Paper: A Semantic Network Approach to Measuring Relatedness

ACL ID C10-2041
Title A Semantic Network Approach to Measuring Relatedness
Venue International Conference on Computational Linguistics
Session Poster Session
Year 2010
Authors

Humans are very good at judging the strength of relationships between two terms, a task which, if it can be au- tomated, would be useful in a range of applications. Systems attempting to solve this problem automatically have traditionally either used relative po- sitioning in lexical resources such as WordNet, or distributional relation- ships in large corpora. This paper pro- poses a new approach, whereby rela- tionships are derived from natural lan- guage text by using existing nlp tools, then integrated into a large scale se- mantic network. Spreading activation is then used on this network in order to judge the strengths of all relation- ships connecting the terms. In com- parisons with human measurements, this approach was able to obtain re- sults on par with the best purpose built sy...