Using data mining techniques to optimise athlete training and recovery programmes in tertiary physical education

Aiming at the demand for scientific training of athletes in college sports education, this paper integrates data mining technology to propose athlete training and optimisation methods, and constructs an athlete training quality monitoring system and intelligent recovery assessment system. The traditional Apriori algorithm is improved by using multidimensional association rules, and multidimensional attribute mining is carried out on the collected data of athletes’ training data to search for frequent item sets and output strong association rules, so as to achieve the monitoring of training quality and adjustment of training programmes. Using the improved fuzzy decision-making method to filter out the optimal feature subset, and integrating the improved whale algorithm and random forest to achieve intelligent recovery effect evaluation. By carrying out the practice of training and recovery optimisation, it can be seen that the total score of physical fitness test of track and field athletes increased from 18.19 to 19.8 before the experiment, and the training quality was significantly improved. Various health indicators such as heart rate, blood lactate, serum creatine kinase, etc. gained significant improvement in adopting the recovery optimisation method of athletes in this paper. The mean values of training status, coaching factors, and personal situation satisfaction evaluation dimensions were 4.35, 4.425, and 4.38, respectively, and the training and recovery plan of this experiment was well received by the subject athletes.