A Neural Networks Based Model to Predict the Interest of College Students in Sports Activities

Abstract

The widespread application of big data technology in various fields, including research in education and sports in colleges, has also been deeply influenced. College students are the future strength of a country, and their habits and interests in sports activities have profound significance for their physical and mental health, teamwork, and outlook on life. However, traditional research methods, such as questionnaire surveys, observations, or interviews, have obvious limitations when dealing with large amounts of complex high-dimensional data. This study aimed to extract the interesting features of college students regarding sports activities using graph neural network (GNN) technology. Then, the labels of those interest features were further predicted, and a feature matrix was constructed. Finally, the K-means clustering method was used to achieve accurate feature clustering. This study presents a novel idea and approach for physical education and event planning in colleges, offering both practical value and theoretical significance.