Abstract: Word sense disambiguation (WSD) is still an open research area in natural language processing and computational linguistics. It is from both theoretical and practical point of view. Here, the problem is to find the sense for word in given a context, It is a technique of natural language processing(NLP) ,which requires queries and documents in NLP or texts from Machine Translation (MT). MT is an automated system which involves Marathi, Urdu, Bengali, Punjabi, Hindi, and English etc. Most of the work has been completed in English, and now the focus has been shifted to other languages. The applications of WSD are disambiguation of content in information retrieval (IR), machine translation (MT), speech processing, lexicography, and text processing. In this paper, we have used knowledge based approach along with selectional restriction. It is used to block the formation of component word meanings representation that contains selectional restriction. We have developed a WSD tool using Hindi wordnet. Wordnet is built from co-occurrence, and collocation and it includes synset or synonyms which belong to either noun, verb, adjective, or adverb. In this paper we shall discuss the implementation of our tool and its evaluation.