Live Cricket Score Prediction Web Application using Machine Learning

Cricket is an uncertain sport in which two teams compete against each other, and no Machine Learning model can predict the outcome with 100% accuracy. However, with some inputs such as conditions, required factors, circumstances, and past data, it is possible to predict an approximate score for either team or the match’s victor. As a result, analysts from the respective teams use all the available data and statistics to improve their team’s performance. Therefore, the authors have developed a web application for performing predictive analysis of a live T-20 match to predict the final score and to also predict the winner of the contest before the match begins. A live score scraper is also included in the app, which extracts accurate match data and loads it into the model for prediction. For a deeper understanding, the application allows to visualize the predictions using graphs. Algorithms such as Multivariate Polynomial Regression and Random Forest Classifier are used to make these predictions, while the web application is developed using Flask.