarding campus network security during the construction of digital campus.

With the continuous promotion of the integration of industry and education, constructing a quality evaluation system for the integration of industry and education in vocational education has become a key issue to improve the level of vocational colleges and universities’ curricula. Based on the CIPP model, the article builds a quality evaluation system of vocational education industry-teaching integration that includes 4 first-level indicators, 12 second-level indicators and 34 third-level indicators, and empirically analyzes the quality of industry-teaching integration in three higher vocational colleges, H1, H2 and H3, using the fuzzy comprehensive evaluation method through the questionnaire survey from the viewpoint of empirical application. According to the results of the fuzzy comprehensive evaluation, the quality of industry-teaching integration in H1 and H2 higher vocational colleges and universities belongs to the good level, and its comprehensive judgment value is 78.2 and 78.395 respectively.The comprehensive judgment value of the quality of industry-teaching integration in H3 higher vocational colleges and universities is 82.037, which belongs to the excellent level. The three sample higher vocational colleges have achieved outstanding results in the integration of industry and education, providing an example for the development of integration of industry and education for higher vocational colleges in the region.

Under the background of big data era, big data mining technology is widely used, through data mining technology, deeper exploration of data, discovering the relevance of data, can provide decision support for decision makers. This paper analyzes the Internet big data of college students’ employment decision-making based on big data mining technology, uses Apriori algorithm to mine the influencing factors of college students’ vocational skills generation, meanwhile applies ID3 decision tree algorithm to analyze the college students’ tendency of vocational choice, and explores the relevant factors affecting college students’ employment through correlation analysis and clustering analysis. The results of the study show that students’ personal, family and school have strong correlation with students’ vocational skills generation, which affects the improvement of students’ personal job-seeking ability. Meanwhile, the ID3 decision tree algorithm is applied to the employment consulting service for graduates to construct a career decision tree for individual college students, which visualizes their career choice paths under the influence of career values and helps them make more appropriate career choices. In addition, qualification certificates, social practice experience, academic performance, expected salary, ideal employment unit and other factors will affect the employment choice of college students, and there are individual differences among different students.

In this paper, the 3D reconstruction of the finite element model of the knee joint is completed by first generating and editing the 3D images of the martial arts movements through Mimics software. After that, Hypermesh and Abaqus software are used to pre- and post-process the properties of materials in the knee joint biomechanical finite element model. Visual 3D software and low-pass filter smoothing technique were used to obtain and process the kinematic and kinetic data of the martial arts maneuvers, and the processed data were used as boundary and loading conditions to import the data of the three martial arts maneuvers, namely, horse stance, lunge stance, and servant stance into the finite element model for calculating and comparing the biomechanical responses of the articular cartilage and meniscus. The results showed that the movement pattern of horse stance has a larger knee range of motion and a smaller peak ground reaction force compared to the lunge and servant stance movements in the martial arts maneuvers. Finite element simulations showed that the straddling knee stance produced smaller peak contact stresses on the knee cartilage and meniscus, and the peak stress area changed more during the movement. Three-dimensional finite element simulation analysis obtained four characteristic moments, namely: the first peak ground reaction force moment, the maximum external rotation-external rotation moment, the maximum dorsiflexion moment, and the second peak ground reaction force moment, which corresponded to a greater difference in ground reaction force values. Therefore, it is recommended to wear protective equipment in advance for the injury-prone areas to reduce the risk of injury before the wushu performance.

In this paper, we use a large language model for business English translation and context analysis, and propose an adaptive parameter unfreezing method based on the quantization difference between adjacent layers within the decoder to fine-tune the layers of the language model related to the translation task, and to understand the behavior of the model in the relevant layers. Then the method of combining different encoders is proposed as a dual encoding-decoding framework on top of the traditional encoding-decoding framework, which is applied to the task of context analysis in business English translation. The fine-tuning method in this paper significantly improves the text translation quality of the language model, especially in the English-X tri-lingualization, which improves the COMET and BLEU metrics by 3.22 and 2.58 points respectively. In addition, the dual encoding-decoding model proposed in this paper is applicable to the task of contextual analysis in business English translation, which significantly improves the performance of contextual analysis in business English, and the F1 value on the HIT-CDTB dataset is improved by 11.60% compared with that of Rutherford’s model. The experiment proves that the proposed method of text has made progress in the research of the task of analyzing textual contextual relations in business English.

Intellectualization of agricultural machinery can effectively improve the efficiency and quality of operations, and has an important role in promoting agricultural development. Based on AR technology, this paper introduces the key technology to build the interactive control system of agricultural machinery, uses NURBS to realize virtual agricultural machinery modeling, uses VRML technology to design a prototype of the scene environment of interactive farmland virtual reality, and details the methods of virtual modeling, virtual roaming, interactive control and collision detection in the process of system development. A four-degree-of-freedom simulation test bed is established to realize the simulation of the tractor’s attitude when walking in the field. The position information of the crop rows is extracted from the virtual scene, and the control signals are given according to this information to carry out the speed, direction and balance control of the traveling of the agricultural machine, so that the tractor travels along the crop rows. The maximum deviations of the roll angle, pitch angle and yaw angle are within 0.36°, and the maximum deviations of the elevation and traveling speed are 2.11 mm and 0.14 km/h. The simulation analysis and the physical test show the feasibility of the interactive control system of the farm machine.

At present, the evaluation of spoken English in domestic universities is affected by the evaluation teachers’ personal cognition, preference, time, energy and other factors, and it is difficult to unify the standard of oral evaluation in the implementation, and the evaluation frequency and timeliness are insufficient to meet the students’ willingness to improve their oral language. In this paper, multimodal speech recognition technology is utilized to firstly collect students’ speech signals through microphone arrays, secondly extract acoustic and linguistic features of speech, and construct multimodal feature vectors by combining visual information such as students’ lip movements and facial expressions. Subsequently, the feature vectors are input into a deep neural network model for training and recognition, fusing LSTM network with attention mechanism to analyze the speech emotion and capture the emotional changes in speech. Meanwhile, the interaction behavior in speech is analyzed by combining temporal convolutional network. Construct a deep reinforcement learning model, introduce a user item interaction layer, design a user interaction simulator, and obtain user feedback on the smart English classroom. Using multimodal speech recognition technology, the temporal waveform of classroom speech is analyzed for sound pressure value, and the normalized sound pressure value range fluctuates around [-1.5,1.5].The average recognition rate of the six emotions rises to 67.86% with the joint effect of LSTM and attention mechanism. By comparing the experiment, analyzing the difference between the experimental class and the control class before and after the reading aloud ability, the average score of the experimental class is 23.945, and the average score of the control class is 21.464, at the same time, the post-test of reading aloud ability corresponding to the experimental class and the control class P=0.005<0.05. It can be seen that the intelligent interactive classroom of English language constructed in this paper has a facilitating effect in the process of teaching reading aloud in the aspect of reading aloud ability of students The classroom can be seen that the intelligent English interactive classroom constructed in this paper has a promoting effect in the process of teaching reading aloud in terms of students' reading ability.

Image alignment is a fundamental problem in the field of computer vision and an important prerequisite for carrying out many other tasks. Firstly, the theoretical basis and realization method of image alignment as well as the process and the method of alignment are introduced to provide alignment ideas. Subsequently, an image alignment method based on the union of multi-scale features is proposed, and a new loss term is introduced to the small-scale features therein, which further improves the distinguishability of the small-scale feature descriptors while guaranteeing the invariance of the large-scale feature descriptor matching therein. Three common alignment algorithms (RIFT algorithm, HAPCG algorithm, and SAR-SIFT algorithm) are selected for stability assessment and quantitative evaluation on the dataset, and an image enhancement algorithm with histogram equalization is used to enhance the dataset. The results show that the feature stability of this paper’s method is described as 99.1%, which is better than other algorithms. Meanwhile the desired effect is achieved on the dataset.

The continuous development of digital informatization has opened the era of intelligent education in the field of education. Higher education has accumulated a huge amount of data, but it is not fully utilized, and in-depth mining and analysis of these data can reveal the students’ learning and life status and provide powerful support for teaching management. Therefore, the research of using clustering algorithm to build a hierarchical management model for English teaching is very necessary. Clustering algorithm provides an effective way for the analysis of students’ learning behavior, and for the research needs of English teaching, this paper proposes a multi-factor improved K-means clustering algorithm and compares and verifies its clustering effect. For the problem of stratified division of student groups, firstly, the clustering index system of students’ book borrowing behavior and English course learning behavior constructed is used. Then, the improved K-Means clustering algorithm is used to cluster and mine the data of each student’s behavior to discover the student groups under different behaviors, so as to realize the hierarchical clustering of students in hierarchical management. Finally, for English teaching, a student stratification management model is established from three aspects: student stratification, teaching goal stratification and teaching process stratification, which provides important decision support for student stratification determination in English teaching and provides a more rationalized management model for student management workers.

