US 12,332,939 B2
Virtual knowledge graph construction for zero-shot domain-specific document retrieval
Yeon Seonwoo, Daejeon (KR); Seunghyun Yoon, San Jose, CA (US); Trung Huu Bui, San Jose, CA (US); Franck Dernoncourt, San Jose, CA (US); Roger K. Brooks, Palo Alto, CA (US); and Mihir Naware, Redwood City, CA (US)
Assigned to ADOBE INC., San Jose, CA (US)
Filed by ADOBE INC., San Jose, CA (US)
Filed on Jun. 24, 2022, as Appl. No. 17/808,599.
Prior Publication US 2023/0418868 A1, Dec. 28, 2023
Int. Cl. G06F 16/90 (2019.01); G06F 16/901 (2019.01); G06F 16/903 (2019.01); G06F 16/9038 (2019.01); G06F 16/93 (2019.01)
CPC G06F 16/9024 (2019.01) [G06F 16/90335 (2019.01); G06F 16/9038 (2019.01); G06F 16/93 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method for text processing, comprising:
receiving a query comprising a natural language expression;
extracting a plurality of query mentions from the query;
generating a virtual knowledge graph of the query by masking each of the plurality of query mentions and generating, using a relation encoder network, a query relation vector between a pair of the plurality of query mentions;
identifying a virtual knowledge graph of a document including a document relation vector between a pair of a plurality of document mentions;
computing a graph similarity score between the virtual knowledge graph of the query and the virtual knowledge graph of the document based on the query relation vector, the document relation vector, and a match between the pair of the plurality of document mentions of the document relation vector and the pair of the plurality of query mentions of the query relation vector by computing a relation similarity between the query relation vector and the document relation vector based on the match; and
transmitting a response to the query based on the document and the graph similarity score.