Many data mining techniques have been proposed so far to mine text documents. However, those are all not giving importance to similarity and relatedness between words. Since most existing text mining methods adopted term-based approaches, they all suffer from the problem of synonymy between words. Synonym is a word or phrase that means exactly or nearly the same as another word or phrase in the same language. This paper presents an innovative and effective text mining technique which includes the processes of detecting and scoring of similar and closely related words, to improve the efficiency of the text mining. Substantial experiments on an open forum demonstrate that the proposed solution achieves encouraging performance. Here similar words refers to the words which are similar by meaning and relatedness refers to the contextual relationship between words.
