An Intelligent Prediction Model for IPL Winners Using Deep Learning and KNN Algorithm

The Indian Premier League (IPL) is very widely acknowledged across the globe and commercially successful cricket leagues anywhere in the globe, which attracts high public interest and participations from its fans and spectators themselves. Predicting IPL matches can prove to be a real challenge and provides sports analytics with a unique and rich source of insights, being beneficial for multiple stakeholders: teams, analysts, and organizations involved in betting. This paper outlines an all-inclusive and comprehensive research study that involves current machine learning algorithms, specifically the K-Nearest Neighbours (KNN) regression model. Along with a varied multitude of neural learning algorithms, this particular model is designed to predict IPL match winners. The methodology involved pre-processing data, feature selection, model training, and evaluation that ultimately led to a solid prediction framework. For that, performance of the regression model KNN is considered against baseline as well as traditional models so that we can shed light over its productiveness in the world of sports prediction. Further, the findings of this research add further value to the body of sports analytics literature and pave the way for further explorations into predictive modeling strategies.