US 11,755,673 B1
Systems and methods for recommending pages
Bradley Ray Green, Bellevue, WA (US); Vishal Vusirikala, Seattle, WA (US); Feng Wang, Redmond, WA (US); and Komal Kapoor, Issaquah, WA (US)
Assigned to Meta Platforms, Inc., Menlo Park, CA (US)
Filed by Meta Platforms, Inc., Menlo Park, CA (US)
Filed on Mar. 30, 2022, as Appl. No. 17/708,846.
Application 17/708,846 is a continuation of application No. 16/234,488, filed on Dec. 27, 2018, granted, now 11,301,533.
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 7/00 (2006.01); G06F 16/9536 (2019.01); G06F 16/2457 (2019.01); G06N 5/04 (2023.01); G06F 16/9535 (2019.01); G06N 20/00 (2019.01)
CPC G06F 16/9536 (2019.01) [G06F 16/24578 (2019.01); G06F 16/9535 (2019.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
identifying, by a computing system, an interest of a user in trending content based on page embeddings associated with pages visited by the user during a session;
applying, by the computing system, the page embeddings to a machine learning model to generate a pooled page embedding, the machine learning model trained with training data including pages visited by users during their respective sessions; and
based at least in part on the pooled page embedding, determining, by the computing system, a plurality of candidate pages as potential recommendations for the user, the plurality of candidate pages associated with the trending content.