Research Article
Evolution and Comprehensive Analysis of Machine Translation Systems for Regional Languages
@INPROCEEDINGS{10.4108/eai.19-12-2025.2362616, author={Sachin Majithia and Harjeet Singh and Rajeev Puri}, title={ Evolution and Comprehensive Analysis of Machine Translation Systems for Regional Languages}, proceedings={Proceedings of the 1st International Conference on Computational Techniques in Data Science, IICTDS 2025, 19-20 December 2025, Chandigarh, India}, publisher={EAI}, proceedings_a={IICTDS}, year={2026}, month={11}, keywords={Machine Translation regional languages statistical models neural network Hindi Punjabi deep learning machine translation}, doi={10.4108/eai.19-12-2025.2362616} }- Sachin Majithia
Harjeet Singh
Rajeev Puri
Year: 2026
Evolution and Comprehensive Analysis of Machine Translation Systems for Regional Languages
IICTDS
EAI
DOI: 10.4108/eai.19-12-2025.2362616
Abstract
India, which is a vast multilingual community, has 22 Indian Languages approved by the constitution, so there is much need of machine translation of documents between different language pairs. They have similar scripts and grammar. Because of all of these factors, language translation is essential. Over the last several years, much research has been going on in Machine Translation in India. During the early 90s, advanced studies in natural language processing made more favorable machine translation systems. This aided in the creation of practical machine translation systems in a few specific domains [4]. This research article presents the evolution of Machine Translation for regional languages from rule-based approaches to Neural Architectures, which deliver near-human quality machine translations. Most of the Machine Translation systems are evaluated using standard metrics like accuracy and BLEU score. We compare BLEU, Accuracy, ChrF Score and Comet Score for the evaluation of machine translation approaches used for regional Languages.