Abstract: Stemming is integral part of many natural language processing and information retrieval application. Stemmer for a given language basically extracts root or base word for the input word. Marathi word being rich in morphological variation requires an efficient stemmer which can deal with various morphological structures associated with words. Marathi WordNet consists mainly of Marathi root and base words with their Part of Speech information which is useful to reduce over and under stemming issues. Proposed system augments rule based approach with WordNet to perform stemming of the Marathi words with the help of Name entity and stem exception dataset.

Keywords: Stemmer, Root words, Marathi WordNet, Stem Word, Suffix, Inflection and Marathi