Football Match Result Prediction Using the Random Forest Classifier

Nowadays, people are turning their attention to football as the business investment more and more, especially invest in joining football club. In addition, If the results of some matches do not meet the goal of the clubs, the investors will not invest in the club and the club may be loss a lot of money that they should be. So, we have developed the web system for a football match result prediction method in order to help making investment decisions for investors and generating the guidance for developing their football teams. The objective of this project is to predict the football match results for the English Premier League, and to analyze factors affecting the outcome of the match for guiding team improvement. This project collected previous three-season match information from www.premierleague.com to predict the current league season match results. All collected data were analyzed by the machine learning technique for building a football match result prediction model, and for finding factors affecting on football match results to give advice for improving their football teams. The testing results of the prediction are shown that the accuracy and the precision are more than 70%. Therefore, this system can help the user get the guidance for improving the football team and the precise prediction of football match results.