Inferring Binary Relation Schemas for Open Information Extraction

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 555–560, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. Inferring Binary Relation Schemas for Open Information Extraction Kangqi Luo1 and Xusheng Luo2 and Kenny Q. Zhu3 Department of Computer Science & Engineering Shanghai Jiao Tong University, Shanghai, China 1luokangqi@sjtu.edu.cn 2freefish 6174@sjtu.edu.cn 3kzhu@cs.sjtu.edu.cn Abstract This paper presents a framework to model the semantic representation of binary re- lations produced by open information ex- traction systems. For each binary relation, we infer a set of preferred types on the two arguments simultaneously, and gen- erate a ranked list of type pairs which we call schemas. All inferred types are drawn from the Freebase type taxonomy, which are human readable. Our system collects 171,168 binary relations from Re- Verb, and is able to produce top-ranking relation schemas with a mean reciprocal rank of 0.337. 1 Introduction Open information extraction (or Open IE) is a task of extracting all sorts of relations between named entities or concepts from open-domain text cor- pora, without restraining itself to specific rela- tions or patterns. State-of-the-art Open IE sys- tems (Carlson et al., 2010; Fader et al., 2011; Schmitz et al., 2012; Nakashole et al., 2012) ex- tract millions of binary relations with high preci- sion from the web corpus. Each extracted relation instance is a triple of the form ⟨arg1, rel, arg2⟩, where the relation rel is a lexical or syntactic pat- tern, and both arguments are multi-word expres- sions representing the argument entities or con- cepts. Whereas Open IE provides concrete relation in- stances, we are interested in generalizing these instances into more abstract semantic representa- tions. In this paper, we focus on inferring the schemas of binary relations. For example, given the binary relation “play in”, an Open IE system extracts many triples of the form ⟨X, play in, Y ⟩. The following relation triples are extracted in ReVerb: ⟨Goel Grey, played in, Cabaret⟩ ⟨Tom Brady, play in, National Football League⟩ Informally, the goal of our system is to automatically infer a set of schemas such as ⟨t1, play in, t2⟩, where t1 and t2 are two seman- tic types drawn from a standard knowledge base such as WordNet (Miller, 1995), Yago (Suchanek et al., 2007), Freebase (Bollacker et al., 2008), and Probase (Wu et al., 2012), and each such schema can be used to represent a set of “play in” relation instances. For the above example, two possible schemas for “play in” are: ⟨film actor, play in, film⟩ ⟨athlete, play in, sports league⟩ The schema of a binary relation is useful in- formation in NLP tasks, such as context-oriented entity recognition and open domain question an- swering. Suppose we are to recognize the enti- ties in the sentence “Granger played in the NBA”. “Granger” is a highly ambiguous term, while “the NBA” is probably a sports league. Then with the the above relation schemas for “play in”, the entity recognizer knows that “Granger” is more likely to be an …

Content truncated to 500 words