Amazon was a Silver sponsor of the 2020 Conference on Information and Knowledge Management (CIKM), which provides an international forum for presentation and discussion of research on information and knowledge management, as well as advances on data and knowledge bases. During the conference, which was held Oct. 19 – 23, senior principal scientist Xin Luna Dong presented a keynote talk on harvesting knowledge from the semi-structured web. The mission, Dong explained, is to build a product graph “to answer any question about products and related knowledge in the world.”
Knowledge graphs have been used to support a wide range of applications and enhance search and QA, but we often miss long-tail knowledge, including unpopular entities, unpopular relations, and unpopular verticals, she said. Dong described Amazon’s AutoCeres ClosedIE system, which improves the accuracy of fully automatic knowledge extraction from 60%+ of state-of-the-art to 90%+ on semi-structured data. She also described OpenCeres, the first ever OpenIE system on semi-structured data able to identify new relations not readily included in existing ontologies. Finally, Dong’s keynote described Amazon’s other efforts in ontology alignment, entity linkage, graph mining, and QA that enable Amazon to leverage the knowledge we extract for search and QA.
Watch a replay of Dong’s keynote talk below.
AutoCeres: Harvesting knowledge from semi-structured web | Amazon Science
