With the continuous growth of the national economy and the expansion of market demand, traditional finance has gradually turned to digital transformation, and the emergence of digital finance has brought new breakthroughs to the economy. With the continuous development of the times, in order to meet the needs of the market, digital finance and commercial investment are constantly integrated. Therefore, this paper selects the returns and risks of digital financial investment as the research topic, and predicts the investment returns of the five major online banks by analyzing the digital financial portfolio investment return prediction system. The machine learning algorithm is introduced to optimize the digital financial portfolio investment return prediction system. The investment return rate is predicted by the optimized digital financial portfolio investment return prediction system, and then compared with the actual investment return rate. The experimental results show that the predicted value of the traditional digital financial portfolio investment return prediction system for the online bank financial management return rate differs from the actual return rate by 1%-2%, while the predicted value range of the digital financial portfolio investment return prediction system for the online bank financial management return rate is the same as the fluctuation range of the actual return rate. From the experimental data, it can be seen that the digital financial portfolio investment return prediction system based on machine learning can effectively improve the prediction ability of the digital financial portfolio investment return prediction system, making the predicted value closer to the actual value and increasing the reliability of the prediction. This paper provides reference value for the optimization and improvement of the digital financial portfolio investment return prediction system and contributes to the development of digital finance. Keywords:

Visual design and security and communication network semantic technology are hot topics in recent years. As a typical representative of visual design and security and communication network semantic technology, visual design and traditional cultural element environment design have attracted many scholars’ attention. With the rapid development of modern computer technology, contemporary visual design methods have also changed. It has great development space in conveying information and aesthetic feeling. The combination of Chinese traditional culture and visual design is not only the inheritance of Chinese traditional culture, but also the trend of visual design in the new era. China’s traditional culture is a very valuable resource, which not only has a wide range of themes but also has rich connotations. It is incomparable to any other country. After thousands of years of development, China’s traditional culture has no doubt about its artistic value. As a new design method, dynamic visual design is rising with the development of market economy. With the passage of time, all aspects of human life have encountered a variety of dynamic visual design. In further exploration, it was found that the highest score of users’ visual perception of static visual design was only 6. The integration of cultural elements was also very low, and user satisfaction had not changed for a long time. The visual experience of dynamic visual design can often reach full score, but the integration degree of cultural elements is as high as 97%. High user satisfaction has laid a foundation for the inheritance of Chinese traditional culture. It can be seen that the latter has a broader development prospect and can better meet the requirements of the times. The research in this paper has important guiding significance for the application of visual design and security and communication network semantic technology.

Low earth orbit satellites can help narrow the digital divide and provide low latency and high-speed Internet connections. However, they are extremely fast and cannot stay over a single place. In addition, during the process of circumnavigating the ground, it can only be exposed to a fixed location for a short period of time. In response to the above issues, a collaborative spectrum sensing algorithm based on fuzzy integration is proposed, and an intelligent algorithm is obtained by combining alliance game algorithm. Finally, a multi-satellite low orbit satellite collaborative spectrum sensing method based on multi-satellite cooperative beamforming and intelligent algorithm is designed. The research results indicated that the correct detection probability of intelligent algorithms was positively correlated with the signal-to-noise ratio. At a signal-to-noise ratio of -11dB, the probability of correct detection reached a steady state of 1. Under strong interference conditions, when the number of participating satellites in the array was 10, the detection probability of the research method approached 1, and the optimal satellite array power utilization rate obtained was 93.4%. The above results indicate that the research method can reduce the impact of strong ground interference signals and fully tap into the spatial resources available for low orbit satellites.

With the acceleration of economic globalization and industrialization process, the processing technology of natural gas and oil is being more and more challenged and influenced. This elevates the likelihood of oil and gas leakage impacting the surrounding environment during the loading and unloading processes. To enhance the safety of oil and gas handling, an index system has been developed which is based on an improved correlation analysis algorithm and a hierarchical analysis method, as well as a correlation analysis network model of risk source. The results proved that in the night experiment, the accuracy of the correlation rule of the improved algorithm increased from 90% to 95%, and the error value was even close to 0, while the traditional algorithm fluctuated between-1 and 1. In general, the proposed evaluation system and model effectively improve the prediction and identification probability of operation safety in the oil and gas processing process.

