US 12,293,629 B2
Self-evolving AI-based playstyle models
David Froy, Lakeville-Westmorland (CA); Thomas Trenkler, Thal bei Graz (AT); Karen Van Niekerk, Dieppe (CA); Michael Russ, Graz (AT); and David Small, Moncton (CA)
Assigned to IGT, Las Vegas, NV (US)
Filed by IGT, Las Vegas, NV (US)
Filed on Dec. 5, 2023, as Appl. No. 18/529,248.
Application 18/529,248 is a continuation of application No. 17/537,054, filed on Nov. 29, 2021, granted, now 11,875,637.
Prior Publication US 2024/0104998 A1, Mar. 28, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G07F 17/32 (2006.01); G06N 3/12 (2023.01)
CPC G07F 17/3227 (2013.01) [G06N 3/12 (2013.01); G07F 17/3206 (2013.01); G07F 17/323 (2013.01); G07F 17/3237 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A method for monitoring play of an electronic game, the method comprising:
generating, by a gaming server, a playstyle model for the electronic game based on a set of gameplay decisions associated with a plurality of previous gameplay sessions for the electronic game;
associating, by the gaming server, the generated playstyle model with a player of the electronic game;
identifying, by the gaming server, gameplay data for a gaming session of the electronic game with the player at a gaming system, the gameplay data comprising data indicating a gameplay decision by the player during the gaming session;
modifying, by the gaming server, the playstyle model associated with the player based on the identified gameplay data, wherein modifying the playstyle model associated with the player based on the identified gameplay data comprises comparing an action of the player indicated by the gameplay data to an action indicated by the playstyle model; and
in response to not identifying a match between the action of the player indicated by the gameplay data and an action indicated by the playstyle model, identifying a reason for no match between the action of the player indicated by the gameplay data and an action indicated by the playstyle model.