A Novel Data Mining Algorithm and Its Applications in Basketball Match Technique and Tactical Analysis

Abstract

The basketball game is a process in which players use different basic basketball methods to change their actions according to certain tactical structural forms. In the field of basketball, a huge data is generated during training, matches or competitions, sports management, and national physical fitness tests. During basketball matches, the management uses numerous methods to collect data about opponent teams, some of which are intuitive, while others may not be able to directly display their important information. In the sports domain, the coaches and managers use data mining techniques for transforming sports data into actionable knowledge and training their athletes for possible predictions of the outcomes of the games. This article focuses on the analysis application of data acquisition and preprocessing in basketball techniques, and tactical analysis is studied using the proposed data mining algorithm. The proposed algorithm and data of basketball games were used to make association rules analysis to analyze technical and tactical characteristics. The algorithm generates association rules based on the frequent item sets of basketball technical moves. It is evident from the experimental results that the proposed algorithm leads to high accuracy and better outcomes in terms of prediction.