Introduction: The physical health of students is an indispensable part of the education system. Objectives: The existing methods for evaluating physical fitness and health lack sufficient analysis of test data. Methods: Therefore, the study proposed an improved student physical health evaluation algorithm using K-means and decision tree algorithms. The initial cluster center of K-means was determined using cuckoo optimization, and the median distance of data points was used instead of the mean. The minimum Gini coefficient was used as the optimal binary value for the decision tree algorithm. Results: Experiments showed that the root mean square error of each item in the improved K-means algorithm was on average 0.056 lower than that of the fuzzy C-means algorithm. The recall rate and F1 value were on average 0.084 and 0.093 higher, respectively. The accuracy of clustering analysis was 3.3% and 5.1% higher than that of the FC-MC algorithm and SC algorithm, respectively. The decision tree algorithm approached convergence after 200 iterations, with the maximum values being 1.4%, 6.3%, and 13.5% higher than other algorithms. In the randomly selected class, the contribution of male students’ sitting forward bending, long-distance running, and pull-up projects to the total score was relatively low and need to be prioritized for improvement. Conclusion: From this, the proposed physical health evaluation method can effectively minimize the impact of extreme value data on the calculation outcomes, raise the accuracy of clustering analysis and evaluation, and accurately determine the overall and individual physical weakness items of the class.

A method for automatic recognition and anomaly detection of electrocardiogram signals based on deep neural network structure search has been proposed. Firstly, the raw ECG signals are converted into various image representations, including Gram angle field, recursive mapping, Markov transition field, etc., which enables the deep learning model to better handle these complex signal features. Meanwhile, this study utilizes convolutional neural networks for feature extraction and learns the complex relationships between features through fully connected layers. The results demonstrated that the improved method achieved a maximum accuracy of 98.5% and an average accuracy of 94.0% on the PhysioNet MIT-BIH dataset. Additionally, on the PTB dataset, the average recall rate of the improved method reached 98.4%, surpassing the performance of traditional neural networks and Canny algorithm. The experimental results indicate that the research method effectively optimizes the key patterns’ recognition ability in electrocardiogram signals and has excellent performance in detection results. This study offers a more reliable tool for early diagnosis and health management of human health diseases.

Traditional green building performance evaluation methods usually rely on static design data and a single evaluation indicator, and lack dynamic monitoring and multi-dimensional data integration, which makes it difficult to connect data at different stages, resulting in a large gap between prediction and actual performance. This paper applies digital construction technology to construct a green building performance prediction and evaluation system that integrates data across stages and dynamically, thereby improving the accuracy and reliability of the evaluation. First, according to Building Information Modeling (BIM) technology, a digital building model containing data such as energy efficiency, environmental impact, and resource utilization is constructed. By deploying Internet of Things (IoT) sensors, energy consumption, temperature, humidity, and air quality in the building are monitored in real-time. The data is transmitted to the cloud platform for centralized processing and visualization, and compared with the design data in the BIM model to provide timely feedback on performance differences. Using big data analysis, support vector machine (SVM), and particle swarm optimization (PSO), data from the full life cycle are analyzed to predict building performance and optimize it. Finally, the LCA (Life Cycle Assessment) method is utilized to comprehensively consider the environmental impacts of buildings such as carbon footprint and resource consumption, and combined with multi-objective decision analysis tools to optimize the green building design and operation plan. The experiment shows that the energy efficiency comparison difference of green buildings is within 10kWh/m², and all indicators are accurately predicted, providing a scientific basis for the design and operation of green buildings.

