A Perspective on Analyzing IPL Match Results using Machine Learning

Indian Premier League (IPL) is a T20 league which was started in the year 2008 and it is the most belaud T20 cricket carnival in the world. Since IPL has huge popularity, it is needful to inspect the possible predictors that affect the overall result of the matches. The expected solution depends on time series analysis and Machine learning techniques which minimizes the use of domain knowledge. This paper attempts to predict the runs for every ball by using the runs scored by the batsman previously as the observed data. Data from all past IPL matches is collected for the analysis and the problem is modeled as a classification problem. In this work, we build predictive models for predicting the runs based on the observed data. We have used Recurrent Neural Network (RNN) and Hidden Markov Model (HMM) for generating the models for the problem. Finally, the model created using RNN and HMM provided the highest prediction accuracy.