Machine Learning-Based Besemah Language Translator Model with Recurrent Neural Network (RNN) Model Algorithm

Andika, Muhamad (2025) Machine Learning-Based Besemah Language Translator Model with Recurrent Neural Network (RNN) Model Algorithm. -.

Text Machine Learning-Based Besemah Language Translator Model with Recurrent Neural Network (RNN) Model Algorithm.pdf Download (1MB)

Abstract

Abstract—Indonesia consists of various tribes with their respective regional languages, one of which is the Besemah tribe in South Sumatra province with its language culture, namely Besemah Language. Until now Besemah language is still used by the Besemah tribe but over time the number of Besemah language speakers is decreasing not to mention most of the wider community do not know what Besemah language is. Machine translation is a tool that can switch one language to another. This research aims to collect datasets in the form of sentences and words from Besemah Language and then create a Besemah Language translation machine to Indonesian and vice versa. In this research, Neural Machine Translation (NMT) technology with Recurrent Neural Network (RNN) approach is applied. The results for val_accuracy besemah-indonesia is 0.8469 and for Indonesia-besemah get a val_accuracy value of 0.8492, in translation trials conducted using the RNN model, 100 epochs, 64 batch sizes and 0.2 validation split.

Actions (login required)

View Item