Abstract
Cricket is the second most popular sport ever. It is generally played in three different standard formats namely, test-cricket, one-day cricket, and twenty-twenty cricket (T-20). The T-20 format has been gaining popularity day by day due to its small window of play-time and adventurous materialization. It has become a sport of million dollars. In this scenario researchers are proposing automated and prompt data mining techniques for predictions of various aspects related to T-20 matches. In this paper, a data mining approach has been presented that takes into account the match historical statistics, the real-time match scenario, and predicts the innings-wise score and final score. Suitable mathematical models has been proposed for the purpose of generating the analytical outputs. The proposed scheme is validated through experimental evaluations. Satisfactory rate of accuracy has been obtained that outperforms the other state-of-the-art schemes while compared. The overall rate of accuracy thus obtained is 94% which is a good figure when we are considering the real time data.
Get full access to this article
View all access options for this article.