ANN and Rule Based Method for English to Arabic Machine Translation

Machine translation is the process by which computer software is used to translate a text from one natural language into another language with or without minimal human intervention. This definition involves accounting for the grammatical structure of each language and using rules and grammars to transfer the grammatical structure of the source language into the target language. This paper presents an English into Arabic Machine Translation (MT) system for translating simple well-structured English sentences into well-structured Arabic sentences using a rule-based approach and feed-forward back-propagation Artificial Neural Network (ANN). Our system is able to translate sentences having gerunds, infinitives, prepositions and prepositional objects, direct objects, indirect objects, etc. Neural network works as bilingual dictionary, which does not only store the meaning of English word in Arabic but it also stores linguistic features attached to the word. The performance is evaluated by different MT evaluation methods. The n-gram blue score achieved by the system is 0.6029, METEOR score achieved is 0.8221 and 0.8386 on F-measure.