An approach for measuring semantic similarity between words using multiple information sources

Li, Y; Bandar, ZA; McLean, D

Abstract

Semantic similarity between words is becoming a generic problem for many applications of computational linguistics and artificial intelligence. This paper explores the determination of semantic similarity by a number of information sources, which consist of structural semantic information from a lexical taxonomy and information content from a corpus. To investigate how information sources could be used effectively, a variety of strategies for using various possible information sources are implemented. A new measure is then proposed which combines information sources nonlinearly. Experimental evaluation against a benchmark set of human similarity ratings demonstrates that the proposed measure significantly outperforms traditional similarity measures.

Citation

Li, Y., Bandar, Z., & McLean, D. (2003). An approach for measuring semantic similarity between words using multiple information sources. IEEE Transactions on Knowledge and Data Engineering, 15(4), 871-882. https://doi.org/10.1109/TKDE.2003.1209005

• Journal Article Type: Article
• Publication Date: Jul 1, 2003
• Deposit Date: Jul 28, 2015
• Journal: IEEE Transactions on Knowledge and Data Engineering
• Print ISSN: 1041-4347
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Volume: 15
• Issue: 4
• Pages: 871-882
• DOI: https://doi.org/10.1109/TKDE.2003.1209005
• Publisher URL: http://dx.doi.org/10.1109/TKDE.2003.1209005
• Related Public URLs: http://ieeexplore.ieee.org/xpl/aboutJournal.jsp?punumber=69
• Additional Information: Additional Information : This paper rigorously investigates the contributions of different information sources to similarity between words. It presents word similarity measures by nonlinearly combining structural semantic information from lexical taxonomy and information content from corpus. Our approach outperforms previously published measures: best published correlation against the benchmark set of word pairs of Rubenstein-Goodenough's human similarity ratings has been 0.8484, whilst ours is 0.8914. The paper has been cited over 70 times (SCI) and 300 times (Google Scholar) as of Jan 2010, selected as advanced reading material in CIS526 Machine Learning, Temple University, Philadelphia, and adopted by other researchers in real system developments, e.g., Bibster - a semantics-based bibliographic peer-to-peer system.