Applying clustering to data analysis of Physical Healthy Standard

A great deal of sports data are recording year by year, including training data of athletes, test data of students in sports course, and test data of Physical Health Standard (PHS). In the past, usage of these records is limited to basic statistics analysis. With the development of artificial intelligence and data analysis technologies, sports data analysis became more and more technical. Data mining is one of the efficient techniques, which can find unknown patterns of certain datasets and relationships among the data. However it was seldom applied in sports field. In this paper, we use one of the data mining algorithms (clustering) to analyze PHS test data with the help of SQL Server 2005. In the experiments, the scores of vital capacity, grip strength, standing long jump and step test of a student are used for input attributes, and total score of the student is used for prediction attribute. The purposes of experiments are to discover how to construct a training set and how to set parameters of Microsoft clustering algorithm. Some valuable conclusions are achieved.