Research And Application Of Badminton Tactic Analysis System Based On Association Rules

Badminton is an important strength of the Chinese sports project, and it is avictorious army of the Chinese Olympic team. But as more and more informationtechnology is applied to the field of sports, especially the decision support systembased on data mining technology is applied increasingly popular. Badminton sport inChina can continue to keep defending in the international high level competition,faced with new challenges. The information level of Badminton sports in our countryis limited at present, the analysis of the relevant data for the preservation of a largenumber of game only limited in the field of statistics, so it is necessary to make adeeper analysis of the existing data by taking association rules mining, and find therules which related to the athletes’ winning factors. Since the traditional associationrules algorithm exists many shortcomings such as time-consuming, redundancy rules,it is unable to adapt to the fast pace of the badminton match on the efficiency. In orderto solve this problem, in this paper, I made an exploratory research of the data miningtechnology in the field of badminton sport. This paper’s topic ifs form “Thedevelopment and application of badminton training quality monitoring and statisticaltactic based on tablet PC” which is set up by the National Sports Bureau. It can solvehow to implement the badminton match data quickly on-the-spot collection andanalysis, and provide an auxiliary support to make targeted training programs andtechniques for coaches.The work for this paper reflected in the following aspects:1.Research the basicknowledge of Association rule data mining, analyses the defects of the traditionalApriori algorithm, and make a comparative analysis for the related improvedalgorithm.2. Propose an improved algorithm(ACARMI) which based on Reorder andDirect algorithm, it can solve the problem that the Reorder and Direct algorithmgenerated large numbers of candidate item-sets during the operation and improve the efficiency of the algorithm.3. Designed and implemented a data preprocessing modelaccording to the characteristics of the badminton statistics, it can turn the game datainto the form which adapt to the Boolean association rules mining.4. Designed andimplemented the Badminton tactic analysis system, and implemented the wholeprocess which from the data collection to the data analysis.The main contributions of the paper are: Improved the Reorder and Directalgorithm which focus on the first kind of constraint problem, and proposed a newimproved algorithm (ACARMI) which can improve the efficiency of data analysis.Designed and implemented a data preprocessing model which can adapt to thebadminton data features, and the raw data after the transformation can be used toanalyze by the Boolean association rules mining algorithm. Analyzed the collectedexperimental data in the f badminton tactic analysis system through ACARMIalgorithm, validate that the system can play a good supporting role in makingscientific training plan and tactic decision for coaches.