Prediction Model for Horse Racing

Abstract

Because of rapid development of data collection and storage technology, data become a kind of important resource and interesting knowledge and prediction can be discover through the data. In order to predict some unknown result, machine learning is applied to different industries widely. A predictive model will be trained and tested with large datasets. Horse racing is a legal gambling sport in Hong Kong and most of the historical result and information are shown on the Hong Kong Jockey Club official website. The public will alaylze the data on the newspaper or website to predict the race result manually. However, it is difficult to find out any pattern. Therefore, a horse racing result prediction model is proposed in this project. The data of race meetings, horses and jockeys will are used as features. The model will focus on the prediction of the champion horse of each race meeting. The accuracy will be evaluate with the Favorite Odd which means the specific horse will be the first to finish the race with the highest chance.