Synonyms Based Term Weighting Scheme: An Extension to TF.IDF

Information retrieval is pivotal task in any web search and navigation on World Wide Web. Therefore effective document retrieval techniques can affect not only the efficiency of search engines but also the users experience in terms of relevant information. TF.IDF (Term Frequency Inverse Document Frequency) is the most widely used weighting scheme for key words to facilitate the relevant documents each. But whereas the existing TF.IDF approach does not take into account the semantic correlations between the terms which may lead to less relevant document retrieval. In order to overcome this challenge, we propose a novel Synonyms-Based Term weighting scheme (SBT) which changes Inverse Document Frequency (IDF) according to the synonyms based cluster of any term. We have employed MeSH to compute the dynamic cluster of synonyms of the terms in biomedical text documents. The efficacy of the proposed scheme is corroborated through experiments.