Association rule mining is an important data mining technique that finds association rules by mining frequent attributes. This work aims to construct association rules that determine cricket players` strengths and weaknesses. We propose an approach to learn the association of strengths or weaknesses exhibited by batters (or bowlers) with the type of delivery they have faced (or bowled). In essence, the bowling (or batting) features that may be associated with the batter`s (or bowler`s) strengths or weaknesses are investigated. Each delivery is represented as a set of bowling and batting features, similar to the set of items representing a transaction in association rule mining. Apriori algorithm of association rule mining is used to obtain the strength association rules and weakness association rules. Cricket text commentary data are obtained from the EspnCricInfo website and utilized for finding player`s strength and weakness rules. Rules for more than 250 players are constructed by analyzing text commentaries over one million deliveries for 13 years (2006-2019). The data, codes, and results are shared at https://bit.ly/3rj6k6c.
© Copyright 2022 Machine Learning and Data Mining for Sports Analytics. MLSA 2021. Communications in Computer and Information Science. Veröffentlicht von Springer. Alle Rechte vorbehalten.