Driven by the core qualities of the Civics discipline, the requirements of curriculum reform and the needs of teaching practice, the optimization of teaching strategies has become particularly urgent in the field of Civics education. The article introduces the Markov decision-making process and basic elements of reinforcement learning, combines the Q learning algorithm with neural networks, and constructs a deep reinforcement learning model (IDQN) for multiple intelligences with collaborative scheduling. Based on this, a numerical simulation experiment of deep reinforcement learning strategy in Civics teaching was designed and implemented. Through experimental analysis: when the recommended path is 30, the IDQN model has the best learning path recommendation effect, with an IKL of 0.477. The model also has excellent performance in the allocation of teaching resources, with the accuracy, recall and F1 value of 5 tests above 90%. After the numerical simulation of Civic Education teaching, the learning interest, attitude, and motivation of students in the experimental group increased by 27.52% to 34.49%. Under this influence, combined with the learning path and resource allocation provided by the IDQN model, students in the experimental group showed a significant improvement in their learning effect, and the average score of Civic Education Theory was 6.06 points higher than that of the control group.

According to the principle, characteristics and use of CCD, this paper designs a laser beam quality measurement program using CCD as a beacon light capture detector and proposes a laser spot detection method based on CCD. The experimental steps and calculation steps for laser beam width measurement and laser power measurement by CCD camera are proposed respectively. The beacon light is used as a light source, and the spot image is processed according to the principle of gray-scale image thresholding to capture the beacon light and present it in the form of a spot on the CCD image sensor. Then, through binarization processing, the spot of the beacon light is distinguished from the background, so as to realize the spot position detection of the beacon light beam. The image data are collected to experimentally detect the laser spot position detection algorithm based on CCD image sensor proposed in this paper, respectively. In the fine-tracking spot position detection, the spot is adjusted in the range of ±9.25mrad, and the solution value is set to be determined every 0.78mrad. The spot center is kept in the range of ±9.05mrad, and centering is carried out every 0.003mrad according to the fine-centering algorithm. The experimental results show that the spots after fine centering are all within the range of ±0.78mrad, and the change trend is consistent with the simulation results, so the laser spot position detection algorithm proposed in this paper is feasible in fine tracking spot position detection.

This paper studies the 3D target modeling method under multi-view video based on deep convolutional network. Through the detailed exposition of the basic theory of 3D target modeling technology and the complete derivation of non-uniform rational B spline curve, this paper establishes technical support such as camera coordinate system for the generation of 3D target model. According to the basic structure of Deep Convolutional Network (DCNN), a DCNN network model suitable for the research scenario of this paper is established, and the model is utilized for feature extraction of images in multi-view videos. The softargmin algorithm is used to generate the parallax map for parallax estimation in the parallax calculation stage. According to the parallax map, voxel-based 3D reconstruction of the target in the multiview video is performed, and the surface reconstruction of the voxel model is performed using the Marching Cubes algorithm, and after obtaining the surface model of the target object, texture mapping is performed to enhance the realism of the model. The deep convolutional network based 3D building method in this paper can effectively realize the feature extraction of target objects in multi-view video. In 3D target modeling, the model in this paper achieves good results on both public and measured datasets, and has obvious performance superiority and generalization ability compared with other methods.

With the deepening of education modernization, improving teachers’ digital literacy has become the key to promoting the digital transformation of education. The growing demand for professionals in modern society has made the digital literacy of physical education teachers in vocational undergraduate colleges more and more important. This paper defines digital literacy and the digital literacy of vocational undergraduate teachers in turn, explores the four connotations of digital literacy, and proposes strategies to improve the digital literacy of physical education teachers in vocational undergraduate colleges. The entropy value method was used to measure the digital literacy level of physical education teachers in vocational undergraduate colleges, determine the weight of teachers’ digital literacy evaluation indexes, and select and analyze the influencing factors of teachers’ digital literacy. Pearson correlation analysis was conducted on teachers’ digital literacy and influencing factors, as well as various dimensions and influencing factors, and multiple linear regression models were constructed to analyze the improvement path. The measurement results show that in the dimension of digital awareness, the mean values of digital willingness, digital cognition, and digital will are 4.4269, 4.3484, and 4.3748, respectively, indicating that the subject vocational undergraduate physical education teachers are highly willing to learn and use digital technology resources. The correlation coefficients between the dimensions and influencing factors of digital literacy were roughly in the range of 0.4~0.7, and the P values were all < 0.01, indicating that there was a significant positive correlation between them. The path coefficients of "TS→DA", "TE→DA" and "TM→DA" were 0.0533, 0.0796 and 0.0789, which did not reach the significance level, while the other paths reached the significance level (P<0.05), indicating that there was a significant positive impact.

The research selects the documents related to the legal regulation of civil abuse of rights of action as the research object, crawls the central and local legal regulation database through Python, and uses the social network analysis method to quantitatively analyze the dimensions of the subject of legal regulation from the composition of the subject of legal regulation, the density of the network, the centrality, and the cohesive subgroups, etc. The data preprocessing is carried out on the valid data obtained. Secondly, we pre-processed the acquired valid data, extracted high-frequency words using the improved TF-IDF algorithm, and obtained the probability distribution of the subject strength of “document-subject” and “subject-phrase-item” by calculating the degree of perplexity and utilizing the LDA subject model, and obtained the probability distribution of the subject strength at different stages of civil abuse litigation. In order to obtain the themes and evolution characteristics of the legal regulation of civil abuse of rights of action at different stages, the research results are combined with the results of the study from multiple dimensions. Finally, the research results are combined to design the strategy of legal regulation of civil abuse of rights of action from multiple dimensions.

AIGC-driven development and innovation of regional education has become an important issue, and in the context of the era when AIGC technology has triggered profound changes in education, the traditional education model is experiencing a paradigm shift from the transmission of knowledge to the cultivation of innovation ability. Based on this, we first construct a model of influencing factors in the application of AIGC in course management based on the rooting theory, and verify the proposed hypotheses to provide a theoretical basis for the construction of course management optimization and multi-level decision-making model. Then we optimize the course management of foreign language teachers in colleges and universities by relying on the all-round and multi-level innovation of AIGC in the field of education, and construct a multi-level decision-making model. In the teaching application practice, the scores of the experimental class on learning interest, learning attitude and learning motivation are all higher than 75 points after practice, and the average score is 8.87 points higher than that of the control class, and the P is less than 0.05. The learning achievement of the experimental class is increased from 73.95 to 80.95 (P < 0.05), and the optimized multilevel decision-making model of this paper has a significant effect on improving students' learning interest, learning attitude, learning motivation and learning achievement, learning attitude, learning motivation as well as learning achievement, which further validates the application effectiveness of the multilevel decision-making model and provides case references for researchers of AIGC-based instructional decision-making.

This paper points out that dance movements can be regarded as the carrier of the fusion of traditional cultural elements and styles, and ethnic folk dance movements are used as the dynamic expression of inheriting traditional cultural elements and styles. Analyze the characteristics of non-negative matrix decomposition algorithm, and use the non-negative matrix decomposition algorithm to reduce the dimensionality of dance action images. In order to optimize the classification effect of the classifier on the data after dimensionality reduction, SVM algorithm is selected to form a dance movement recognition method based on matrix decomposition technology and SVM classifier. By adjusting the values of penalty factor and kernel parameter , the effectiveness of matrix decomposition algorithm for image dimensionality reduction is verified. Analyze the feasibility of the dance movement recognition method based on matrix decomposition technique and SVM classifier by selecting different data sets. Establish the dance movement evaluation model based on matrix decomposition technology, compare the evaluation model scores with the dance expert scores, and test the effect of matrix decomposition technology on the classification of dance movement styles. The Spearman’s correlation coefficient between the expert’s score and the model’s score remains above 90% in the evaluation of different dance movements. Combined with the evaluation guidance of dance experts, the dance style movement evaluation model proposed in this paper can effectively evaluate and analyze dance movement styles.

Students’ mental health problems are increasingly becoming an important part of the educational and teaching process in colleges and universities. In this paper, we collect students’ psychological data through the students’ mental health early warning system and preprocess the data through data cleaning and other data. The features of the processed mental health data are extracted using Global Chaos Bat Based Algorithm (GCBA). Construct a mental health early warning system for college students and build a decision tree model into the system for categorizing students’ mental health status. The performance of the decision tree model in this paper is verified by evaluating the finger with other models and comparing the actual classification prediction results, constructing the decision tree model with the psychological condition of interpersonal relationship of college students as an example, and conducting the visualization analysis of the decision tree. Independent sample t-test is conducted on three measures such as using the mental health early warning system constructed in this paper, and according to the results, the application of the system in this paper highlights the role of the enhancement of the level of students’ mental health and the significant improvement of depression and other psychological conditions.

Six historical building clusters in the main city of Changchun, namely People’s Street, Xinmin Street, the Palace of the Forged Manchus, the South Square, the First Automobile Manufacturing Plant, and the Kuanchengzi Station of the Middle East Railway, with a total of 2,501 historical building sites, are taken as the research objects. Using ArcGIS software, the morphology and spatial distribution pattern of the historic building clusters in the main city are discussed based on the perspective of spatial layout by invoking spatial measurement methods such as kernel density, standard deviation ellipse, algebraic geometry, and spatial correlation, etc. The results are summarized in the following table. The results show that the spatial distribution of historic buildings in the main city of Changchun is dominated by a “single center (People’s Square)” agglomeration, with a maximum kernel density of 0.9950. At the same time, the periphery also appeared to diffuse re-agglomeration, hierarchically showing a “two-axis” diffusion pattern. Among them, the main axis resides in the center of the city and extends infinitely from north to south. The secondary axis is the administrative office and center of the pseudo-Manchukuo State, which is the pseudo-Manchu Imperial Palace and Xinmin Street respectively. Finally, from the perspective of planning and design, it tries to put forward the strategy of protection and utilization, including environment, function, and culture, etc., to provide methods and bases for the holistic protection and utilization of Changchun’s historical buildings.

