Open AccessA System for Machine Translation from English to Ebira using the Rule-based Approach
Author(s) -
Tijani Musari Abdulmusawir,
Sani Felix Ayegba,
Yahaya Musa Kayode,
Eze Christian Chinemerem
Publication year - 2021
Publication title -
journal of scientific research and reports
Language(s) - English
Resource type - Journals
ISSN - 2320-0227
DOI - 10.9734/jsrr/2021/v27i1130465
Subject(s) - computer science , machine translation , artificial intelligence , universal networking language , natural language processing , language industry , linguistics , natural language , comprehension approach , philosophy
This research work is aimed at bridging the knowledge gap between the most popular knowledge rich English language and the minority Ebira language spoken by the Ebira people, a minority ethnic group in part of Nigeria. Across the globe and on the internet, English language has become the most widely used language for knowledge dissemination. And presently, the majority of the indigenous people of Ebiral and also known as “Anebira” are still not proficient in their use of English language which as a result prevents them from gaining full knowledge disseminated in English language. Hence, the need to develop an automated Machine Translation System capable of translating English text to Ebira text which will help the people to tap from the abundant knowledge conveyed in English language for effective and fast development in their social, political, scientific, philosophical and economic areas of life. The system was designed to consolidate on human translators’ effort and not to replace them. A comprehensive study and analysis of the two languages was carried out with the help of Ebira native speakers in Ebiraland Kogi central and some professional English language tutors at FCE Okene. The knowledge gathered provided the basis for the design and testing of the rule base, inference engine, bilingual dictionary which are important components for the proposed automated system for translation of English text to Ebira text using PHP. Making use of the word in the bilingual dictionary, the system will successfully translate your English text to Ebira. The system was evaluated using one of the popular automatic method of evaluating MT systems BLEU (Bilingual Evaluation Understudy). And an accuracy of 81.5% in translation was achieved. An improved system in the future is recommended to accommodate more complex sentences for the more benefit of the good people of Enebira.