Word Sense Disambiguation-based Sentence Similarity

Previous works tend to compute the similarity between two sentences based on the comparison of their nearest meanings. However, the nearest meanings do not always represent their actual meanings. This paper presents a method which computes the similarity between two sentences based on a comparison of their actual meanings. This is achieved by transforming an existing most-outstanding corpus-based measure into a knowledge-based measure, which is then integrated with word sense disambiguation. The experimental results on a standard data set show that the proposed method outperforms the baseline and the improvement achieved is statistically significant at 0.025 levels.

• Type: Journal
• Year: 2010
• Where: COLING
• Authors: ChukFong Ho, Masrah Azrifah Azmi Murad, Rabiah Abdul Kadir, Shyamala C. Doraisamy
• Keywords: Actual Meanings, COLING 2010, Computational Linguistics, Most-outstanding Corpus-based Measure, Nearest Meanings