Using Data Mining Techniques to Predict Win-Loss in Korean Professional Baseball Games

• DOI: 10.7232/jkiie.2014.40.1.008
• Publication Date: 2014-03-14T06:35:42Z
• Authors: Younhak Oh; Han Kim; Jaesub Yun; Jong-Seok Lee
• Keywords: Binary classification; Predictive modelling; Lasso; Tree (set theory)

In this research, we employed various data mining techniques to build predictive models for win-loss prediction in Korean professional baseball games. The historical containing information about players and teams was obtained from the official materials that are provided by KBO website. Using collected raw data, additionally prepared two more types of dataset, which ratio binary format respectively. Dividing away-team's records corresponding home-team generated while dataset comparing record values. We applied seven classification three (raw, ratio, binary) datasets. decision tree, random forest, logistic regression, neural network, support vector machine, linear discriminant analysis, quadratic analysis. Among 21(= 3 datasets<TEX>${\times}$</TEX>7 techniques) scenarios, most accurate model forest technique based on accuracy 84.14%. It also observed using helped better than data. From capability variable selection stepwise found annual salary, earned run, strikeout, pitcher's winning percentage, four balls important factors a game. This research is distinct existing studies used different