A Study on the Methodology of Analysis of Competition Data to Improve Sports Performance

Abstract

The purpose of this study is to propose and discuss an analysis theory suitable for the field of sports game analysis in line with the trend of the modern era. We would like to explore various analytical examples and issues used in the field of sports game analysis and propose data mining, an analysis methodology with high convergence interaction with future innovative technologies based on modern data science. Traditional analysis methods in the past overlooked the influence of independent factors that occur between specific performance factors and a number of factors.BR In particular, research results derived from the combination of economic factors through interaction between economic factors were limited. However, in the case of data mining analysis methods, regular data generated in various phenomena and situations are combined to increase the utilization and accessibility of analysis, and the purity inherent in the data is derived as a result to infer meaningful results compared to traditional statistical techniques. The use of data mining analysis methods in the field of sports competition analysis can infer and predict the results of various diversified and multifaceted phenomena in the sports field in three dimensions, and is very popular in using analysis results. In addition, the field applicability of research results for research purposes is secured, and the efficiency of practical values is high, so advanced behavioral pattern analysis will be possible through existing numerical analysis. Finally, the use of data mining from the perspective of sports game analysis (image analysis, data analysis, media analysis) will be able to upgrade sports data analysis in conjunction with big data analysis such as predictive analysis, machine learning, streaming analysis, and cluster analysis in the database.