In this paper, the financial structure is defined as two parts, asset structure and capital structure, with respect to the mechanism of enterprise financial management on the economic performance of enterprises. The multivariate regression model of asset structure and business performance is constructed with the dimensions of asset turnover efficiency and asset structure ratio. In order to represent the operating performance, total return on assets and return on net assets are chosen as the measures of operating performance and as the explanatory variables. It is proposed that there is a linear correlation between capital structure and corporate profitability, and the linear model between capital structure and corporate operating profitability is constructed. Combined with empirical tests to verify the relationship between asset structure or capital structure on business operations. The curve estimation method of the regression model is used to analyze the effects of inventory ratio, money fund ratio and fixed asset ratio in asset structure and capital structure on the total return on assets and return on net assets. The coefficients of fixed asset turnover on performance are 0.033 and 0.025 respectively, i.e., for every increase of 1 in fixed assets, total return on assets and return on net assets increase by 0.033 and 0.025. Similarly, the fixed asset turnover, inventory turnover, and the ratio of long term financial assets are positively correlated with the performance of the enterprise. The correlation coefficients of equity ratio and state-owned ratio of enterprise capital structure are positive, which bring positive impact on enterprise operating profitability.

In the era of digital economy, the digital transformation of enterprise financial management has become an important topic that needs to be studied and solved at present. In this paper, based on analyzing the internal and external drivers on the digital transformation of enterprise financial management, the financial data of 3,498 Shanghai and Shenzhen A-share listed enterprises were obtained using Python technology. Then a fixed effect model was constructed by combining the multiple linear regression model to analyze the degree of influence of internal and external drivers on the level of digital transformation of enterprise financial management. Policy support, digital technology environment, leadership support, team awareness, and digital technology investment all have a significant effect at the 1% level on the level of digital transformation of enterprise financial management. Among them, the influence of digital technology investment is the largest, that is, every 1 percentage point increase in the enterprise’s digital technology investment in financial management, the level of digital transformation of enterprise financial management will increase by 0.204 percentage points. And there is significant regional and equity heterogeneity in the level of digital transformation of enterprise financial management, and the effect of digital transformation of financial management is stronger in the eastern region and state-owned enterprises. Therefore, in the era of digital economy, enterprises need to build a digital financial management system, strengthen cross-departmental collaboration and communication, and combine composite talents to realize the digital transformation of financial management.

In this paper, based on the knowledge graph, word vectors and other personalized path generation related technologies, based on the graph convolutional neural network to complete the construction of the English knowledge graph model, to generate a personalized English knowledge graph, drawing on the data structure in the graph, to generate a personalized learning path, in order to make the generation of personalized learning path is more reasonable, in accordance with the difficulty value of the exercises for the exercises to be sorted. Simulation experiments are designed to evaluate the difficulty level of the generated exercises. The difficulty level of most of the English exercises generated by the personalized recommendation path is concentrated in the easy and general levels, and there are a total of 2,229 questions in these two difficulty levels, so the difficulty level of the generated questions is moderate. After a period of personalized path-generated English learning, six teaching activities were carried out, and the average score of the first post-test of the experimental group was higher than that of the control group, and the Sig values were all less than 0.05, indicating that the difference in the scores of the two groups of students was significant, which side by side reflected the accuracy of personalized path-generated English teaching.

Under the accelerated process of economic globalization and the booming development of Internet technology, cross-border e-commerce, as a new mode of international trade, is becoming a new driving force for the transformation and upgrading of foreign trade with its high efficiency and convenience, low cost and high benefit. This study uses data cleaning and missing value filling methods to preprocess user behavior data and merchandise sales marketing data in cross-border e-commerce Wish platform, and discretizes user behavior data using rough set method. Then, we select the merchandise sales and user behavior as the dependent and independent variables to construct a multiple nonlinear regression model in order to analyze the influence of user data on sales in cross-border e-commerce Wish platform. The results of the multivariate nonlinear regression model show that user behavior in cross-border e-commerce Wish platform has a significant effect on merchandise sales (P=0.005243). It is also found that the sales strategy adjusted according to the regression results can improve the sales and promotion effect of enterprises in cross-border e-commerce platform. The research results of this paper enrich the theoretical and practical research on the optimization and adjustment of cross-border e-commerce enterprises’ sales strategies, provide theoretical basis and decision-making reference for the subsequent adjustment of cross-border e-commerce enterprises’ sales strategies, and help cross-border e-commerce enterprises to go global.

The continuous improvement of judicial construction has led to the emergence of a large amount of judicial data on the Internet, and how to make full use of judicial data to promote judicial openness, fairness and efficiency has become an important issue in the construction of judicial informatization. In the article, the word vector generation technique is used to obtain the annotation sequence of legal text, and then the BiLSTM model is combined with the CRF model to realize the recognition of legal text entities, and the Adam algorithm is used to optimize the training of the model, so as to improve the recognition effect of the model on legal text entities. The GCN model in the graph representation learning algorithm is introduced, and the legal text entity recognition results are used as inputs for the construction of sequential and semantic relationships, and the GCN-BiLSTM model for legal text entity relationship extraction is constructed by combining the graph representation attention network and the BiLSTM model. Based on the self-constructed legal text dataset, the validation analysis of the above model is carried out through simulation experiments.The accuracy of the BILSTM-CRF model in legal text entity recognition is 85.67%, which is 7.35% higher than that of the single LSTM-CRF model. The GCN-BiLSTM model improves its accuracy by 2.14 percentage points compared with the CasRel model in extracting the entity relationships of legal texts with multi-entity overlapping. Combined with the legal text entity relationship extraction results, the knowledge map of legal cases can be constructed to provide accurate knowledge relationship support for sorting out the veins of legal cases.

This paper measures the international trade efficiency of developing countries based on the data envelopment analysis (DEA) model, and explores the impact of digital transformation on trade efficiency differentiation using regression analysis. Relevant data of 19 developing countries, including China, are selected, and the trade efficiency at each stage is calculated separately using the three-stage DEA model in this paper. The regression model is constructed to quantitatively analyze the impact of digital transformation in the differentiation of trade efficiency of developing countries. From 2011 to 2020, the trade efficiency of each developing country shows a wave-like upward trend, and the average value of the comprehensive average efficiency in the third stage is 0.728, but only China, Peru and Colombia have a higher than average level of trade efficiency, which intuitively demonstrates the trade efficiency differentiation of developing countries. Differentiation. The overall regression results show that the elasticity coefficient of digital transformation on the international trade efficiency gap is -0.274, indicating that digital transformation has a greater effect on narrowing the trade efficiency gap than widening it. And in the subregional regression, the elasticity coefficient of digital transformation in Asia is 1.398, and the elasticity coefficients in Africa and Latin America regions are -0.953 and -0.603 respectively, and the digital transformation has significantly different impacts on trade efficiency differentiation in different regions.

This study aims to construct an effective pathway for students’ career planning and innovative industry education by integrating support vector machine algorithm with big data analysis technology. By effectively integrating multi-source data and combining the improved genetic algorithm for feature selection and extraction of student data, the support vector machine algorithm is used to conduct in-depth analysis of the data related to students’ career planning and innovation and entrepreneurship education, to provide students with accurate and personalized career and entrepreneurship guidance, and based on which, the career planning and innovation and entrepreneurship education path is constructed. Experimental analysis of the classification prediction performance of the support vector machine algorithm and comparison with other classification prediction algorithms show that the support vector machine algorithm used in this paper has the highest classification accuracy in the assessment of students’ career planning and innovation and entrepreneurship ability, and the model performance is the most stable. The results of the educational experiment show that after using the educational path proposed in this paper, the students’ satisfaction with career planning and the mean value of the assessment score of innovation and entrepreneurship ability increase by 70.89% and 170.73%, respectively. The above results fully demonstrate the effectiveness of the educational path constructed in this paper, which provides a useful reference for efficient education and teaching reform.

With the progress of the times, the scientific and reasonable planning of physical education infrastructure and resources is an important way to realize the fair development of education. Firstly, a physical education resource input-output evaluation index system and a multi-objective optimization model of resource allocation to improve the utilization rate of physical education resources are constructed for the integration of physical education resources in Wuhan private colleges. In order to achieve the effect of enhanced spatial traversal ability, the collision range of raindrops is expanded by adding the hybrid collision strategy and introducing the adaptive collision factor, and the artificial raindrop algorithm with the introduction of hybrid collision and stretching is proposed on the basis of the original artificial raindrop algorithm. The improved artificial raindrop algorithm is compared with different optimization algorithms for simulation comparison experiments and model solving. The results show that the improved artificial raindrop algorithm converges faster and with higher accuracy, while the multi-objective optimization model proposed in this paper achieves the balanced development goal of physical education resources integration and allocation in Wuhan private colleges and universities.

