Constructing a Computational Lexicographic Framework for a Comprehensive Digital Yoruba Dictionary

The computational lexicography of under-resourced languages requires the development of specialised theoretical and technical frameworks that go beyond adaptation of existing European language models to address the structural, tonal, and dialectal specificities of each target language. This dissertation constructs a computational lexicographic framework for a comprehensive digital Yoruba dictionary, making an original contribution to both Yoruba lexicography and African computational linguistics. The framework is developed through three integrated research strands: a systematic review of existing Yoruba lexicographic resources and their gaps; the development of a formal linguistic specification for Yoruba dictionary entries incorporating tone marking, grammatical category notation, dialectal tagging, and collocational information; and a proof-of-concept prototype development using natural language processing tools adapted for tonal language lexicographic applications. The theoretical framework integrates lexicographic theory, specifically Wiegand's general theory of lexicography, with computational linguistics and Yoruba descriptive grammar. Existing Yoruba dictionaries, including those by Afolabi Olabimtan and the Dictionary of Modern Yoruba, lack digital searchability, tonal annotation, and systematic coverage of dialectal variation. This dissertation fills that critical resource gap while making theoretical contributions to computational lexicography for tone languages. Findings will demonstrate the feasibility of a tone-aware Yoruba digital dictionary architecture and identify key natural language processing challenges specific to Yoruba lexicographic data. Recommendations will address a collaborative national Yoruba lexicographic project and international partnership with African language natural language processing research initiatives. Keywords: computational lexicography, digital dictionary, Yoruba, natural language processing, tone languages.