Affordable Access

Integrating Distributional Lexical Contrast into Word Embeddings for Antonym-Synonym Distinction

Authors
  • Nguyen, Kim Anh
  • Walde, Sabine Schulte im
  • Vu, Ngoc Thang
Type
Preprint
Publication Date
May 25, 2016
Submission Date
May 25, 2016
Identifiers
arXiv ID: 1605.07766
Source
arXiv
License
Yellow
External links

Abstract

We propose a novel vector representation that integrates lexical contrast into distributional vectors and strengthens the most salient features for determining degrees of word similarity. The improved vectors significantly outperform standard models and distinguish antonyms from synonyms with an average precision of 0.66-0.76 across word classes (adjectives, nouns, verbs). Moreover, we integrate the lexical contrast vectors into the objective function of a skip-gram model. The novel embedding outperforms state-of-the-art models on predicting word similarities in SimLex-999, and on distinguishing antonyms from synonyms.

Report this publication

Statistics

Seen <100 times