Experiments in Idiom Recognition

Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, pages 2752–2761, Osaka, Japan, December 11-17 2016. Experiments in Idiom Recognition Jing Peng and Anna Feldman Department of Computer Science Department of Linguistics Montclair State University Montclair, New Jersey, USA 07043 Abstract Some expressions can be ambiguous between idiomatic and literal interpretations depending on the context they occur in, e.g., sales hit the roof vs. hit the roof of the car. We present a novel method of classifying whether a given instance is literal or idiomatic, focusing on verb-noun constructions. We report state-of-the-art results on this task using an approach based on the hypothesis that the distributions of the contexts of the idiomatic phrases will be different from the contexts of the literal usages. We measure contexts by using projections of the words into vector space. For comparison, we implement Fazly et al. (2009)’s, Sporleder and Li (2009)’s, and Li and Sporleder (2010b)’s methods and apply them to our data. We provide experimental results validating the proposed techniques. 1 Introduction Researchers have been investigating idioms and their properties for many years. According to traditional approaches, an idiom is — in its simplest form— a string of two or more words for which meaning is not derived from the meanings of the individual words comprising that string (Swinney and Cutler, 1979). As such, the meaning of kick the bucket (‘die’) cannot be obtained by breaking down the idiom and an- alyzing the meanings of its constituent parts, to kick and the bucket. In addition to being influenced by the principle of compositionality, the traditional approaches are also influenced by theories of generative grammar (Flores, 1993; Langlotz, 2006) The properties that traditional approaches attribute to idiomatic expressions are also the properties that make them difficult for generative grammars to describe. For instance, idioms can be syntactically ill-formed (e.g., by and large), resistant to grammatical transfor- mations (e.g., the bucket was kicked by him ̸= ‘die’), impervious to lexical substitutions (e.g., kick the pail ̸= ‘die’), and semantically ambiguous without context. This last property of the idioms is what we address in our work. The examples below illustrate the ambiguity1. (A1) After the last page was sent to the printer, an editor would ring a bell, walk toward the door, and holler ” Good night! ” (Literal) (A2) His name never fails to ring a bell among local voters. Nearly 40 years ago, Carthan was elected mayor of Tchula. . . (Idiomatic) (B1) . . . that caused the reactor to literally blow its top. About 50 tons of nuclear fuel evaporated in the explosion. . . (Literal) (B2) . . . He didn’t pound the table, he didn’t blow his top. He always kept his composure. (Idiomatic) (C1) . . . coming out of the fourth turn, slid down the track, hit the inside wall and then hit the attenuator at the start of pit road. (Literal) (C2) . . . job training, research and more have hit a …

Content truncated to 500 words