Abstract:
Cricket is one of the most popular games in many countries. More than 19 countries playing cricket as main game & population for cricket will increasing gradually. However there are no proper tool for analyzing pre outcome of the match stating to end, and available tools are not support to simulate match using batting partnerships. The ultimate goal of predicting pre outcome of cricket match is to identify key players and their batting performances also prevent wrong players selecting and toss decision by making statistical predictions.
This project is focusing on One Day International (ODI) cricket match and predicts the outcome of the particular match. Our proposed solution consists of three major modules namely; Web UI Module, CRIC-Win Analytic Engine and Backend Data Module. CRIC-Win Analytic Engine has two sub data models, one for predict overall match outcome based on given pre match data and next for predicting match outcome based on batting partnership both home team and opponent team. All sub models in the CRIC-Win Analytic Engine are developed base on Naïve Bayes algorithm and use for generating the classifier model which can be used to predict the outcome of the cricket match.
Mainly prediction result divided into two segments. First predict overall pre outcome of the match based on given details and next predict how each partnership will affect to win the match. This work suggests that the relative team strength between the competing teams forms a distinctive feature for predicting the winner. Modeling the team strength boils down to modeling individual players batting partnerships and bowling performances forming the basis of our approach. We use partnership statistics as well as the recent performances of home team & opponent team, partnership records, wicket falling pattern, remaining overs, required runs, toss decision, ground and day/night effects have also been considered in order to predict the outcome of a match. The solution can be simulated match outcome before the match start and while playing. Also this research describes in detail the different attribute selection techniques as well as the data mining algorithms used to solve this problem of outcome prediction in cricket. We have also used accuracy as the evaluation criteria to evaluate how well the prediction performs.
