Phrase-Level Metaphor Identification Using Distributed Representations of Word Meaning

Abstract

Metaphor is an essential element of human cognition which is often used to express ideas and emotions that might be difficult to express using literal language. Processing metaphoric language is a challenging task for a wide range of applications ranging from text simplification to psychotherapy. Despite the variety of approaches that are trying to process metaphor, there is still a need for better models that mimic the human cognition while exploiting fewer resources. In this paper, we present an approach based on distributional semantics to identify metaphors on the phrase-level. We investigated the use of different word embeddings models to identify verb-noun pairs where the verb is used metaphorically. Several experiments are conducted to show the performance of the proposed approach on benchmark datasets.

Perceptional and actional enrichment for metaphor detection with sensorimotor norms

“…Another characteristic of metaphor that different theories generally agree upon is the embodied cognition hypothesis: that a metaphor is the use of a more embodied (i.e., concrete) concept to describe a less embodied (i.e., abstract) concept [e.g., Gibbs et al (2004); Lakoff (2012)]. Since embodied concepts are attested by sensory inputs (Barsalou 1999), NLP studies explored the incorporation of sensory input-related resources, such as sensory lexicon, synaesthesia, and vision-based information (Shutova et al 2016;Tekiroglu, Özbal, and Strapparava 2015), property norms (Zayed, McCrae, and Buitelaar 2018), information about concreteness, imageability (Maudslay et al 2020), or emotion (Rai et al 2019), etc. An emergent trend that is highly interdisciplinary is to leverage neuro-cognitive research outcomes in NLP. Although no metaphor processing study has adopted this new paradigm yet, this new trend may well be the next direction to go.…”

A Survey on Computational Metaphor Processing Techniques: From Identification, Interpretation, Generation to Application

“…Zayed et al (2019) presented a crowd-sourcing approach for building a metaphor identification dataset and released one (ZayTw) sourced from Twitter on general and political topics. They applied a weakly supervised classifier (Zayed et al, 2018) on the source data and filtered data with criteria, such as verb balance, sense coverage, and size before crowd-sourcing.…”

Contextual Modulation for Relation-Level Metaphor Identification

“…Over the last decades, the focus of computational metaphor identification has shifted from rule-based (Fass, 1991) and knowledge-based approaches (Krishnakumaran and Zhu, 2007;Wilks et al, 2013) to statistical and machine learning approaches including supervised (Gedigian et al, 2006;Turney et al, 2011;Dunn, 2013a,b;Tsvetkov et al, 2013;Hovy et al, 2013;Mohler et al, 2013;Klebanov et al, 2014;Bracewell et al, 2014;Jang et al, 2015;Gargett and Barnden, 2015;Rai et al, 2016;Köper and Schulte im Walde, 2017), semi-supervised (Birke and Sarkar, 2006;Zayed et al, 2018) and unsupervised methods (Shutova and Sun, 2013;Heintz et al, 2013;Strzalkowski et al, 2013). These approaches employed a variety of features to design their models.…”