U.S. Patent no. 10,286,323: Dynamic difficulty adjustment
Issued May 14, 2019, to Electronic Arts Inc.
Priority Date March 8, 2016
U.S. Patent No. 10,286,323 (the ‘323 Patent) relates to level of difficulty in video games. Games that are too easy or too difficult will result in less enjoyment for a user which will lead to less playtime. A challenge for game developers is designing a game difficulty levels that are more likely to keep users engaged for longer periods of time. Based on user data and monitoring the user against the expected results the video game may modify the difficulty level by selecting a seed value to change the level accordingly. A prediction model may be generated using machine learning as well to affect the difficulty level. The computer processor will continue to monitor progress and can continuously alter the video game to make these changes. This process can then further increase engagement with the video game.
Embodiments of systems presented herein may perform automatic granular difficulty adjustment. In some embodiments, the difficulty adjustment is undetectable by a user. Further, embodiments of systems disclosed herein can review historical user activity data with respect to one or more video games to generate a game retention prediction model that predicts an indication of an expected duration of game play. The game retention prediction model may be applied to a user’s activity data to determine an indication of the user’s expected duration of game play. Based on the determined expected duration of game play, the difficulty level of the video game may be automatically adjusted.
1. A computer-implemented method comprising: as implemented by an interactive computing system configured with specific computer-executable instructions, accessing a set of user interaction data of a user, the set of user interaction data corresponding to a user’s interaction with a video game; selecting a parameter function from a plurality of parameter functions based at least in part on a penalty value associated with at least some of the parameter functions from the plurality of parameter functions, the penalty value derived for the parameter function based at least in part on characteristics of the parameter function; based at least in part on the set of user interaction data, determining a predicted churn rate for the user by applying the set of user interaction data to the selected parameter function, the predicted churn rate corresponding to a probability that the user ceases playing the video game; and modifying execution of the video game based at least in part on the predicted churn rate for the user.