Analysis of Various Stemming Algorithms for Limbu Language

Natural Language Processing requires various preprocessing before words are processed by the system. Stemming plays a very crucial role as a preprocessing technique, which extracts the root word (stem) or lemma from the lexeme. Raw texts can be inflected or derivative words. Those words require cleansing before analysis which can be done by removing affixes or with the required morphological analysis. This paper analyzes various stemming algorithms for Limbu language. We explore the prevalent utilization of stemming algorithms, in both common practice and specifically tailored for less-resourced languages. We conduct a comprehensive assessment and analysis of existing stemmers on the Limbu language, examining their performance. Drawing from this comparative study and experimentation, we introduce an innovative model for the Limbu language that enhances certain aspects of established algorithms.