Morphological Analysis of Malay Words for Resolving Ambiguity

The issue of morphological uncertainty is broadly tended to in the cutting edge in Natural Language Processing (NLP). For the most part, vagueness is settled with the utilization of substantial physically explained corpora and machine learning. Be that as it may, such strategies do not generally accessible, as great preparing information is not available for all dialects. In this paper, we introduce a technique for disambiguation without highest quality level corpora utilizing a few factual models, to be specific, Braille Translation Algorithms and unambiguous N-grams from the naturally explained corpus. Every one of the strategies was tried on the Corpus of Glosbe and on the Corpus of Dewan Bahasa Pustaka (DBP). Therefore, more than a half of words with uncertain examinations were disambiguated in the two corpora, exhibiting high exactness. Our technique for morphological disambiguation shows that it is conceivable to dispose of a portion of the uncertain examinations in the corpus without particular phonetic assets, just with the utilization of crude information, where all conceivable morphological investigations for each word are shown.