Bayesian Approach to Classification of Football Match Outcome

Authors

  • Muhammad Haleq Azhar Abdul Rahman Faculty of Computer Science and Information Technology, Universiti Tun Hussein Onn Malaysia, 86400 Parit Raja, Batu Pahat, Johor, Malaysia
  • Aida Mustapha Faculty of Computer Science and Information Technology, Universiti Tun Hussein Onn Malaysia, 86400 Parit Raja, Batu Pahat, Johor, Malaysia
  • Nazim Razali Faculty of Computer Science and Information Technology, Universiti Tun Hussein Onn Malaysia, 86400 Parit Raja, Batu Pahat, Johor, Malaysia
  • Rahmat Fauzi

Abstract

The football match outcome prediction particularly has gained popularity in recent years. It attract lots type of fan from the analyst expert, managerial of football team and others to predict the football match result before the match start. There are three types of approaches had been proposed to predict win, lose or draw; and evaluate the attributes of the football team. The approaches are statistical approach, machine learning approach and Bayesian approach. This paper propose the Bayesian approaches within machine learning approaches such as Naive Bayes (NB) , Tree Augmented Naive Bayes (TAN) and General Bayesian Network (K2) to predict the football match outcome. The required of football data is the English Premier League match results for three seasons; 2016 – 2017, 2015 – 2016 and 2014 – 2015 downloaded from http://www.football-data.co.uk. The experimental results showed that TAN achieved the highest predictive accuracy of 90.0 % in average across three seasons among others Bayesian approach (K2 and NB). The result from this research is hope that it can be used in future research for predicting the football match outcome.

Downloads

Download data is not yet available.

Downloads

Published

25-11-2018

How to Cite

Abdul Rahman, M. H. A., Mustapha, A., Razali, N., & Fauzi, R. (2018). Bayesian Approach to Classification of Football Match Outcome. International Journal of Integrated Engineering, 10(6). https://penerbit.uthm.edu.my/ojs/index.php/ijie/article/view/2780

Most read articles by the same author(s)