Article,
Setswana Noun Analyzer and Generator
International Journal of Computational Linguistics (IJCL), 9 (2): 32-40 (June 2018)
Abstract
Word morphology is a process of analysing word formation. Morphological analysis is one of the pre-processing steps in natural language processing tasks. Few studies have looked at Setswana noun morphology analysis and generation computationally. In this paper we present a rule-based Setswana noun morphological analyzer and generator. The analyser and generator implement morphological rules which are supported by a dictionary of root words with some attributes. Results show that Setswana nouns could mostly be analysed using morphological rules and the rules could also be used to generate other words. Adjectives, pronouns, adverbs and enumeratives are also included. The generator shows that Setswana nouns, adjectives and adverbs are less productive compared to verbs. The analyzer gives a 79% performance rate and the generator 92%. The analyser rules fail when multiple words have the same intermediate word and with homographs. The generator failures are due to over generation and under generation.
