Title :
A machine learning strategy for predicting march madness winners
Author :
Gumm, Jordan ; Barrett, Andrew ; Gongzhu Hu
Author_Institution :
Dept. of Comput. Sci., Central Michigan Univ., Mount Pleasant, MI, USA
Abstract :
The Division I NCAA Men´s Basketball Tournament is a popular sporting event held annually to determine the leagues National Champion. Over the past several years the betting scene surrounding the tournament has become arguably more popular than the tournament itself, drawing in fans who bet billions overall on its outcome. In this paper, we discuss the statistical challenges in correctly predicting winners in the tournament and present a machine learning strategy for predicting the games. The Kaggle Machine Learning March Mania Competition was used to test the effectiveness of the model by comparing it against other machine-learning-based models submitted to the competition. Overall, the project was considered successful as it scored in the top 15 percentile of all submissions.
Keywords :
learning (artificial intelligence); sport; Kaggle Machine Learning March Mania Competition; March Madness winner prediction; game winner prediction; machine learning strategy; Algorithm design and analysis; Correlation; Economic indicators; Games; Indexes; Prediction algorithms; Predictive models; Kaggle Machine Learning contest; NCAA Men´s basketball tournament; non-linear regression; predictive modeling;
Conference_Titel :
Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD), 2015 16th IEEE/ACIS International Conference on
Conference_Location :
Takamatsu
DOI :
10.1109/SNPD.2015.7176206
