Data-Driven Cricket: A Machine Learning Approach to IPL Score Prognostication

The Indian Premier League (IPL) is a very popular and globally recognized cricket tournament, attracting many spectators worldwide. The proposed work aims to predict cumulative scores of teams in IPL using attributes such as batting and bowling team data, total runs, wickets, overs, runs scored in the last 5 overs, batting taken in the last 5 overs, and Total Score. This work employs Machine Learning (ML) techniques such as Linear Regression (LR), Random Forest (RF), and Support Vector Machine (SVM) models to predict IPL scores. The database considered in this work was carefully obtained from well-known websites such as ESPN and Cricbuzz, ensuring the reliability and credibility of the research. In addition to the prediction models, a user-friendly graphical user interface (GUI) is implemented for easy interaction with the prediction system. This GUI allows the user to enter the necessary match data and displays the predicted score. Uniquely, this system offers a range of points that can be extended up to 5 runs from the estimated score. This range recognizes the unpredictable nature of cricket matches and provides more realistic and practical predictions. The combination of various ML models and an intuitive GUI makes this system a powerful tool for IPL score prediction.