This paper proposes a vocal music teaching system architecture integrating multimedia technology, aiming to enhance the intuitiveness, interactivity and personalization of vocal music teaching through technical means. The system is equipped with virtual reality and voice interaction technologies to realize the digital presentation of the functional modules of the architecture. In addition, in order to evaluate the teaching effectiveness of the system, a number of evaluation indicators are designed. The fuzzy comprehensive evaluation algorithm is used as the main method, supplemented by hierarchical analysis method, to comprehensively evaluate the teaching effectiveness. Multimedia technology can improve students’ vocal ability and mastery of theoretical knowledge, in which the vocal ability is improved by 5.98% to 10.48% compared with the control class, and at the same time, there is a promotion effect on students’ positive interest in vocal learning. The students’ recognition of the system in terms of technology application, learning interaction experience, learning content and process, and teaching effect ranged from 4.077 to 4.608, with a high degree of recognition. The experts’ comprehensive evaluation of the classroom effectiveness of vocal music teaching under the system of this paper is 93.437, which is highly satisfactory. This study not only provides new technical support for vocal music teaching, but also provides a scientific assessment method for teaching evaluation, which is of great significance to improve the level of vocal music teaching.

Tang poetry, as a treasure of ancient Chinese literature, contains a wealth of natural imagery, which not only add to the picture sense of Tang poetry, but are also important carriers of the poet’s emotions and thoughts. The study outlines the nature imagery from the perspective of Tang poetry, as well as the key elements and intrinsic connections among them, and borrows k-means clustering to categorize the nature imagery groups. In addition, the study improves the principal component model by using index homogenization, homogenization, and entropy weighting, so that it achieves the best dimensionality reduction effect while guaranteeing the integrity of the data of Tang poetry text.The F1 value of SVM and KNN classifiers for classifying the natural imagery and emotional expression of Tang poetry text is more than 0.9 after dimensionality reduction of the method in this paper, which is a good classification performance. Cluster analysis divides the natural imagery of Tang poetry into astronomical imagery, landscape imagery, and animal imagery, which account for 38%, 53%, and 9%, respectively. “Old times – bright moon”, “Thinking – slanting sun”, “Looking back – west wind”, “the end of the world – west wind” natural discourse is more likely to form word clusters in the natural imagery of Tang poetry. The analysis of principal component model shows that poets are more willing to express their emotions through natural imagery, and the proportion of neutral emotional expression is 5.17% to 7.43%.

In wireless sensor networks in industrial control systems, wireless communication security is challenged due to the broadcast nature of the wireless channel, where information is more easily eavesdropped by illegal nodes on the network. The article establishes a secure communication system based on ZigBee wireless communication technology applied to wireless sensor networks in industrial control systems. In order to improve the secure communication performance of wireless sensor networks, this paper combines the Merkle tree with the μTesla protocol to establish a key management scheme for wireless communication. Then from the node trust degree, the node two-way authentication mechanism for data transmission is constructed by combining the digital signature algorithm. For the effectiveness of the secure communication mechanism of wireless sensor networks, this paper carries out data analysis through performance testing. The key management scheme takes about 17.37 μs and 3.24 μs to add and revoke a key, respectively, and the local optimal value of user time consumption is 7.26 s when the connectivity frequency is 12 min and the revocation threshold is 60. The average value of the node bidirectional authentication mechanism can reach 96.17% for the accuracy of identifying the malicious nodes in the wireless sensor network, and the bit error rate is lower than 0.5 % for the communication transmission with the mesh topology. The bit error rate is less than 0.1%. The introduction of Merkle tree and digital signature algorithms into the construction of secure communication mechanisms in wireless sensor networks can significantly improve the data transmission security performance of industrial control systems.

The emotional curve of a story is the core embodiment of the reading value of a novel, and good novels tend to have similar patterns of emotional changes, which are explored in novels by combining artificial intelligence technology. After collecting modern Chinese novel texts, Chinese word segmentation and de-duplication are performed to complete the novel text preprocessing. In view of the limitations of convolutional neural network (CNN) and recurrent neural network (RNN) in text feature extraction, this paper proposes a multi-channel convolutional and bi-directionally gated recurrent unit (BiGRU) deep learning model, Pt-MCBGA, to mine the emotional polarity in the text and analyze the emotional trend of modern Chinese novels. After a series of comparison experiments, it is demonstrated that the model performance achieves a relatively excellent performance, and the recall rate on the two datasets is improved to 83.53% and 83.69%, respectively. According to the Pt-MCBGA model, the sentiment analysis of the modern Chinese novel The Legend of the Eagle Shooting Heroes finds that the novel is dominated by positive sentiment, with both positive and negative sentiment values being relatively high, and that the characters are rich in emotions and have great emotional ups and downs.

Artificial Intelligence Generated Content (AIGC), as a computer technology mainly characterized by intelligent content generation, has caused significant changes in film and television performances and creations, and has greatly broadened the creation and development space of film and television performances. In this paper, we use motion capture technology to obtain the character movement data in film and television performances, and combine it with the skeletal motion data generation algorithm to realize the mapping of skeletal motion data. Using ResNet-122 as the backbone network, a 3D action pose estimation model is constructed by combining multi-view and multi-feature fusion networks. Based on the 3D action pose estimation sequence, the character animation generation model is constructed by combining GAN and action detail attention mechanism, and the action detail feature loss function is designed to improve the generalization ability of the animation generation model. In order to verify the effectiveness of the above method, data analysis is carried out through simulation verification. The average value of PCP3D index of the 3D action pose estimation model is 98.37, which is 0.28 percentage points higher than the sub-optimal model, and the average joint position error is only 16.07 mm. The animation generation model combining GAN and the action detail attention mechanism has the values of animation generation diversity and richness index of 5.104 and 3.997, respectively, and the animation generation diversity and richness indexes of the animation generation model combining GAN and the action detail attention mechanism are 5.104 and 3.997, respectively. 3Ds MAX software can map the generated animation sequences into the virtual space, providing assistance for optimizing the motion design of film and television performances.

This paper improves the deep residual network, proposes 3DResNet network and carries out particle swarm optimization, constitutes the PSO-3DResNet model, and designs the coal mill fault diagnosis model based on PSO-3DResNet model. The technical parameters, common fault types and fault characteristics of the coal mill are analyzed, and the relationship between the input and output parameters of the coal mill is decomposed by the residual-based condition monitoring method. Combining the numerical simulation model of coal mill and historical operation data, the typical fault condition monitoring of coal mill is constructed. Compare the classification accuracy of each model on the working state of blast furnace wind mouth, and get the anomaly detection performance of each model. The PSO-3DResNet model is analyzed to monitor the normal operating state of the coal mill, and the model is tested using the historical current and outlet wind temperature anomaly data of the coal mill. When the coal mill is in an abnormal state, the estimated residuals of the current abnormal condition fluctuate within [-16,3] with a small range, and the weighted average residuals of the current abnormal condition index remain within [-4,1].

The construction of ecological civilization is a fundamental plan related to the sustainable development of economy and society, and the dispute settlement mechanism of environmental damages is its innovative and important content. Starting from the environmental legal dispute resolution mechanism, the article analyzes the legal basis of environmental dispute mediation and the process related to pre-litigation mediation. Considering environmental legal dispute resolution as a kind of multi-objective decision-making optimization problem, a multi-objective decision-making optimization model for environmental legal disputes is constructed with the objective functions of legal effectiveness, legal applicability and subject interest rate. Then adaptive inertia weights and dynamic image Pareto solution set updating strategy are introduced to improve the multi-objective particle swarm algorithm, and combined with information entropy-based TOPSIS decision-making to realize the optimal solution selection for environmental legal dispute resolution. In the multi-objective decision-making optimization model, the improved multi-objective particle swarm algorithm achieves the optimum for a total of 15 data, and the simulation time in solving the optimal solution of the 10*10*5 case problem is only 2.314s, and the optimal solution of environmental legal dispute resolution can be obtained based on different objective functions. Environmental legal dispute resolution needs to aim at effectiveness, applicability and subject’s interests, introduce appropriate punitive damages, realize the effective connection between administrative law and criminal law, and promote the high efficiency of environmental legal dispute resolution.

Financial sharing has become an important trend in the process of enterprise development in the era of big data. This topic centers on the research of the application of cloud computing technology in financial shared services, and introduces machine learning algorithms into financial risk early warning. Financial and non-financial indicators are selected to construct the financial analysis index system, K-tuning and mean value algorithm is used to realize the risk level division, SVM algorithm is used to construct the financial risk early warning model, the parameters are continuously adjusted according to the model accuracy rate, and the model is applied to the benefit analysis. Dividing the samples into four financial risk levels of none, low, medium and high can more accurately reflect the specific situation of enterprise finance. It is proved through experiments that the financial risk prediction performance of SVM model in this paper far exceeds the logistic regression model and Gaussian plain Bayesian model, the accuracy rate is improved by 9.7% and 18.6% respectively, and the average accuracy rate in the test set reaches more than 93%. Therefore, it is feasible as well as of great research value to apply cloud computing technology in artificial intelligence to the research field of risk warning of financial shared services.

In this paper, OpenCV technology is used to produce the distribution network defects dataset, which can be used as a training set, validation set, and test set in the ratio of 6:2:2. Combining the dataset and the Transformer framework, the S-Transformer based distribution network key quality defect identification model is constructed together. At this level, the degree of equipment deterioration is fitted, the distribution network intelligent operation and maintenance optimization strategy is formulated, and the experimental method is applied to evaluate the distribution defect identification and intelligent operation and maintenance. The identification rate of S-Transformer network for the six collected distribution network equipment defects is 0.9~0.95, which accurately controls the potential dangers, and is conducive to the subsequent intelligent equipment operation and maintenance of the distribution grid and its management and control, compared to the Compared with the traditional operation and maintenance program, the operation and maintenance program in this paper can reduce the operation and maintenance time by 52 hours per month, which greatly provides the efficiency of operation and maintenance labor.

