Word Activation Forces: Distinctive Statistics Revealing Word Associations

Authors:
Jun Guo;Guang Chen;Weiran Xu
Affiliations:
School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China;School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China;School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China
Venue:
Wireless Personal Communications: An International Journal
Year:
2012

Citing 9
Cited 0

Word association norms, mutual information, and lexicography

Computational Linguistics
A vector space model for automatic indexing

Communications of the ACM
Contextual correlates of synonymy

Communications of the ACM
Automatic word sense discrimination

Computational Linguistics - Special issue on word sense disambiguation
Automatic retrieval and clustering of similar words

COLING '98 Proceedings of the 17th international conference on Computational linguistics - Volume 2
Finding predominant word senses in untagged text

ACL '04 Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics
Cross-talk between Language Processes and Overt Motor Behavior in the First 200 msec of Processing

Journal of Cognitive Neuroscience
Dependency-Based Construction of Semantic Space Models

Computational Linguistics
A structured vector space model for word meaning in context

EMNLP '08 Proceedings of the Conference on Empirical Methods in Natural Language Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Word activation has caused ample investigations in many different scientific areas. Various theories have long been debated in predicting and interpreting the fundamental language phenomenon. From a perspective of mechanics, this study considers the word activations as imaginary forces and quantifies their amount by adapting the formula of the universal gravitation to the corresponding imaginary masses and distance that are estimated via the statistics of language experience. In large scale experiments, we found that the word activation forces not only straightforwardly predict various kinds of word activations, but also lead to a simple and human-comparably accurate measure to identify word closest associates including synonyms, near-synonyms, antonyms and similar functional words. The plausibility of identified closest associates with over 10,000 popular English words is highly inspiring.