Representing Verbs with Rich Contexts: an Evaluation on Verb Similarity

TL;DR

This paper introduces a new dependency-based DSM for verbs that captures rich syntactic contexts, improving verb similarity assessments and addressing data sparsity issues.

Contribution

It proposes a novel joint dependency-based context representation for verbs, enhancing similarity evaluation and robustness over traditional separate feature approaches.

Findings

Joint contexts outperform single dependencies in verb similarity tasks.

The approach mitigates data sparsity despite limited training data.

Performance is comparable or better than existing methods.

Abstract

Several studies on sentence processing suggest that the mental lexicon keeps track of the mutual expectations between words. Current DSMs, however, represent context words as separate features, thereby loosing important information for word expectations, such as word interrelations. In this paper, we present a DSM that addresses this issue by defining verb contexts as joint syntactic dependencies. We test our representation in a verb similarity task on two datasets, showing that joint contexts achieve performances comparable to single dependencies or even better. Moreover, they are able to overcome the data sparsity problem of joint feature spaces, in spite of the limited size of our training corpus.