The electric power industry is an important basic industry of the country, and among all the electric power equipment, the distribution lines are directly facing the end-users, which is an important infrastructure to serve the people’s livelihood. In this study, we first transformed the distribution line engineering quality defect acceptance problem into a sequential decision-making problem, and constructed an improved reinforcement learning network model DDQN based on it, and introduced a reward function into the model to improve the intelligent adjustment ability of the intelligent bodies in the model to the data related to the distribution line, so as to improve the detection performance of the DDQN model in the distribution line engineering quality defect acceptance. The results show that the improved DDQN model is highly feasible and effective in the detection of quality defects in distribution line engineering compared with other comparative models. The simulation test of distribution line engineering quality defects found that the accuracy of the DDQN model-based distribution line engineering quality defects acceptance technique in detecting line quality defects is 95%. It is verified that the accurate and reliable distribution network line engineering quality defect acceptance technology based on the improved DDQN model is conducive to guaranteeing the safe and stable operation of the power grid system.

Science and technology innovation talents are the center of gravity of the national strategic power, which is crucial for promoting social development and scientific and technological progress. The purpose of this paper is to study the scientific and technological innovation talents of power grid enterprises, build the evaluation index system of scientific and technological innovation talents with reference to the CIPP model, select a power grid enterprise to analyze the examples, and use the fuzzy AHP model to evaluate its scientific and technological innovation talents training. Then build the role mechanism model of science and technology innovation talent cultivation, conduct regression analysis of the influence factors of science and technology innovation talent cultivation, and verify the research hypothesis. The evaluation results of the STI talents of the sample grid enterprises range from 3.6 to 4.0 points, and the evaluation grades are all good, confirming the practicality of the proposed STI talent evaluation method. Except for years of education, high focus in research field and teamwork, the selected personal factors, organizational factors and environmental factors have positive and significant effects on the quality of STI talents training. It is suggested that power grid enterprises improve and promote the development of the training system of scientific and technological innovation talents by building a training and development channel, developing a layered training model, innovating training methods as well as building a research platform.

Participatory culture, as one of the characteristics of audience performance in the current communication environment, provides imaginative space for stimulating the power of audience participation in the communication of non-heritage culture, and at the same time provides new thinking direction and inspiration for the current communication of non-heritage culture. In this paper, we mainly apply recurrent neural networks to model sequence data, and control the flow of information by adding special gating structures, so as to be able to effectively memorize and process long sequence data. Self-attention is constructed so that the network can better focus on the important parts of the sequence while ignoring the irrelevant information in the sequence. Identify non-heritage communication behaviors based on time-series data, and model non-heritage cultural communication behaviors based on the length of time the behaviors occur under the framework of situational awareness. The research experimental model is designed, relevant hypotheses are proposed, and examined through empirical evidence. The number of borrowings by visitors under 18 years old, which is the main group of visitors, declined from 737 in 2016 to 357 in 2022, with an overall decline of 51.56%, and the overall visiting behavior also showed a declining trend. In order to test the mediating role of perceived value in the relationship between interactive behavior and the communication effect of intangible cultural heritage, the benchmark model M3 model was constructed with the communication effect as the dependent variable and gender and whether the only child was the controlling variable, and the independent variables “interactive behavior” and “perceived value” were added on this basis, and the perceived value had a significant positive impact on the communication effect, β=0.485, p<0.001. The influence of interactive behavior on communication effect remains significant, at this time the β-value is 0.487 and p<0.001, the mediating role of perceived value between interactive behavior and non-heritage culture communication effect.

Under the background of carbon peak carbon neutrality, the competition among ports is not only the competition among terminal scale, throughput, and service level, but also the competition of low energy consumption and low pollution, and with the development of China’s carbon trading mechanism, the cost of carbon emission has become more and more a part of the enterprise that cannot be ignored. In this paper, the berths and shore bridges of the port are taken as the target variables, and the fuel consumption in the process of ships traveling to the port is inferred according to the assumed conditions, and the BAP model under the carbon peak carbon neutrality is deduced, and the relevant constraints are proposed. The initial population is randomly generated, and the first generation of offspring population is obtained through the selection, crossover and mutation operations of multi-objective genetic algorithm, which then continues until the end conditions of the program are satisfied. Through the empirical method, comparing the effect of carbon cost optimization scheme generated by multi-objective genetic algorithm and traditional method, the value of the objective function under the multi-objective genetic algorithm model decreased by 10.48%, the operation cost of the port decreased by 4.54%, the cost of the ship’s in-port time decreased by 24.9%, and the ship’s average in-port time decreased by 11.01%, as compared with the traditional allocation scheme. The multi-objective genetic optimization model of berth shore bridge considering carbon cost can shorten the ship’s time in port, which reduces the carbon emission from the side and achieves the promotion purpose of green port. In the model sensitivity analysis, with the increase of carbon trading price, the four indicators F, F1, F2 and T also showed linear growth, with the growth rate of 17.24%, 18.44%, 14.37% and 18.02%, respectively, and the model sensitivity is good.

This paper constructs an improved Changsha city brand image communication model on the basis of the traditional contagion model, and studies the communication effect of Changsha in the process of city brand image transformation from “online star city” to “long-term famous city”. By summarizing and analyzing the current situation of Changsha’s city brand image communication, the evaluation index system of Changsha’s city brand image communication effectiveness is constructed, and the collected evaluation index data are downscaled using principal component analysis. The support vector regression machine combined with differential evolution algorithm is used to quantitatively analyze the communication benefits of Changsha city brand image. The improved city brand image communication model in this paper has a higher accuracy compared with the traditional contagion model, and can accurately grasp the communication effect of Changsha city brand image. The average relative error of the support vector regression machine model in the quantitative analysis of communication benefits for the test samples from 2020 to 2023 is only 1.53%, which is 27.86% lower than that of the BP neural network model. It strongly demonstrates the effectiveness of the regression model selected based on the communication big data in this paper, and provides a useful reference for accurately measuring the communication benefits of Changsha’s city brand image.

This paper constructs a heterogeneous network adjacency matrix containing multiple user relationships from the connotation of professional organizations and other guides to individual behaviors covered by the take-read mechanism. The GAT algorithm is used to learn the embedding of its heterogeneous network in order to obtain the embedding vectors of user nodes, which serves as the basis for the analysis of the spreading influence of group behavior. An event recognition method based on word embedding and hierarchical cohesive clustering is proposed to analyze the recognition and evolution of social media essay-carrying behavioral events (group behavioral events) for complex networks. We point out that the distribution of group behavior affects the dynamics of information dissemination, set the adoption threshold parameter of the group, and analyze the dissemination pattern of individuals’ (individual information) participation in essay-reading behaviors. Analyze the emergence and evolution of thesis-reading behavior in social media, and explore the influence of individual’s own attributes and the attitude of neighboring nodes on the evolution of group behavioral events in complex networks. The spreading degree analysis is conducted for different relational social media bandwagon behaviors. When =0.6 and =0.8, the individual’s decision is supported by the neighbor’s viewpoints, and the users who have already participated in the paper band-reading activities have a strong attraction to the individual. When the strong degree increases to a certain value, the individual decides to participate in the dissertation banding activity, at which point the individual is no longer influenced by the external environment. The degree of the initial node for the propagation of thesis banding behavior in random networks and small-world networks is linearly and negatively correlated with the percentage of the information audience.

With the progress of modern technology, smart wearable devices have been gradually applied in the field of sports. This paper focuses on the experiments of motion recognition of the main joints realized by convolutional neural network-assisted smart wearable devices. Using smart wearable devices to feature extraction of a variety of sports signals, using GAF algorithm for sports signal image coding, and using convolutional neural network and gated recurrent unit, a CNN-GRU-based motion recognition method is proposed. Through the training and evaluation experiments of the model, it is found that the average accuracy of the CNN-GRU model training and testing is higher than 96%, and the loss value is lower than 1.5%, and the performance of sports recognition is better than that of CNN and CNN-LSTM models. Meanwhile, it presents excellent performance in the recognition of sports with different classifications and different signal durations, reaching 97.02% and 92.63% accuracy in the recognition of three and four types of sports, respectively, and the distribution of the values of human body indexes in different sports in the case study presents a certain degree of regularity, which verifies the effectiveness and feasibility of the CNN-GRU model in different application scenarios. It also shows that the method has great development potential in the field of intelligent sports.

Aiming at the current problems of low level of intelligent development and backward infrastructure in the countryside, this paper proposes a multi-objective optimization model for rural construction. According to the overall principle of optimization and the current situation of rural infrastructure construction, model assumptions, objective functions and constraints are determined. Facing the problem of calculating the optimal values of the four objective functions, NSGA-II method is chosen to solve and analyze the problem. NSGA-II algorithm is calculated in 100 iterations, and the optimal solutions of the four objective functions are 0.813, 0.943, 0.852, and 0.886, which are better than NSGA and GA algorithms in terms of performance. In order to improve the intelligent development of the countryside, two targeted development proposals are put forward.

