Computer-aided Game Analysis of Net Sports in Preparation of Chinese Teams for Beijing Olympics.
Hui Zhang,Lijuan Yu,Jinju Hu
2010-01-01
Abstract:In the preparation of the 2008 Beijing Olympics, the Technique and Tactic Research Center of Shanghai University of Sport developed a series of softwares and conducted a large number of game analyses for national teams of table tennis, badminton, tennis and volleyball, and the results proved to be quite effective. Based on the methods of data collection and computation, the present paper classifies the computer-aided game analysis into non-systematic analysis, systematic analysis and intelligent analysis. The non-systematic analysis is mainly conducted by collecting the technique and tactic attributes of the last stroke in every rally (such as the technique, position, striking sequence, scoring and losing). The strength of non-systematic analysis lies in its quick feedback of the results to the coaches and players, but this method may fail to obtain the technical and tactical attributes of other strokes. The systematic analysis is done according to the strictly defined technical and tactical observation system by collecting all the technical and tactical attributes of every stroke of the player in the competition (match, set, score, player, striking technique, striking position, striking placement, technical state, effect, scoring and losing). It can provide complete information of the techniques and tactics and conduct different types of detailed analysis, but due to the heavy workload of data collection for systematic analysis, it may need longer time to provide feedback to coaches and players. Intelligent analysis is the method based on data mining and artificial neural network techniques. Data mining (association analysis) could reveal the association characteristics between several consecutive strokes or stroke positions or placements and scoring or losing. It could help coaches understand better the characteristics of the techniques and tactics and thus is likely to become one of the major methods in future ball game analysis. Through automatic training with only the input data (such as technique and tactic indexes) and output data (such as winning probability), the artificial neural network can construct models with very high precision. Therefore, further investigation in this field is necessary and worthwhile.