Identification of Reduplication in Bengali Corpus and their Semantic Analysis: A Rule Based Approach

In linguistic studies, reduplication generally means the repetition of any linguistic unit such as a phoneme, morpheme, word, phrase, clause or the utterance as a whole. The identification of reduplication is a part of general task of identification of multiword expressions (MWE). In the present work, reduplications have been identified from the Bengali corpus of the articles of Rabindranath Tagore. The present rule-based approach is divided into two phases. In the first phase, identification of reduplications has been done mainly at general expression level and in the second phase, their structural and semantics classifications are analyzed. The system has been evaluated with average Precision, Recall and FScore values of 92.82%, 91.50% and 92.15% respectively.