Data Mining Analysis of Overall Team Information Based on Internet of Things

• Published in: IEEE access Vol. 8; pp. 41822 - 41829
• Main Authors: Lee, Yueh-Shiu, Wang, Jun-Ren, Zhan, Jun-We, Zhang, Jing-Mi
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Big Data; Cameras; Data analysis; Data mining; e-sports; Games; High school basketball; Internet of Things; Linear regression; Mathematical analysis; Regression analysis; spectator sport; Support vector machines; Training; Trends
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2020.2976728

In professional basketball games, big data has been largely used in analyzing the reasons for winning or losing games and further to design relevant stratagem according to the analytic results to attain victory. Nonetheless, the High School Basketball League (HBL) in Taiwan never used big data or relevant research to analyze game results. The study aims to conduct big data analyses to discuss the key winning factors and trends for HBL. Using Excel and multiple linear regression to understand the importance level and trend of each variable to the winning rate. Additionally, combining with the Support Vector Machine (SVM) prediction to confirm whether the big data analytic result is applicable for implementing in realistic games. After implementing the analysis of multiple linear regression, based on the yearly trends, the significant influence factors are 2P%, 3P%, FTM, TRB, OREB, STL, and TOV. Consequently, the prediction has reached 85% after inputting these data into SVM.