ABSTRACT

In this paper, a new stemmer has been proposed named as “Maulik” for Hindi Language. This stemmer is purely based on Devanagari script and it uses the Hybrid approach (combination of brute force and suffix removal approach). Stemming can be used to improve the effectiveness of information retrieval. The proposed stemmer is both computationally inexpensive and domain independent. The results are favorable and indicate that the proposed stemmer can be used effectively in Information Retrieval systems. This stemmer also reduces the problem of over-stemming and under-stemming which was found in A Light weight Stemmer for Hindi.