The purpose of this study was to propose a new methodology for the automatic identification and the classification of the swimmers' kinematical information during the interval trainig on competitive swimming. Forty-seven college swimmers attached the newly developed chest band sensor unit, which has a triple-axes accelerometer and then performed a controlled interval training set with four stroke styles. The author identified swimmer's states, such as the swimming/rest phases and the start, turn and goal touch instants by using the Ay acceleration. With the inductive inference based on the experimental results and the deductive inference based on the empirical rule on the interval training brought the estimation of the swimming time. As for the classification of the swimming strokes, extracted swimming phase signal, the mean, variance and skewness of each axis acceleration were calculated for each bout. The authors compared some of data mining algorithms for the stroke style classification with four descriptive statistics, such as Var(Ax), Skewness(Ay), Mean(Az), Skewness(Az) as the independent variables and stroke style as the depending variable. The accuracy of the stroke style classification by both the multi-layered neural network and the C4.5 decision tree were 91.1%.