OpenKI: Integrating Open Information Extraction and Knowledge Bases with Relation Inference

Proceedings of NAACL-HLT 2019, pages 762–772 Minneapolis, Minnesota, June 2 - June 7, 2019. c⃝2019 Association for Computational Linguistics 762 OpenKI: Integrating Open Information Extraction and Knowledge Bases with Relation Inference Dongxu Zhang1∗, Subhabrata Mukherjee2, Colin Lockard2, Xin Luna Dong2, Andrew McCallum1 1University of Massachusetts Amherst {dongxuzhang,mccallum}@cs.umass.com 2Amazon {subhomj,clockard,lunadong}@amazon.com Abstract In this paper, we consider advancing web- scale knowledge extraction and alignment by integrating OpenIE extractions in the form of (subject, predicate, object) triples with Knowl- edge Bases (KB). Traditional techniques from universal schema and from schema mapping fall in two extremes: either they perform instance-level inference relying on embedding for (subject, object) pairs, thus cannot handle pairs absent in any existing triples; or they per- form predicate-level mapping and completely ignore background evidence from individual entities, thus cannot achieve satisfying quality. We propose OpenKI to handle sparsity of Ope- nIE extractions by performing instance-level inference: for each entity, we encode the rich information in its neighborhood in both KB and OpenIE extractions, and leverage this in- formation in relation inference by exploring different methods of aggregation and attention. In order to handle unseen entities, our model is designed without creating entity-specific pa- rameters. Extensive experiments show that this method not only significantly improves state-of-the-art for conventional OpenIE ex- tractions like ReVerb, but also boosts the performance on OpenIE from semi-structured data, where new entity pairs are abundant and data are fairly sparse. 1 Introduction Web-scale knowledge extraction and alignment has been a vision held by different communities for decades. The Natural Language Processing (NLP) community has been focusing on knowl- edge extraction from texts. They apply either closed information extraction according to an on- tology (Mintz et al., 2009; Zhou et al., 2005), re- stricting to a subset of relations pre-defined in the ontology, or open information extraction (OpenIE) ∗ This work was performed while at Amazon. to extract free-text relations (Banko et al., 2007; Fader et al., 2011), leaving the relations unaligned and thus potentially duplicated. The Database (DB) community has been focusing on aligning re- lational data or WebTables (Cafarella et al., 2008) by schema mapping (Rahm and Bernstein, 2001), but the quality is far below adequate for assuring correct data integration. We propose advancing progress in this direc- tion by applying knowledge integration from Ope- nIE extractions. OpenIE extracts SPO (subject, predicate, object) triples, where each element is a text phrase, such as E1: (“Robin Hood”, “Full Cast and Crew”, “Leonardo Decaprio”) and E2: (“Ang Lee”, “was named best direc- tor for”, “Brokeback”). OpenIE has been stud- ied for text extraction extensively (Yates et al., 2007; Fader et al., 2011; Mausam et al., 2012), and also for semi-structured sources (Bronzi et al., 2013), thus serves an effective tool for web-scale knowledge extraction. The remaining problem is to align text-phrase predicates1 from OpenIE to knowledge bases (KB). Knowledge integration an- swers the following question: given an OpenIE ex- traction (s, p, o), how can one populate an existing KB using relations in the pre-defined ontology? The problem …

Content truncated to 500 words