Logo image
Home Academic units
Sign in
Integrating Vision and Language Datasets to Measure Word Concreteness
Conference paper

Integrating Vision and Language Datasets to Measure Word Concreteness

James Pustejovsky
IJCNLP 2017 (International Joint Conference on Natural Language Processing) , Eighth (Taipei (Taiwan), 11/2017)
11/2017

Abstract

Data sets Corpora (Linguistics) Computational Linguistics
We present and take advantage of the inherent visualizability properties of words in visual corpora (the textual components of vision-language datasets) to compute concreteness scores for words. Our simple method does not require hand-annotated concreteness score lists for training, and yields state-of-the-art results when evaluated against concreteness scores lists and previously derived scores, as well as when used for metaphor detection.

Metrics

7 Record Views

Details