Literal and Metaphorical Sense Identification through Concrete and Abstract Context

Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, pages 680–690, Edinburgh, Scotland, UK, July 27–31, 2011. c⃝2011 Association for Computational Linguistics Literal and Metaphorical Sense Identification through Concrete and Abstract Context Peter D. Turney Inst. for Info. Tech. NRC Canada Ottawa, Canada peter.turney@nrc-cnrc.gc.ca Yair Neuman Dept. of Education Ben-Gurion Univ. Beer-Sheva, Israel yneuman@bgu.ac.il Dan Assaf Dept. of Education Ben-Gurion Univ. Beer-Sheva, Israel dan.assaf4@googlemail.com Yohai Cohen Gilasio Coding Tel-Aviv, Israel yohai@gilasio.com Abstract Metaphor is ubiquitous in text, even in highly technical text. Correct inference about tex- tual entailment requires computers to distin- guish the literal and metaphorical senses of a word. Past work has treated this problem as a classical word sense disambiguation task. In this paper, we take a new approach, based on research in cognitive linguistics that views metaphor as a method for transferring knowl- edge from a familiar, well-understood, or con- crete domain to an unfamiliar, less understood, or more abstract domain. This view leads to the hypothesis that metaphorical word usage is correlated with the degree of abstractness of the word’s context. We introduce an algorithm that uses this hypothesis to classify a word sense in a given context as either literal (de- notative) or metaphorical (connotative). We evaluate this algorithm with a set of adjective- noun phrases (e.g., in dark comedy, the adjec- tive dark is used metaphorically; in dark hair, it is used literally) and with the TroFi (Trope Finder) Example Base of literal and nonliteral usage for fifty verbs. We achieve state-of-the- art performance on both datasets. 1 Introduction Metaphor is a natural consequence of our ability to reason by analogy (Gentner et al., 2001). It is so common in our daily language that we rarely notice it (Lakoff and Johnson, 1980). Identifying metaphorical word usage is important for reasoning about the implications of text. Past work on the problem of distinguishing lit- eral and metaphorical senses has approached it as a classical word sense disambiguation (WSD) task (Birke and Sarkar, 2006). Here, we take a differ- ent approach to the problem. Lakoff and Johnson (1980) argue that metaphor is a method for trans- ferring knowledge from a concrete domain to an ab- stract domain. Therefore we hypothesize that the de- gree of abstractness in a word’s context is correlated with the likelihood that the word is used metaphori- cally. This hypothesis is the basis for our algorithm for distinguishing literal and metaphorical senses. Consider the following sentences: L: He shot down my plane. → C1: He fired at my plane. ↛ A1: He refuted my plane. M: He shot down my argument. ↛ C2: He fired at my argument. → A2: He refuted my argument. The literal sense of shot down in L invokes knowl- edge from the domain of war. The metaphorical us- age of shot down in M transfers knowledge from the concrete domain of war to the abstract domain of debate (Lakoff and Johnson, 1980). The entailments of L and M depend on the in- tended …

Content truncated to 500 words