Predicting player disengagement and first purchase with event-frequency based data representation

In the game industry, especially for free to play games, player retention and purchases are important issues. There have been several approaches investigated towards predicting them by players' behaviours during game sessions. However, most current methods are only available for specific games because the data representations utilised are usually game specific. This work intends to use frequency of game events as data representations to predict both players' disengagement from game and the decisions of their first purchases. This method is able to provide better generality because events exist in every game and no knowledge of any event but their frequency is needed. In addition, this event frequency based method will also be compared with a recent work by Runge et al. [1] in terms of disengagement prediction.