This paper presents a hardware encryption system based on FPGA (Field-Programmable Gate Array) implementing the elliptic curve cryptography algorithm. Using FPGA as the core control unit, IoT (Internet of Things) data transmission terminals are connected to FPGA-specific external interfaces via USB/SPI interfaces. Data collected into the FPGA undergoes encryption and decryption using the FPGA’s internal hardware resources. The encrypted data is then converted into TCP/IP protocol packets and transmitted to a cloud server through the FPGA’s internal Ethernet interface circuit module. A detailed analysis and design of the hardware implementation of the elliptic curve encryption algorithm are provided. Simulation validation of the point multiplication algorithm was conducted on a computer platform with a quad-core 3.2GHz processor and 8GB of memory, using the Xilinx 5vlx20tff323 chip. The simulation results indicate that the maximum execution frequency reached 372.686 MHz, with a single point multiplication operation completed in 3328 . This significantly enhances the processing speed of the algorithm, bearing significant theoretical value and practical implications for advancing the security of the IoT ecosystem.

It has identified and presented a unified machine-learning-based malware defense system that can handle dynamic features in cyber-security challenges. This approach will leverage recent deep learning models, ensembles, and automatic generation of defense strategies to construct an effective and adaptive framework for malware detection and mitigation. These results tend to indicate significant gains compared with traditional signature-based approaches, whereby known malware detection rates reached 99.2%, and zero-day vulnerabilities reached 87.5%. The system also recorded an extra 68% reduction in false positives after one month of operations due to the adaptive learning component, while real-time detection features yielded less than a one-second response time for 95% of the threatened records. The generated defense strategy module can demonstrate a 92% success rate in the automated mitigation or containment of identified threats. The paper further presents that even with such advances, much potential still exists for optimizing resource use, enhancing model interpretability, and building more robust defenses against adversarial attacks. It enhances the area of cybersecurity and adds a new dimension by showing the capability of AI-enabled methodology to create much more efficient, agile, and flexible malware protection systems-thereby paving the way for more advanced cybersecurity innovations.

The presented article develops the detailed analysis of battery performance degradation profiles for EVs, based on operational data collected in real-world use. Based on data points gathered for 150 vehicles over 24 months, we have developed and then validated an integrated degradation prediction model incorporating several degradation mechanisms. Our study applies a novel hybrid approach that will combine physics-based principles with data-driven methods for outlining the battery aging profile. The model proposed in this paper realizes a better prediction performance of 94.3% under different operational conditions and thus proves to be considerably superior to the existing techniques. Indeed, the change of temperature and charging behavior becomes the main influence factor with the correlation coefficient of 0.85 and 0.78, respectively. After applying the proposed model to a fleet management system, there are 32.4% maintenance cost reduction and 15.8% increasing of the cycle life for batteries. It represents in detail the continuous degradation assessment and predictive maintenance framework, validated on different vehicle platforms under varying operational conditions. These findings provide valuable inputs related to the improvement of battery management strategies and life extension of a battery in electric vehicle applications, hence benefiting theoretical understanding and practical application in electric vehicle battery management.

With the reforms in competition rules and equipment by the International Table Tennis Federation (ITTF), the number of rounds in table tennis matches has increased, placing higher demands on athletes’ abilities to transition between technical and tactical offensive and defensive strategies, as well as on their physical and psychological qualities. Therefore, this study employs methods such as the strength difference evaluation, competition performance (CP), and multiple regression. Using 48 international important matches in which Player W (anonymous) participated as case studies, the study evaluates and predicts the competitive performance of W when facing athletes of different world rankings. The results indicate that in matches against athletes with different skill rankings, the phases where the technical strength difference significantly affects competition performance (CP) are the attack-after-serve phase and attack-after-receive phase, followed by the rally phase, and finally, the rally phase Ⅱ. The competitive level in serve rounds is superior to that in receive rounds. The Kruskal-Wallis test results reveal significant fluctuations in the competition performance (CP) during the rally phase Ⅱ, demonstrating highly significant differences (P < 0.01). In matches, Player W has a very low probability of winning when not holding an absolute advantage in key techniques (the first four strokes) – particularly pronounced when facing athletes ranked in the world top 20. The multiple regression model for the technical strength difference in table tennis matches plays a certain role in predicting the performance of athletes in terms of technical and tactical indicators during matches, offering a clear reflection of the effectiveness of these indicators.

With the unprecedented growth of technological advancement, effective technological transfer has become increasingly important in all dimensions of human lives. Technological transfer is a multi-level and complex ecosystem network with complicated inter-relational elements and effective fac-tors. This complexity raises the question of how to rearrange the elements of the technology transfer to improve its positive performance. To address this issue, this study aims to compare the perfor-mance and gaps of the three modes of technology transfer, which are technology entrepreneurship, technology licensing, and technology shareholding, by evaluating the three participants, which are universities/research institutes, corporations, and intermediary agencies, using related attributes. This study applies a hybrid multiple attribute decision-making (HMADM) model including the DE-MATEL for constructing the INRM, DANP for computing influence weights, modified VIKOR for evalu-ating the performances and gaps among the three technology transfer modes so that to develop sus-tainable and systemic improvement strategies. At the macro level, the results show that, the technol-ogy transfers modes receive an overall positive effect, especially universities/research institutions. At the micro level, the technology licensing has not only the highest performance but also the largest gap. According to this finding, technology licensing is the most feasible way to cater to technology transfer at the macro level from the micro level. The findings suggest decision makers pay attention to the role of universities/research institutes as the main factor influencing technology transfer ef-fectiveness. Also, they should focus on influential attributes such as researcher participation and technical collaboration ability for reducing the gap.

This paper presents an innovative optimization framework aimed at data mining in social networks, guaranteeing solutions for some of the basic challenges of computational efficiency, scalability, and accuracy. This work presents a precise approach that integrates state-of-the-art algorithmic enhancements with dynamic resource management techniques. Extensive experimental validation using real and synthetic datasets has marked the significant performance gains achieved within the framework. These results point to a 70.2% reduction in processing time and a 71.2% saving in memory consumption, all while maintaining accuracy rates above 95%. This optimization framework is very stable under different operation conditions, since its responses have always remained below 85 ms under peak loads of up to 245,000 requests per second. The empirical evaluation of the framework across diverse social networking platforms bears testimony to the fact of practical efficacy and has emerged strongly while dealing with dynamic network architecture with extensive data processing needs. The application results in significant improvement in resource utilization efficiency, providing sub-linear increase in memory consumption for maintaining consistent performance under fluctuating load scenarios. The present study extends the scope of social network analysis by proposing a scalable, efficient, and reliable optimization framework that might be of vital importance in both research and practical implementation contexts.

This study investigates the application of artificial intelligence techniques in coal mine gas monitoring and prediction, aiming to construct more efficient and accurate gas concentration prediction models to reduce the risk of gas explosion in coal mine production. Due to the limited performance of traditional prediction methods in dealing with high-dimensional and dynamic three-dimensional mining environments, this study employs a fusion model based on temporal convolutional network (TCN) and temporal generative adversarial network (TimeGAN), TCN-TimeGAN, to predict the gas concentration. The model combines the interval sampling advantage of TCN and the time series characteristics of TimeGAN, and through four processes of embedding, recovering, by generating and discriminating gas concentration time sequences, the time-dependent features of gas concentration data can be effectively captured, thus improving the prediction accuracy and timeliness. In this study, gas concentration data from September 2020 through December 2021 were used as the basis, through data cleaning and outlier processing, it is found that the gas concentration data has obvious time-dependence, which is suitable for using time series modeling. Embedding and Recovery Networks via TCN-TimeGAN Modeling, the gas concentration data are mapped to a low-dimensional feature space, a generative network is then used to generate new time series data from random noise, and the model parameters are optimized by combining the discriminative network in order to improve the quality and consistency of the generated data. In particular, to cope with the problem of gradient instability of generative adversarial networks during training, In this paper, Wasserstein distance is introduced to optimize the loss function and a gradient penalty term is added during the training process to improve the stability of model training and the realism in the samples generated. In addition, this study also explores the prediction performance of combining LSTM networks for gas concentration. The standard recurrent neural network (RNN) faces the problem of gradient vanishing in the processing of long time-dependent data, whereas the improved LSTM overcomes this problem through memory cells and gating mechanism for real-time prediction task of gas concentration. In this study, the LSTM is further extended to three-dimensional spatial input data, experiments demonstrate the prediction accuracy of the improved LSTM. To verify the validity of the model, this paper adopts a hierarchical K-fold cross-validation method, which divides the data into a training set and a validation set to ensure that the model can be generalized. Experimental results indicate that TCN-TimeGAN and improved LSTM significantly outperform traditional methods in gas concentration prediction. By analyzing the training and validation accuracies, the models showed high prediction accuracy (89.1% to 93.8%) after 20 epochs, verifying the stability and applicability of the models.
In conclusion, this study shows that the gas prediction model based on TCN-TimeGAN and improved LSTM can more accurately predict the gas concentration in coal mines, improve the intelligence level of coal mine gas monitoring, and provide technical support for safe production in coal mines. Meanwhile, the methods and models in this study also provide new thoughts and methods for time series data prediction in other fields.

