A Complete Bengali Stop Word Detection Mechanism

Stop word lists of different languages have been developed already around the world. The Bengali language is a grammatically enriched language. It has a large vocabulary containing a massive number of stop words. There is no proper set of stop words for the Bengali language. A proper set of stop word list is highly required in modern information retrieval systems. In this paper, a corpus-based methodology for detection and extraction of Bengali stop word is proposed. We have used machine learning library NLTK from python environment. As there is no introduction and classification available for Bengali stop word, this paper discusses those issues of Bengali Stop word. There is no work available for Bengali stop word removal. This motivates us to develop Bengali stop word elimination algorithm for the Bengali language. Our proposed approach is the first attempt for this task in the Bengali language. In our experiment, we have achieved 100% precision but we have an accuracy of 70 - 75%.