Re-Representing Metaphor: Modeling Metaphor Perception Using Dynamically Contextual Distributional Semantics

• Published in: Frontiers in psychology Vol. 10; p. 765
• Main Authors: McGregor, Stephen, Agres, Kat, Rataj, Karolina, Purver, Matthew, Wiggins, Geraint
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Media S.A 15.04.2019
• Subjects: computational creativity; computational linguistics; conceptual models; distributional semantics; metaphor; Psychology; vector space models; computational creativity; vector space models; conceptual models; metaphor; computational linguistics; distributional semantics
• ISSN: 1664-1078; 1664-1078
• DOI: 10.3389/fpsyg.2019.00765

In this paper, we present a novel context-dependent approach to modeling word meaning, and apply it to the modeling of metaphor. In distributional semantic approaches, words are represented as points in a high dimensional space generated from co-occurrence statistics; the distances between points may then be used to quantifying semantic relationships. Contrary to other approaches which use static, global representations, our approach discovers contextualized representations by dynamically projecting low-dimensional subspaces; in these spaces, words can be re-represented in an open-ended assortment of geometrical and conceptual configurations as appropriate for particular contexts. We hypothesize that this context-specific re-representation enables a more effective model of the semantics of metaphor than standard static approaches. We test this hypothesis on a dataset of English word dyads rated for degrees of metaphoricity, meaningfulness, and familiarity by human participants. We demonstrate that our model captures these ratings more effectively than a state-of-the-art static model, and does so via the amount of contextualizing work inherent in the re-representational process.