Amharic-Kistanigna Bi-directional Machine Translation using Deep Learning

Machine translation is an application of NLP, which can be used to translate text from one natural language to another natural language. This research attempted to design Amharic-Kistanigna bidirectional machine translation. Previously, there is no research conducted on machine translation between Kistanigna and Amharic. Due to an increase in the number of language users, to address the issues of the endangered of the Kistanigna language and to increases the content of the language in web, it is essential to develop machine translation between Kistanigna and Amharic language. Neural Machine Translation (NMT) is a recently proposed approach to machine translation that has achieved the state-of-the-art translation quality. So, we conduct the experiments using five deep learning encoder-decoder model. Such as, LSTM, Bi-LSTM, LSTM + attention, CNN + attention, and Transformer. We have used 9,225 parallel sentences for each model. Finally, we proposed Transformers model with a BLEU score of 7.73 and 4.43 from Amharic to Kistanigna and Kistanigna-Amharic translation respectively. The lack of sufficient data to conduct an extensive experiment is the study’s major weakness. As a result, additional parallel corpora for conducting similar research are required.