Journal: International Journal Of Research And Development (ijrd)
ISSN Number:
0
Downloads
18
Views
Abstract
Advancements in data crunching technologies have hugely impacted sports science especially soccer with most football teams having a data science department aimed at providing data driven solutions for various issues not limited to : Match strategy, tactics, and analysis, Identifying players’ playing styles, Player acquisition, player valuation, and team spending. Identifying the right features to be incorporated in predictive models for soccer matches however remains a viable research area. This study presents a machine learning based predictive model for soccer outcomes using historical data. Data preprocessing algorithms are presented as well as a framework for computing expected goals and incorporating individual player performance in predicting soccer outcomes.