TY - JOUR AU - A. Aina, Akindele AU - Táíwò, Oyè PY - 2019/01/01 Y2 - 2024/03/28 TI - Yorùbá Noun Ontology from Functional Perspectives JF - Journal of The Linguistic Association of Nigeria JA - JOLAN VL - 22 IS - 1 SE - Articles DO - UR - https://jolan.com.ng/index.php/home/article/view/221 SP - 169-211 AB - <p>The present shift in language analysis from theoretical and descriptive analysis to applied, adaptable and implementation design has being the practice resulting from modern Information and Communication Technology world. It is noteworthy that Awobuluyi as a renowned Yorùbá scholar have always maintained the positon that a good grammar of any language should be from functional perspectives. Eventually, Ontological annotation as a modern practice by which theoretical knowledge occurring in its implicit form is prepared in both human and machine readable format to aid Natural Language Processing (NLP) is an edge in development. However, the fact that Yorùbá Scholars have considered other perspectives to identify and classify Yorùbá Noun make the knowledge heterogeneous and a big challenge for machine accessibility. This study in order to meet this challenge, extracts the implicit knowledge of Yorùbá nouns being described from Awobuluyi’s functional perspectives, to implements an ontology annotation model named Yorùbá Noun Ontology according to Awobuluyi (YORNOA), in an explicit, formal and machine readable format which make the knowledge available for NLP. The data for this study is the classification of Yorùbá noun as contained in Awobuluyi (1978)’s <em>Essentials of Yorùbá Grammar </em>and Awobuluyi (2013)’s <em>E</em><em>̩̀ ko</em><em>̩̀ Gírámà Yorùbá</em>. An additional 106 Nouns were purposively selected from <em>Modern Yorùbá Dictionary </em>to complement&nbsp;the twenty three (23) concept term extracted from the original text. Methontology as proposed by Gómez- Pérez, Fernandez-Lopez, and Corcho (2004) was followed to design the annotation, and Protégé 4.5 was used to implement the model. From the implementation thirteen (13) subclasses were specified of which two (2) are contrastive, the rest were non-contrastive because the (+) value that is&nbsp;returned for Human nouns cannot be returned for Non- Human nouns, (-) value was returned. This study demonstrate and concludes that the knowledge issues proposed from Awobuluyi functional perspectives can be extracted and presented in the format to shore up knowledge representation for NLP activities.</p> ER -