User experience modeling for gaming applications
In various embodiments, a method is presented that includes: generating, via a system including a processor, behavioral experience analysis (BEA) tools based on a preference learning model; receiving, via the system, game telemetry data from a gaming application; generating, via the system, a predicted user motivation by applying the BEA tools to the game telemetry data; and facilitating adaptation of the gaming application based on the predicted user motivation.
1. A method comprises:
generating, via a system including a processor, behavioral experience analysis (BEA) tools based on a preference learning model that includes an artificial intelligence (AI) model trained to infer a preference relation between variables of the AI model;
receiving, via the system, game telemetry data from a gaming application associated with game play of a player;
generating, via the system, a predicted user motivation by applying the BEA tools to the game telemetry data, wherein the predicted user motivation includes motivation data that indicates motivation scores for the player for each of a plurality of motivation factors associated with a corresponding one of a plurality of different motivations, and wherein each of the motivation scores for the player predicts an amount that the player's behavior while playing the gaming application is motivated by one of the plurality of different motivations; and
facilitating adaptation of the gaming application based on the predicted user motivation.
2. The method of claim 1 , wherein the preference learning model is trained based on a plurality of player questionnaires associated with prior game play and further based on prior game telemetry data associated with the prior game play.
3. The method of claim 2 , wherein the preference learning model is trained using second-order data generated combinatorically from the plurality of player questionnaires.
4. The method of claim 2 , wherein the preference learning model is implemented via support vector machines (SVMs).
5. The method of claim 4 , wherein the SVMs include non-linear SVMs with radial basis function kernels.
6. The method of claim 1 , wherein the game telemetry data includes at least one of: playtime data, completion data or progression data.
7. The method of claim 1 , wherein the game telemetry data includes an indication of one of a plurality of player types generated via a clustering analysis on other game telemetry data.
8. The method of claim 1 , wherein the system is implemented via a game development platform that further includes a gaming development application, and wherein facilitating adaptation of the gaming application includes facilitating adaptation of the gaming application via the gaming development application.
9. The method of claim 1 , wherein the gaming application includes a plurality of optional versions, wherein the system is implemented via a gaming system that executes the gaming application, and wherein facilitating adaptation of the gaming application includes selecting one of the plurality of optional versions based on the predicted user motivation.
10. The method of claim 1 , wherein facilitating adaptation of the gaming application includes identifying a player mismatch.
11. The method of claim 1 , wherein the the motivation scores for the plurality of motivation factors include a competence score, an autonomy score, a relatedness score and a presence score.
12. The method of claim 1 , wherein the plurality of different motivations include three or more of: competition, completion, fantasy, destruction, discovery, strategy, excitement, and power.
13. The method of claim 1 , wherein the player is a gaming bot.
14. The method of claim 1 , wherein the game telemetry data includes pixel data associated with game video.
15. A system comprises:
a processor;
a memory configured to store operational instructions that, when executed by the processor, cause the processor to perform operations that include:
generating, via a system including a processor, behavioral experience analysis (BEA) tools based on a preference learning model that includes an artificial intelligence (AI) model trained to infer a preference relation between variables of the AI model;
receiving, via the system, game telemetry data from a gaming application associated with game play of a player;
generating, via the system, a predicted user motivation by applying the BEA tools to the game telemetry data, wherein the predicted user motivation includes motivation data that indicates motivation scores for the player for each of a plurality of motivation factors associated with a corresponding one of a plurality of different motivations, and wherein each of the motivation scores for the player predicts an amount that the player's behavior while playing the gaming application is motivated by one of the plurality of different motivations; and
facilitating adaptation of the gaming application based on the predicted user motivation.
16. The system of claim 15 , wherein the preference learning model is trained based on a plurality of player questionnaires associated with prior game play and further based on prior game telemetry data associated with the prior game play.
17. The system of claim 16 , wherein the preference learning model is trained using second-order data generated combinatorically from the plurality of player questionnaires.
18. The system of claim 16 , wherein the preference learning model is implemented via support vector machines (SVMs).
19. The system of claim 18 , wherein the SVMs include non-linear SVMs with radial basis function kernels.
20. The system of claim 15 , wherein the game telemetry data includes at least one of: playtime data, completion data or progression data.