Game State Prediction with Ensemble of Machine Learning Techniques

In the game industry, applying machine learning techniques has become more and more important and it will be an essential part in near future. However, most of researches have focused on churn prediction and purchase prediction. In this paper, we try to predict the future state of games by observing the game events. In a game, multiple events are open at the same time, so it is very hard to predict the effect of events. We dame game state that set of variables which reflects user behaviors in the game, such as the number of playing users and the amount of money spent by users. We make a game state prediction models by applying an ensemble of machine learning techniques