Authors:
Sheng-Wei Wang
1
and
Wen-Wei Hsieh
2
Affiliations:
1
Fo Guang University, Taiwan
;
2
National Tsing Hua University, Taiwan
Keyword(s):
Basketball Referee, Performance Analysis, Machine Learning, Pocket Algorithm.
Related
Ontology
Subjects/Areas/Topics:
Computer Systems in Sports
;
Sport Science Research and Technology
;
Sport Statistics and Analyses
Abstract:
Basketball referees are important in a basketball game. In this paper, we analyze the performance of basketball referees in a game from history data and using the machine learning techniques. The data are collected from Taiwan Super Basketball League games. We first observed that the teamwork is a key factor to the performance of referee teams. Furthermore, the degree of teamwork are more important than the personal capabilities. Then, we derived some classifiers by machine learning algorithms to further analyze the data set. Among the three classifiers, a classifier named linear classifier using pocket algorithm, which is able to classify the data points with most correct rate, performs better than the other two classifiers. The classifier also proved the importance of teamwork is much larger than that of personal capability. In the future, the classifier can be used to predict the performance of a referee team in a basketball game.