Joint Syntactic and Semantic Parsing with Combinatory Categorial Grammar

Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, pages 1188–1198, Baltimore, Maryland, USA, June 23-25 2014. c⃝2014 Association for Computational Linguistics Joint Syntactic and Semantic Parsing with Combinatory Categorial Grammar Jayant Krishnamurthy Carnegie Mellon University 5000 Forbes Avenue Pittsburgh, PA 15213 jayantk@cs.cmu.edu Tom M. Mitchell Carnegie Mellon University 5000 Forbes Avenue Pittsburgh, PA 15213 tom.mitchell@cmu.edu Abstract We present an approach to training a joint syntactic and semantic parser that com- bines syntactic training information from CCGbank with semantic training informa- tion from a knowledge base via distant su- pervision. The trained parser produces a full syntactic parse of any sentence, while simultaneously producing logical forms for portions of the sentence that have a se- mantic representation within the parser’s predicate vocabulary. We demonstrate our approach by training a parser whose se- mantic representation contains 130 pred- icates from the NELL ontology. A seman- tic evaluation demonstrates that this parser produces logical forms better than both comparable prior work and a pipelined syntax-then-semantics approach. A syn- tactic evaluation on CCGbank demon- strates that the parser’s dependency F- score is within 2.5% of state-of-the-art. 1 Introduction Integrating syntactic parsing with semantics has long been a goal of natural language processing and is expected to improve both syntactic and se- mantic processing. For example, semantics could help predict the differing prepositional phrase at- tachments in “I caught the butterfly with the net” and “I caught the butterfly with the spots.” A joint analysis could also avoid propagating syntactic parsing errors into semantic processing, thereby improving performance. We suggest that a large populated knowledge base should play a key role in syntactic and se- mantic parsing: in training the parser, in resolv- ing syntactic ambiguities when the trained parser is applied to new text, and in its output semantic representation. Using semantic information from the knowledge base at training and test time will ideally improve the parser’s ability to solve diffi- cult syntactic parsing problems, as in the exam- ples above. A semantic representation tied to a knowledge base allows for powerful inference op- erations – such as identifying the possible entity referents of a noun phrase – that cannot be per- formed with shallower representations (e.g., frame semantics (Baker et al., 1998) or a direct conver- sion of syntax to logic (Bos, 2005)). This paper presents an approach to training a joint syntactic and semantic parser using a large background knowledge base. Our parser produces a full syntactic parse of every sentence, and fur- thermore produces logical forms for portions of the sentence that have a semantic representation within the parser’s predicate vocabulary. For ex- ample, given a phrase like “my favorite town in California,” our parser will assign a logical form like λx.CITY(x) ∧ LOCATEDIN(x, CALIFORNIA) to the “town in California” portion. Additionally, the parser uses predicate and entity type informa- tion during parsing to select a syntactic parse. Our parser is trained by combining a syntactic parsing task with a distantly-supervised relation extraction task. Syntactic information is …

Content truncated to 500 words