US 12,155,558 B2
Method, computer device, and non-transitory computer-readable recording medium for providing optimal path using expert knowledge
Young-Bum Kim, Bellevue, WA (US); Gun Su Lee, Seongnam-si (KR); Dongchan Kim, Bellevue, WA (US); Sun Kim, Bellevue, WA (US); Jinmyung Won, Bellevue, WA (US); Hongyeon Yu, Seongnam-si (KR); and Chanhee Lee, Seongnam-si (KR)
Assigned to NAVER CORPORATION, Seongnam-si (KR)
Filed by NAVER CORPORATION, Seongnam-si (KR)
Filed on Nov. 7, 2022, as Appl. No. 18/053,041.
Claims priority of provisional application 63/363,877, filed on Apr. 29, 2022.
Prior Publication US 2023/0353487 A1, Nov. 2, 2023
Int. Cl. H04L 45/24 (2022.01); G01C 21/36 (2006.01); G06F 16/9535 (2019.01); G06N 3/092 (2023.01); G06Q 30/0601 (2023.01); H04L 41/16 (2022.01); H04L 67/50 (2022.01)
CPC H04L 45/24 (2013.01) [G01C 21/3691 (2013.01); G06F 16/9535 (2019.01); G06N 3/092 (2023.01); G06Q 30/0625 (2013.01); H04L 41/16 (2013.01); H04L 67/535 (2022.05)] 20 Claims
OG exemplary drawing
 
1. A method for providing an optimal path to a goal performed by a computer device having at least one processor configured to execute computer-readable instructions included in a memory, the method comprising:
predicting a subsequent action of a target user through a general optimal path prediction model that is trained in a form of a graph representing a plurality of general user action trajectories of a session in which a plurality of paths of a plurality of general users extends through a plurality of services on a corresponding plurality of domains and an expert-embedded artificial intelligence (AI) model trained based on an expert knowledge-based path of an expert user in a session in which the expert user has expert knowledge, and recommending a path of the predicted action as the optimal path to the goal,
wherein the recommending of the path of the predicted action includes,
identifying a user intent based on a previous action trajectory of the target user in a current session; and
determining a path corresponding to the user intent as the optimal path through the plurality of paths of the plurality of general users and through the expert knowledge-based path of the expert user responsive to a determination from the user intent that the expert knowledge of the expert user should be applied.