Year: 2025
Type: article
Abstract: : This study explores the application of machine learning and deep learning techniques to predict the outcomes of Indian Premier League (IPL) cricket matches by analyzing historical match data. Given the dynamic nature of cricket and the wide range of factors influencing match results, traditional statistical models often fail to capture the complexity of the game. This research implements a hybrid classification approach utilizing models such as Random Forest (RF), Gradient Boosting Classifier (GBC), Support Vector Classifier (SVC), and a Deep Neural Network (DNN) to predict match outcomes. The dataset, sourced from IPL match records and ball-by-ball delivery data from 2008 to 2024, was preprocessed through data encoding, feature engineering, and scaling. Each model was trained and evaluated using Scikit-learn and TensorFlow, followed by hyperparameter tuning to optimize performance. Comparative analysis showed that the Random Forest and Gradient Boosting Classifier achieved the highest accuracy of approximately 86.78%, while the Deep Neural Network demonstrated promising results with room for further improvements through fine-tuning and real-time data integration. The research highlights the effectiveness of ensemble models and deep learning approaches in capturing intricate match patterns and offers a robust framework for real-time IPL match prediction.
DOI: 10.55041/ijsrem46167
Author: Rahul Talwar
Language: English
Subfield: Economics and Econometrics
Domain: Social Sciences
Open Access status: bronze
OpenAlex ID: W4409838644