Player modeling for dynamic difficulty adjustment in top down action games

In video games, Dynamic Difficulty Adjustment (DDA) typically refers to the technique of adapting the video game difficulty to match the skill level of the player. A common practice is to use predictive algorithms to preemptively set up the difficulty of the game so that the player notices the adjustment minimally, and the algorithm tailors the challenge of the game to match the player’s performance. This paper aims to provide a different approach on how DDA is conducted. Combining methods of player modeling, data-driven analysis and clustering algorithms to provide a better insight of the interaction between the player and the game. However, the paper not only proposes a new methodology for DDA solely based on player performance but it also measures the engagement of the player once the adjustment has been applied by using a 7-point likert scale survey after the adjustment has taken place.