This study aims to construct a corporate demand model of brand design for financial central enterprises based on grounded theory, providing a systematic theoretical framework to help financial central enterprises effectively meet their brand design needs. The method of grounded theory is used to extract relevant information from public information and in-depth interviews, and to demonstrate the rationality of the corporate demand model by combining quantitative research. Through an in-depth analysis of the subsystems of demand motivation generation, demand factor analysis and demand design realisation, a feasible theoretical framework is provided for financial central enterprises to achieve effective satisfaction of brand design corporate demands. The problem of accurately grasping requirements in the practice of brand design for financial central enterprises is solved. Useful reference and support are provided for the brand design of financial central enterprises. The innovative practice of brand design is promoted for financial central enterprises.

This paper presents an AdaBoost-DNN (Adaptive Boosting-Deep Neural Network) model for the detection of anomalous electricity consumption in power grid users. Initially, the k-means SMOTE (Synthetic Minority Oversampling Technique) technique is employed to enhance the sample set of the original anomalous consumption data to address the issue of data imbalance. Subsequently, an ensemble learning model based on AdaBoost-DNN is designed for the detection of anomalous consumption. To validate the effectiveness and superiority of the proposed AdaBoost-DNN model, comparative experiments are conducted with three traditional algorithms.

Power metering system is directly related to the production and operation level and benefit of power supply enterprises, and even has a close relationship with the national economic development and people’s life. Numerous scholars have applied deep learning to the field of fault diagnosis. Accordingly, this paper proposes a fault diagnosis method for power metering system based on stacked autoencoder (SAE) algorithm. The deep learning data samples are formed by comprehensively collecting the historical operation data of the system and the feature data provided by the third-party manufacturers. And the fault diagnosis model is designed with the SAE algorithm, and the training and optimization fine-tuning of the algorithm model is realized by BP neural network. Finally, the model is trained using explicit test data samples, and the BP neural network can reach the set accuracy after 3804 training sessions with the output error. Compared to Elman neural network iterations are less and converge faster. Using the trained fault detection model of power metering system for fault diagnosis, the model can successfully classify the faults and achieve the expected diagnostic effect.

The rapid development of digital technology and artificial intelligence has made the improvement and optimization of intelligent warehousing and automated distribution systems important topics for research in modern logistics management. With this as the background, the current study uses a systematic approach to explore critical factors, innovative ways, and implementation strategies related to these factors and their role in improving the effectiveness of intelligent warehousing systems. The study adopts a mixed-methodological approach, establishing a comprehensive evaluation index system including operational efficiency, technical performance, and economic benefits, and simultaneously verifying the implementation of the system through empirical analysis. According to the findings, the intelligent warehousing system increased the efficiency of operations in relation to order processing time and had reduced it by 71.7%, and enhanced the accuracy of picking to 99.8%. The intelligent warehouse system by use of machine learning and meta-heuristic algorithms had greatly improved the efficiency in resources utilization and energy as storage utilization increased by 19.3% while energy consumption dropped by 31.4%. A cost-benefit analysis shows that, despite the significant up-front financial investment, the system achieved a 186% return on investment over three years. This research deepens the theoretical understanding of intelligent warehousing and, at the same time, provides optimization strategies applicable to industry practice. Future research directions should focus on exploring the applications of multi-agent digital twin technology and researching how intelligent warehousing systems contribute to supply chain resilience and sustainability.

With the rapid development of artificial intelligence technology, the education sector is undergoing unprecedented changes. Personalized learning has become a key method to enhance teaching quality and learning outcomes. This paper aims to explore the application of artificial intelligence technology in personalized learning resource recommendation for students, by constructing user profiles, multidimensional models, and personalized recommendation algorithms, in order to provide precise learning resource recommendations for students. This paper proposes a personalized learning resource recommendation algorithm based on a one-dimensional convolutional neural network (1D-CNN). The algorithm first extracts local features of the sequence through convolutional operations, then uses pooling operations to extract long-term features of the sequence, and combines the two features through weighted addition to obtain the user feature information, which allows for the comprehensive extraction of both local and long-term features. Subsequently, the user feature information is multiplied by the linearly transformed sequence information to introduce temporal information. Additionally, student learning records, class performance, and incorrect question records are collected and integrated as user feature information. These user features are passed through a feedforward network to achieve nonlinear transformation and cross-dimensional interaction enhancement. Finally, the user feature vector and item feature vector are computed to obtain their relevance, which is then used for recommendations. Experimental evaluations validate the effectiveness and feasibility of the proposed method, with the aim of providing valuable insights for educational reform and development.

This study introduces a new methodology for the configuration and optimization of algorithm-driven strategies in the digital economy. It puts forward a hybrid optimization algorithm for the efficient handling of complex resource allocation problems. The proposed approach combines adaptive learning mechanisms with traditional optimization methods, showing significant improvement in convergence speed, solution accuracy, and stability of the system. Through extensive experimental validation conducted on a range of benchmark functions and real-world contexts, this algorithm proves to be outstanding at a 48.7% reduction in convergence time, as well as a solution quality enhancement by 66.4% compared with the traditional methods. Robustness analysis confirms consistent effectiveness under all diverse noise conditions and retains high success rates, even in demanding environments. This result greatly contributes to advancing algorithmic optimization approaches for digital economic systems and paves the way toward concrete applicative implementations.

Purpose – This study aimed to explore the internal structure of sustainable employability of liberal arts college students in China and develop a comprehensive scale to facilitate research on this topic and establish a theoretical framework for cultivating sustainable employability of liberal arts college students in China. Design/methodology/approach – Through theoretical derivation and open questionnaire and the Delphi method, the main dimensions of sustainable employability of liberal arts college students are explored. The components elments of each dimension are explored through a text analysis of 189 job advertisements. Through 392 questionnaires and statistical analysis techniques, a scale is developed for measuring the sustainable employability of liberal arts college students. Findings – This study found three dimensions characterizing the sustainable employability of liberal arts students in China: attribute characteristics, general ability of employment, and innovation-driven ability. Additionally, the attribute characteristics encompassed five attribute elements, the general employment ability included six, and the innovation-driven ability included four. This study also developed a 34-item scale for measuring the sustainable employability of liberal arts students that demonstrated good reliability and validity. Originality/value – This study was among the first to investigate the internal structure of sustainable employability of liberal arts students in China.

Amidst the digital economy and ESG policy frameworks, digital transformation emerges as the prime strategy for high-tech companies to enhance their corporate performance. The research investigates the impact of high-tech organizations’ digital transformation on their performance, utilizing data from A-share listed tech firms in Shanghai and Shenzhen spanning 2018 to 2022.The research indicates that digital transformation enhances the performance of high-tech firms in the context of ESG. The modulating mechanism shows that executive compensation will weaken the impact of digital transformation on enterprise performance. The intermediary mechanism demonstrates that internal control and cost effect contribute to the mediating influence on the relationship between enterprise performance and digital transformation. Each of them has successfully cleared multiple tests for robustness. At the same time, there is a certain heterogeneity in the influence of high-tech enterprises on firm performance, and the improvement effect on firm performance is significant in the east and the growth and maturity period. The research presents new empirical evidence and acts as a benchmark for understanding how digital transformation affects high-tech companies’ performance.

This research presents an innovative machine learning framework for predicting library space utilization patterns through the integration of multi-modal deep learning architectures and ensemble methodologies. The proposed system combines Long Short-Term Memory (LSTM) networks with attention mechanisms and sophisticated feature engineering techniques to achieve superior prediction accuracy while maintaining computational efficiency. The methodology encompasses three primary contributions: (1) development of a comprehensive feature extraction pipeline incorporating spatial, temporal, and environmental data streams; (2) implementation of a novel LSTM-Attention hybrid architecture with adaptive learning rate optimization; and (3) integration of ensemble learning techniques for robust prediction performance. The framework demonstrates significant improvements over existing approaches, achieving 96.8% prediction accuracy across diverse operational scenarios. Experimental validation, conducted using an extensive dataset comprising 2.1M samples collected over 33 months from multiple library facilities, demonstrates the framework’s effectiveness. The proposed model achieves a Mean Absolute Error (MAE) of 0.142 and Root Mean Square Error (RMSE) of 0.186, representing a 39.8% reduction in prediction error compared to baseline approaches. The system’s computational efficiency is evidenced by an average processing time of 45.3ms per prediction, with a memory footprint of 512MB. The research contributes to the field of intelligent library management systems by establishing a theoretically grounded and practically implementable solution for space utilization prediction. The framework’s superior performance in capturing complex spatial-temporal patterns, combined with its computational efficiency, makes it suitable for real-time applications in resource-constrained environments. These advances provide a foundation for enhanced space management strategies in modern library systems.

Traditional power load forecasting (PLF) usually uses statistical models or time series analysis methods, but they often only consider historical load data and ignore the impact of meteorological, temperature, humidity and other factors on load, resulting in inaccurate load forecasting. Moreover, traditional methods have limited real-time performance in power load data transmission and cannot respond to changing load demands in a timely manner, which limits the real-time and accuracy of PLF. Wireless networks (WN) and intelligent sensing technology (IST) were used to obtain real-time charge data, and these data were intelligently analyzed to improve prediction performance. WN and IST were used to improve the transmission efficiency and prediction accuracy of PLF. This article studied the transmission delay and integration delay of power load data in WN, and conducted experimental tests on the root mean square error (RMSE) of CER Electricity Data, REFIT Power Data, and Umass Smart Data Set datasets using an intelligent sensing algorithm based on sensors to study their predictive effect on power load. As the number of users continues to increase, the transmission delay and integration delay of power load data were also increasing. During the process of increasing the number of users from 0 to 500, the transmission delay increased from 389ms to 735ms; the integration delay increased from 568ms to 1086ms. The power load prediction algorithm based on intelligent perception technology had average prediction RMSEs of 0.2885, 0.2716, and 0.2618 for CER Electricity Data, REFIT Power Data, and Umass Smart Data Set datasets, respectively. In WN, the transmission delay and integration delay of power load data are relatively small, and with the increase of the number of users, the impact of this delay is relatively small, which can have the effect of supporting the transmission and integration of power data for a large number of users. The power load prediction algorithm based on intelligent perception technology has good prediction results for different datasets and can accurately predict power loads.

