To address the problem that technical and tactical aspects of ball games have a large impact on match-winning, association rules are used to mine and analyse the data. Item constraint terms are added to the original Apriori algorithm to reduce the generation of redundant data, and the operational efficiency of the improved association rules is improved with the aid of the GSA-PSO algorithm to give full play to the fast convergence advantage of the hybrid algorithm. The experimental results show that the algorithm proposed in the study is more convergent than the Apriori algorithm, and the former has fewer iterations on average in the single-peaked function; the overall search accuracy is significantly improved. The algorithm is also less disturbed by the dataset than the Apriori algorithm, with an average running time of 19.23s, 21.54s and 25.61s for dataset sizes of 200, 500 and 1000 respectively. The deviation rate between the predicted and actual coaching strategies was less than 5% and the scoring tolerance rate was between 0.04 and 0.10. It indicates that the algorithm proposed in the study can improve the efficiency of data analysis, which is conducive to the development of a perfect ball sport technical and tactical scheme and promote the stability of ball sport technical and tactical development.
Read full abstract