Augmenting Knowledge Graph Hierarchies using Neural Transformers

Tracy Holloway King, Sanat Sharma, Jayant Kumar, Mayank Poddar, Kosta Blank

Abstract: This work leverages neural transformers to generate hierarchies in an existing knowledge graph. For small (<10,000 node) domain-specific KGs, we find that a combination of few-shot prompting with one-shot generation works well, while larger KG may require cyclical generation. Hierarchy coverage increased by 98\% for intents and 95\% for colors.