Research On Application Of Data Mining In Basketball Technology Actions

With the rapid development of the computer technology, our lifes is changing rapidly by the revolution of information. The data in basketball matchs is increasing too, but in the large amount of these data collections, there is a little really concern us. To control match in time and manage it efficiently, to analyse basketball technology actions in time is urgent for us. On the other hand, data is always unstructured, they are not appeared as traditional structured data, it makes data mining ways based on database are not suitable for them. Therefor, mining and obtaining the interesting hiding information on basketball technology actions in these data resources is a new topic in data mining.This paper discusses the appication of data mining for basketball technology actions in the case of only a few paper and reference about data mining on basketball technology actions (a few words and described references written by sports people), although I have queried and searched them many many times.The aim of this paper is to preprocess basketball technology actions, to classfy these actions with data mining technology, to mine association rules between them.The main works are shown below:The common approaches of data mining are discussed, such as preprocessing technology, classfication technology, clustering technology and mining association rules technology,etc, both ID3 decision tree classfication algorithm and Apriori association rules algorithm are studied in detail.Because of lack of training data, we collected and preprocessed the datum of basketball technology actions by watching TV and match video recording, and studied deeply on the classfication method and process with ID3 and J48 decision tree, as well as Apriori assiociation rules mining algorithm. Based on these algorithms, validation experiment is given and the correspongding results are obtained, some theoritical prooves and practice instructions for basketball game is obtained by integrating the explain result and the fact (basketball match).The paper discusses basketball technology actions mine both on a small scale and a large scale, J48 decision tree classfiction and Apriori association rules mining algorithm are applied, all these research results should have useful instrunction to basketball team.