The popularity of the Internet and mobile smart terminals has changed many forms of learning, and the mobile learning model was born in this environment. As a new learning mode, mobile learning has brought certain development opportunities for college English writing teaching. In the current educational environment, many students hold various mobile devices, which also motivates them to have a strong willingness to learn on mobile. It can be said that the application of mobile learning to English writing is quite suitable. At present, the application of mobile learning in college English writing is not mature enough, and there are often a series of problems such as shortage of resources and network freezes, which also reduces students’ enthusiasm for learning. In order to further improve the fluency and maturity of the mobile learning mode, this paper has combined the wireless network to study the new mobile learning mode of college English writing. By building a mobile learning framework based on wireless network, innovating mobile learning writing content and computing learning resource categories, a new mobile learning mode of college English writing has been finally formed. The experimental results have shown that the new model has mobilized students’ enthusiasm for learning and further improved the writing efficiency. Compared with the old model, the efficiency has increased by 6.73%.

Financial digital management is a new type of financial management method. Through information technology, the financial management process has been digitized, and with the help of technical means such as data analysis and artificial intelligence, financial management automation has been achieved. Traditional financial management methods often require a large amount of manual intervention and processing, which is prone to problems such as cumbersome data processing, time-consuming and labor-intensive, and prone to errors. With the development of computer technology and network technology, digital management has become a new trend in financial management. This article analyzed the application of blockchain and cloud computing technology in financial digital management, and selected 12 enterprises as the research objects. The traditional financial management model and the financial digital management model of blockchain and cloud computing technology were respectively adopted to compare the differences in financial process efficiency, data accuracy, labor cost savings, digital management, and financial risk management between the two models. The experimental results of this article indicated that under the financial digital management mode using blockchain and cloud computing technology, the processing time of the revenue and expenditure process was 4.45 hours in terms of financial process efficiency. In terms of data accuracy, the accuracy rate of accounting was 99.7%. In terms of labor cost savings, the labor cost was 1.505 million yuan/year. In digital management, the data processing efficiency score was 92. In financial risk management, the accuracy score of risk assessment and prediction was 93, which was better than traditional financial management models. The adoption of blockchain and cloud computing technology in financial digital management can significantly improve multiple key indicators such as financial management efficiency, data accuracy, and security. This model has important value and significance for enterprises.

With the rapid development of society, the emergence of society and people’s daily life have put forward higher quality requirements for power supply. The original distribution system cannot monitor and control the circuit condition in real time. The power grid operation efficiency is low, and the loss of electric energy in the transmission process is large, resulting in the unstable power supply to users. With the development of smart grid, distribution automation has become the goal of Power System (PS) development. There are many noise data in the process of medium voltage distribution communication. In this paper, the medium voltage high-speed analog Communication Technology (CT) was applied to distribution automation. By modulating the signal and other operations, automatic power distribution can be realized, which can effectively shorten the maintenance time of fault circuits and quickly share power data resources. This paper compared the traditional medium-voltage distribution with the distribution automation based on the medium-voltage high-speed analog CT. The experimental results showed that the average power supply reliability of the traditional medium-voltage distribution and distribution automation was 88.90% and 95.56% respectively in the 10 kV voltage. In the 20 kV voltage, the average power supply reliability of traditional medium-voltage distribution and distribution automation was 90.24% and 97.04% respectively. Therefore, the application of medium-voltage high-speed analog CT in distribution network to distribution automation can effectively improve the reliability of power supply.

As a result of continuous economic development and accelerated urbanization, the agriculture development has had to change from the traditional mode of agricultural production to the modern mode of agricultural production. What kind of method can better help the development of modern agricultural production mode has become one of the current research topics that has attracted much attention. In response to this problem, the field of modern agricultural production models becomes highly relevant for research. With the in-depth study of modern agricultural production, the research on Internet of Things (IoT) technology in rural characteristic ecological agriculture (ECO) is gradually carried out, and its functional advantages are of great significance to promote the development of modern agriculture. This paper aimed to study the application of IoT technology in the development of rural characteristic ECO. The analysis and research of IoT and ECO enables it to be applied to the construction of an ecological farmland information monitoring system to address the problem of enhancing the ECO development with rural characteristics. In this paper, IoT technology, information detection and ECO were analyzed; the performance of the method was experimentally analyzed; the relevant theoretical formulas were utilized for interpretation. The outcomes demonstrated that the incidence of pests and diseases in field A using the IoT-assisted information monitoring system was 31.11% lower than that in field B, and the use of pesticides was reduced by 15.69%. It can be learned that IoT technology can meet the needs of enhancing the development level of rural characteristic ECO, and the level of agricultural development and work efficiency have been greatly improved.

Purpose – This study investigates the impact of career planning education on university students’ entrepreneurial intentions by examining the mediating roles of self-efficacy and perceived behavioral control, as well as the moderating effects of digital competency and risk propensity. Design/methodology/approach – Data were collected from 450 university students through a structured questionnaire. The research model was tested using structural equation modeling with bootstrapping procedures for mediation analysis and hierarchical regression for moderation effects. Findings – The results reveal that career planning education positively influences entrepreneurial intentions both directly ( =0.312, p<0.01) and indirectly through self-efficacy ( =0.178, p<0.01) and perceived behavioral control ( =0.133, p<0.01). Digital competency ( =0.156, p<0.01) and risk propensity ( =0.143, p<0.01) positively moderate these relationships. Practical implications – The findings suggest that higher education institutions should integrate digital skills development into career planning curricula and tailor educational approaches to students' individual characteristics to enhance entrepreneurial intentions effectively. Originality/value – This study extends the theory of planned behavior by incorporating digital competency as a crucial moderating factor and demonstrating the specific mechanisms through which career planning education influences entrepreneurial intentions in the digital era.

This paper seeks to discuss focused prototype development of self-driving, autonomous, driverless, electric cars with emphasis on subsystem advancement constituting the progress of the technology. The introduction lays special emphasis on the increased role of autonomous technology in transforming transportation by underlining its potential to enhance safety, effectiveness, and sustainability. Some technical background is provided with the definition of what an autonomous car is and its evolution timeline. Electrical vehicle current advancement is also described in detail. At last, comparative analysis of further prototype developments and subsystems with respect to their usefulness and prospects is given. This assessment serves to contribute to the present discourse on self-driving vehicle technology, and the role that these vehicles will play in on-going transport modal shift.

“Internet + medical health” service is an important direction of current medical development. The high interactivity between doctors and patients in online medical services and the massive and dynamic nature of recommended information have brought new challenges to the platform’s analysis of patient perceived trust. It is difficult for the trust transfer model to process massive information in real time. Clustering massive recommended trust is an effective solution, but data clustering is difficult to process simultaneously with the perceived recommendation trust tendency, which brings about the problem of perceived recommendation trust clustering. How to measure the trust tendency reflected in the clustering of patient perceived recommendation trust is a difficult problem faced by the trust transfer model in the context of Internet medical health services. This paper proposes a two-stage research idea of ” conversion first, clustering later”. Intuitive fuzzy sets are used to measure the fuzziness of patient perceived recommendation trust, and combined with sentiment dictionary, density clustering method and other methods to cross and penetrate each other, a patient perceived recommendation trust clustering method is constructed in the context of Internet medical health services. Finally, data experiments were conducted using the real data of the top 17 doctors on the Haodafu online platform to verify the effectiveness of the method. This method can reflect the subjectivity and ambiguity of patients’ perceived trust, provide a solution for the processing of massive recommendation information, contribute to the research on the improvement of trust transfer method system, and provide method support for predicting and analyzing the trust measurement of patients in the context of Internet medical health services. The model proposed in this paper can be used as the core of the trust-based recommendation system in Internet medical care, and help Internet medical platforms formulate precise strategies for doctors.

Upon the arrival of the sharing consumption model, guaranteeing the authenticity of products and the transparency of transactions has emerged as fundamental challenges hindering the industry’s progression. This paper explores the selection and optimization of blockchain technology implementation methods within the shared supply chain. Through a comparative analysis of non-blockchain, private blockchain, and distributed application models, our findings reveal that distributed application generates higher profits when consumers exhibit high sensitivity to blockchain performance and when such performance adheres to specific standards. Conversely, the private blockchain is more suited to customized requirements. Blockchain technology not only increases prices and transparency but also enhances consumer trust, particularly within the distributed application framework. Performance plays a crucial role in decision-making, with the private blockchain relying on corporate investment for optimization and distributed application being constrained by the limitations of the public chain. Based on these findings, it is recommended that enterprises adopt a flexible approach in selecting the most appropriate mode according to their unique needs. Additionally, they should prioritize technological innovation, strive to improve blockchain performance, consider fostering consumer trust, and promote collaborative development throughout the supply chain. These strategies will collectively contribute to the healthy and sustainable growth of the industry.