Using Data Mining Methods to Improve Cross- Gaming Prediction in the Gaming Industry

Using Data Mining Methods to Improve Cross- Gaming Prediction in the Gaming Industry

Considering a variety of casino games and slot machines offered on many casino floors, high utilization of games through cross-selling is likely to lead to an increase in gaming volume and player visitation frequency. Using the real-world gaming data, this study examined various data mining methods to improve the prediction accuracy for cross-gaming. Cross-gaming in this study refers to slot players’ table game play and table game players’ slot play. Of the various data mining methods, C5 and an ensemble model consisting of decision tree classification models outperformed other models in accurately predicting potential cross-gamers. Using the cross-gaming propensity scores derived from these models, casino managers can improve their target marketing efforts for cross-gaming.