中国乒乓球队奥运攻关研究报告——基于人工神经网络的乒乓球比赛诊断模型研究  

Research Report on the Preparations of Chinese Table Tennis Team for the Olympics——On the Diagnostic Model of Table Tennis Competition Based on Artificial Neural Network

作　　者：肖毅[1] 张辉[1] 

机构地区：[1]上海体育学院经济管理学院,上海200438

出　　处：《体育科研》2008年第6期19-22,共4页Sport Science Research

基　　金：2008国家体育总局奥运科研攻关与科技服务项目(07052)

摘　　要：以世界优秀男子乒乓球运动员的比赛数据为训练样本,利用人工神经网络强大的自学习功能建立了乒乓球比赛获胜的诊断模型。根据乒乓球比赛的主要技战术指标数据,利用该模型能较好地诊断比赛的获胜概率,诊断精度较高。同时运用该模型计算了影响乒乓球比赛获胜的主要技战术指标的权重值,为有针对性的指导运动员训练提供了科学参考。Taking the competition data of the world elite men＇s table tennis players as the training sample, the authors used the strong self-learning function of artificial neural network to establish a diagnostic model of table tennis competition. By using the main technical and tactical data in competition, the model may give the probability of winning and the precision of the diagnosis is quite high. The paper uses the model to calculate the weight values of the main indicators that affects the winning of the table tennis games so as to provide scientific reference for coaching the training of the players.

关 键 词：神经网络 乒乓球 技战术 诊断模型 

分 类 号：G846[文化科学—体育训练]