US 12,353,455 B2
Search results for pseudo-content
Matthew Allen Strong Ross, London (CA); Azadeh Haji Hosseini, Toronto (CA); Monique Alves Cruz, Toronto (CA); Albert Jimenez Sanfiz, Toronto (CA); Prabhdeep Singh Cheema, Dublin, CA (US); and Hima Kiran Alladi, Missouri City, TX (US)
Assigned to Scribd, Inc., San Francisco, CA (US)
Filed by Scribd, Inc., San Francisco, CA (US)
Filed on Mar. 29, 2024, as Appl. No. 18/621,820.
Application 18/621,820 is a continuation in part of application No. 18/117,271, filed on Mar. 3, 2023.
Claims priority of provisional application 63/455,775, filed on Mar. 30, 2023.
Prior Publication US 2024/0296179 A1, Sep. 5, 2024
Int. Cl. G06F 16/334 (2025.01); G06F 16/335 (2019.01)
CPC G06F 16/334 (2019.01) [G06F 16/335 (2019.01)] 14 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
receiving a search query, from a user account, for content submitted to a content corpus;
determining the content identified in the search query is unavailable;
generating a pseudo-identifier for the unavailable content identified in the search query;
inserting the pseudo-identifier into a sequence of content identifiers, the sequence associated with the user account;
generating an embedding for the pseudo-identifier;
identifying, via the embedding, a plurality of sequences of content identifiers, wherein a respective sequence corresponds to a different user account that previously sent a search query for the unavailable content identified by the pseudo-identifier;
for one or more of the respective sequences identified via the embedding:
(i) determining content interactions that occurred within a particular time range from an instance of the pseudo-identifier placed in the respective sequence, each of the determined content interactions associated with a proximate content identifier; and
(ii) identifying neighbor content from at least a portion of the proximate content identifiers;
generating search results responsive to the user account's search query based on the neighbor content.