IPL MATCH PREDICTION USING MACHINE LEARNING TECHNIQUES

Abstract

The nation of India is enthralled with cricket, encompassing diverse formats including Test matches, ODI, and T20. The Premier League of India (IPL) exemplifies this fervor, drawing players from regional, national, and international teams. Factors like live streaming, radio coverage, and televised broadcasts buoy the league's widespread popularity among cricket enthusiasts. Predicting IPL match outcomes holds paramount significance for online traders and sponsors. This paper proposes a model for forecasting IPL match results, leveraging K-Nearest Neighbor, Logistic Regression, Random Forest Classifier (RFC), Support Vector Machines, and other machine learning methods and a voting regressor (a fusion of linear regression and support vector regressor). Our approach integrates various factors such as team composition, player batting, and bowling averages, team performance history, toss outcomes, venue dynamics, and the likelihood of winning by batting first against a certain team at a particular location. This research aims to enhance predictive accuracy and facilitate informed decision-making in IPL match analysis.

Keywords – Voting Regressor, SVR, Random Forest Classifier (RFC), K-Nearest Neighbors (KNN), and Logistic Regression (LR).