Tennis Match Situation Analysis and Prediction Method Based on Hybrid LightGBM Modeling

Accurately assessing an athlete's tactical performance is a key element for success in highly technical and competitive sports environments, but traditional analytical approaches are unable to comprehensively analyze the data and still have some limitations. This study proposes and implements a situation analysis and prediction method based on a hybrid LightGBM model and a large amount of game data. Through the combination of linear regression method, Pearson's coefficient, and multilayer perceptron, the LightGBM model has the best overall performance after comparing multiple methods, thus reflecting the model's high accuracy and goodness of fit, which can provide accurate situation analysis and tactical guidance for the game.