DocumentCode
2543290
Title
CricAI: A classification based tool to predict the outcome in ODI cricket
Author
Kaluarachchi, Amal ; Varde, Aparna S.
Author_Institution
Dept. of Comput. Sci., Montclair State Univ., Montclair, NJ, USA
fYear
2010
fDate
17-19 Dec. 2010
Firstpage
250
Lastpage
255
Abstract
Victory is the ultimate goal in any sport. In this work we address the winning factors in the sport of One Day International (ODI) cricket. Winning an ODI cricket match depends on various factors related to scoring as well as physical strength of the two teams. Some of the factors have been described in the literature but there is scope for further research on analyzing them, especially with reference to predicting victory. Interesting factors include home game advantage, day/night effect, winning the toss and batting first. In this article, we have used artificial intelligence techniques, more specifically Bayesian classifiers in machine learning, to predict how these factors affect the outcome of an ODI cricket match. Based on the emerged results, we have developed a software tool called CricAI. This tool outputs the probability of victory in an ODI cricket match using input factors such as home game advantage available at the beginning of the match. The CricAI tool can be used in real-world applications by teams playing cricket. It can accordingly be helpful in adjusting certain factors in order to maximize the chances of winning the real game.
Keywords
Bayes methods; learning (artificial intelligence); pattern classification; sport; Bayesian classifiers; CricAI; ODI cricket; artificial intelligence techniques; machine learning; sport; Africa; Australia; Classification algorithms; Classification tree analysis; Games; Machine learning; AI and Automation; Bayes Thoerem; Classifiers; Predictive Analysis; Probability; Sports Applications;
fLanguage
English
Publisher
ieee
Conference_Titel
Information and Automation for Sustainability (ICIAFs), 2010 5th International Conference on
Conference_Location
Colombo
Print_ISBN
978-1-4244-8549-9
Type
conf
DOI
10.1109/ICIAFS.2010.5715668
Filename
5715668
Link To Document