Research On Men' 1500m Freestyle Competitive Capacity Time Series Characteristics Data Mining

The men' s 1500m freestyle swimming belong to the long distance athletics swimming competition.The international and domestic research on the men' s 1500m freestyle mainly concentrated in the body shape characteristics of the athletes?swimming strokes technical features cooperate with the timing?starts and turns.The study of the pattern of time distribution of swimming whole time less.Training educational circles in the early study of competitive ability in the propose athletic ability including the ability of growth is accord with the characteristics of time series.This article use the method of literature review?expert interview and mathematical statistics and machine learning methods.To collect the international and domestic competition in the men's 1500 m freestyle 50 meters segmentation results were analyzed,and the specific results are as follows:Firstly,in this article,the men's 1500 m freestyle record data is essentially composed of a lot of cross section data of time series.Based on time series theory,put forward the concept of competitive ability of time sequence,characteristics and classification:Competitive ability refers to the time sequence in the process of sports training and sports competition,by reaction,description,and record the related physical,intellectual,technical,tactical and psychological aspects of data in accordance with the time arranged in sequence.The location of the data in time series related to time;Time series is to reflect the competitive ability related to characterize the result variables at different times;Data in time series data can be a period of time can also be a point in time of data;Dependence on the time series usually exist before and after the time.Time series can be divided into one competitive ability time series and multivariate time series competitive abilities.Can be divided into discrete time series competitive ability and continuous competitive ability of time series;Can be divided into smooth and competitive ability of time sequence and non-stationary time series competitive abilities.Secondly,for international and domestic competition performance,four levels of the 1500 meters freestyle athletes were four moments of statistical test,the result shows:the international and domestic level of men's 1500 m four performance of the mean,standard deviation,skewness,and kurtosis are differences.Thirdly,by using support vector machine in machine learning and multiple support vector machine(SVM)method,the collected domestic players in the men' s 1500 m freestyle performance classification model,condition monitoring,key players on the project athletes' selection and forecast cycle performance of athletes has practical application value.Fourthly,for domestic and international competition by using the method of random forest adjacent two each block importance level performance swimming the whole(vertical)and the international domestic same importance level swimming the whole contrast(horizontal)contrast analysis:four level athletes in the international and domestic competition in the 1500 - meter swim period of importance in both similarities and differences.To strengthen all levels it is concluded that the key period of the meshes ability is one of the keys to improve the final grade.Finally,wth the sun in different periods of two levels of data and random game data,respectively for validation of the model(horizontal and vertical),the verification results shows that the importance of analysis piecewise model is valid,athletes competitive state study model in the project monitoring,key athlete selection and cycle performance of athletes in such aspects as forecast has practical application value.