US 11,676,001 B2
Learning graph representations using hierarchical transformers for content recommendation
Jian Jiao, Bellevue, WA (US); Xiaodong Liu, Redmond, WA (US); Ruofei Zhang, Redmond, WA (US); and Jianfeng Gao, Woodinville, WA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Nov. 9, 2020, as Appl. No. 17/93,426.
Claims priority of provisional application 63/072,770, filed on Aug. 31, 2020.
Prior Publication US 2022/0067030 A1, Mar. 3, 2022
Int. Cl. G06N 3/045 (2023.01)
CPC G06N 3/045 (2023.01) 20 Claims
OG exemplary drawing
 
1. A computer-implemented method of completing an incomplete triplet in a knowledge graph comprising:
receiving, by a transformer, a source entity-relation pair input from the knowledge graph;
capturing, by the transformer, interaction information for the source entity-relation pair input;
outputting, by the transformer, link predictions based on the interaction information;
ranking the link predictions based on a plausibility score;
selecting the highest ranked link prediction to be a target node for the incomplete triplet; and
adding the target node to the incomplete triplet in the knowledge graph.