WebSim: A Pathway to Unveiling Term Relationships using a Web Search Technology

We present WebSim (Web-based Similarity metric), whose feature extraction and similarity model is based on a conventional Web search engine. By utilizing the search engine, we can obtain the freshest content for each term that represents the up-to-date knowledge on the term. In comparison with previous text mining approaches that use the certain amount of crawled Web documents as corpus, our method is less sensitive to the problem of data sparseness since we access as much content as possible using the search engine. We also present a method on how to deal with ambiguous terms for the similarity computation. Moreover, we show how existing ontologies can be modified, and demonstrate the characteristics of WebSim by coupling with WordNet. Experimental results show that WebSim can uncover topical relations between terms that are not shown in conventional concept-based ontologies.