Comparative Analysis of Machine Learning Models for Cricket Score and Win Prediction: A Case Study of Linear Regression, Random Forest, Neural Network, and Elastic Net

International Research Journal of Engineering and Technology (IRJET)

Volume: 10 Issue: 11, Pages:251-255, 2023, Link

Cricket is the 2nd most popular sport in theworld behind football. Especially, in Bangladesh thepopularity of cricket and the passion that people show forit, can’t be described in words. Cricket is played all aroundthe year. So, we have built a system that predicts a team'sfinal score at any point of the match. The things that aretaken into consideration are the teams that are playing,where the match is being played, what’s the current runsand wickets, how many runs have been scored, and howmany wickets have fallen in the last 5 overs. We are using 3different machine algorithms; Linear Regression, RandomForest, Elastic Net, and Neural Network to predict thescore. There is also a live win prediction system that willpredict which team is going to win at any moment of thematch. The Random Forest algorithm is used for the winprediction system. Random Forest has the highest R2 value.It also has the lowest mean squared error and meanabsolute error.
(PDF) Comparative Analysis of Machine Learning Models for Cricket Score and Win Prediction: A Case Study of Linear Regression, Random Forest, Neural Network, and Elastic Net.