| International Journal of Applied Information Systems |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 13 - Number 4 |
| Year of Publication: 2026 |
| Authors: O.A. Odeniyi, A.R. Akinwumi, O.O. Ogunsuyi |
10.5120/ijca04844dead7cf
|
O.A. Odeniyi, A.R. Akinwumi, O.O. Ogunsuyi . Development of a Web-based Number-to-Yorùbá Name Translation System: A Rule-based Computational Linguistics Approach to Numeral Text Normalisation for a Low-Resource Nigerian Language. International Journal of Applied Information Systems. 13, 4 ( Aug 2026), 37-50. DOI=10.5120/ijca04844dead7cf
Number-to-word conversion is a foundational text-normalisation task underlying text-to-speech synthesis, machine translation, and educational technology. This task remains underserved for Yorùbá, a tonal, morphologically rich language spoken by an estimated 47 million people across Nigeria, Benin, and Togo, and classified as low resource for computational purposes. This study reports the design, live deployment, and empirical evaluation of YorùbáNum, a browser-based system translating cardinal numbers into Standard Yorùbá lexical forms. The Yorùbá numeral system is vigesimal (base-20) and relies on additive, multiplicative, and subtractive morphological operations, making decimal-style digit-to-word mapping inapplicable [12, 23]. The system was implemented as a full-stack web application (React 19/Vite, Express.js/TypeScript) with a recursive rule-based translation engine. The deployed system was evaluated using an exact-match recall test against an independent gold-standard reference [8], a large-scale randomised robustness test, and a measured latency benchmark. Exact-match recall was 86.96% (40/46), with valid, error-free output for all 9,999 sampled numbers and an average latency of 0.0063 milliseconds. A Mean Opinion Score (MOS) usability evaluation with 62 native Yorùbá speakers returned an overall MOS of 3.91 out of 5.00. The paper contributes a live translation tool and a transparent evaluation benchmark for Yorùbá numeral-translation research.