One Game Fits All: Personalized Content Generation in Mobile Games

Abstract

Procedural content generation uses algorithmic techniques to create large amounts of new content for games and thus reduces the cost of production. However, this content generation is typically the same for all players and is not used to personalize and optimize the game for players’ characteristics. Thus, the core of our research is the improvement of procedural content generation through personalization. We plan to achieve personalization by using modern machine learning algorithms to learn the characteristics of the player. These characteristics will be then used as input parameters for procedural content generation algorithms to produce personalized content. We expect that personalized procedural content generation will have a positive effect on the user’s gameplay experience.