Power distribution and automation stand to gain greatly from the widespread use of connected devices made possible by the advent of the Internet of Things (IoT). The reliability of a SCADA (Supervisory Control and Data Acquisition) system has been extensively shown in the substation environment. The fundamental problem with distribution automation is the lack of distribution-side management, mostly from the field’s geographically dispersed workforce. As a result of their dispersed locations, there has been inadequate tracking of their distribution channels. The smart grid is a power system incorporating evolutionary computing, bidirectional communication, two-way electrical flow, and real-time monitoring. Hence, this paper Internet of Things based Integrated Smart Grid Distribution Management System (IoT-ISGDMS) with fog computing has been presented that addresses issues such as power quality assurance, pole transformers health, and customer consumption in distribution automation. In this paper IoT-ISGDMS uses fog computing which analyzes distribution automation in real-time, making this possible. As a first step, IoT-ISGDMS uses intelligent acceptance systems (IAS) to improve coordination between smart grids and other electronic infrastructures. The second step is to perform comprehensive data analysis, automatically recognize any possible problems, and offer more intelligent fault detection and diagnosis to cut down on time and money spent on maintenance. In conclusion, as the degree of system intelligence rises safeguarding data privacy and the safety of networks will become critical priority areas.

In higher education, physical education courses and ideological and political education are often carried out independently, lacking effective integration and interaction. This paper explores the informationization reform of physical education courses in universities and evaluates its impact on online ideological and political education. First, interdisciplinary integration and informationization methods are used to optimize the physical education course design and integrate ideological and political education content. Then, an online learning platform is constructed to break the limitations of time and space, encourage students to deeply understand the ideological and political education content in physical education teaching, and provide real-time feedback and personalized learning support. Multimedia technology is also adopted to enhance students’ understanding and internalization of sportsmanship and ideological and political education concepts. Students’ learning behavior and ideological and political education absorption in physical education courses are individually evaluated through data analysis. Finally, the profound impact is evaluated through experiments. The results indicate that the informationization reform of physical education courses in universities significantly improves students’ participation and knowledge mastery and effectively promotes the improvement of students’ ideological and political literacy, with the average score increasing by about 14%. These results provide strong empirical support for future teaching design and also provide valuable experience for further exploring the deep integration of physical education courses and online ideological and political education.

As society progresses and science and technology evolve, the need for skilled professionals in the country continues to rise. Ideological and political education (IPE) in courses, crucial for fostering students’ overall development, has shifted from theoretical exploration to practical implementation. Sports colleges should incorporate ideological and political elements into professional courses like sports and computers based on their unique characteristics, aiming to develop high-quality, multidisciplinary talents that align with national requirements. Professional courses are often disconnected from IPE in traditional sports college teaching methods. There is excessive focus on knowledge and technology instruction, while insufficient attention is given to guiding students in developing correct values through the courses. Additionally, there is a lack of well-targeted course design and a comprehensive evaluation system. In response to the above problems, this paper studies and constructs a framework that integrates IPE with basic computer application knowledge, and designs a variety of course designs and teaching methods. In addition, a progressive assessment is designed to introduce the Convolutional Neural Networks (CNN) model and the Bidirectional Encoder Representations from Transformers (BERT) model in the early stage of teaching to conduct preliminary assessments of students’ basic computer skills and ideological and political qualities. In the later stage of the course, the Long Short-Term Memory (LSTM) network model is introduced to analyze student learning behavior and assess overall student quality based on prior evaluations. The experiment shows that the course design studied in this paper can enable students of the School of Physical Education to learn basic computer knowledge while also receiving certain ideological and political quality education. The designed teaching method is better than traditional offline teaching, online teaching and online and offline mixed teaching. When conducting comprehensive quality assessment, a 40% excellent rate can be achieved. Students are randomly selected to track their growth analysis. After the application of this teaching design, 90% of the students’ comprehensive quality assessment remains the same or improves. The findings suggest that the teaching model discussed in this paper not only improves students’ computer skills but also plays a crucial role in their JIANG ideological and political education, effectively achieving the intended teaching goals. The success of the innovative teaching design has been confirmed through practical application, providing a feasible teaching model for sports universities and other specialized institutions.