Machine Learning Application to Performance for Forecasting Horse Racing

다국어 초록 (Multilingual Abstract)

Each country's horse racing industry is seeking a survival strategy in the non-contact environment. In particular, some countries are trying to overcome the current crisis by allowing online ticket sales. In order for online ticket sales to be carried out and activated smoothly, trust-based game management is essential. This study aims to suggest a method that can be applied to the horse racing industry by using data science method and machine leaning techniques. A combination of various variables affecting the horse race is presented and the critical influencing factors are analyzed through the Korean horse racing data over 10 years to investigate the relative importance of the influence of each variable on the race. These attempts highlight the availability and the need for competency in data science technique-based analysis that can bring innovations to the horse racing industry

목차 (Table of Contents)

• 1. Introduction 1
• 2. Literature Reivew 3
• 2.1 The Characteristics of Horse Racing Industry and Data Analysis 3
• 2.2 Data Analysis by Machine Learning 7
• 3. Method 10
• 3.1 Data 10
• 3.2 Research Model 12
• 3.3 Analysis 13
• 3.3.1 Random Forest Model 13
• 3.3.2 XGBoost Model 14
• 3.3.3 CatBoost Model 15
• 3.4 Results 16
• 4. Conclusion 17
• 5. Limitation and Further Study 19
• Reference 20