A semantic similarity approach based on web resources
M. Karthiga, P. C. D. Kalaivaani, S Sankarananth · 2013
The ability to accurately judge the semantic similarity is important in various tasks on the web such as extracting the relation, document clustering, and automatic metadata extraction. An empirical method is proposed to provide a semantic wise search that uses in one hand, a technical English dictionary and on the other hand, a page count based metric and a text snippet based metric retrieved from a web search engine for two words. To identify the numerous semantic relations between the words, a novel pattern extraction algorithm and a pattern clustering algorithm is proposed. The page counts based co-occurrence measures and lexical pattern clusters extracted from snippets is learned using support vector machines. Integrate the page count, text snippet and dictionary based metric to accurately measure the semantic similarity search compared to normal search.