Abstract
ABSTRACT Stemming is a technique used to transform words to their root forms. It is used in various Natural Language Processing applications to improve performance and accuracy. In this paper, the development and evaluation of the first stemmer for Kreol Morisien (KM), the native language of Mauritius, is described. This proposed stemmer is expected to unlock countless opportunities for more efficient applications in KM. The hybrid approach, which combines both rule-based and table-lookup techniques, is presented. The results of the three-fold evaluation, consisting of direct evaluation, gold standard assessment and indirect evaluation are also presented.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call