Keyword Extraction from Hindi Documents Using Document Statistics and Fuzzy Modelling

摘要

In this paper, we put forward a novel unsupervised, domain independent and corpus independent approach for automatic keyword extraction. Our approach combines the document statistics of frequency and spatial distribution of a word in order to extract the keywords. We have extracted keywords from Hindi documents using document statistics and utilized the power of fuzzy logic to combine those document statistics effectively for better results. Further, we use this information to frame fuzzy rules for keyword extraction. Main advantages of our approach are that it uses the fuzzy membership for the variables instead of dealing with crisp thresholds and corpus independent setting of fuzzy membership boundaries. Our work is especially significant in the light that it has been implemented and tested on Hindi which is a resource poor and underrepresented language.