Statistical Machine Translation Model for Enhancing Multilingual Education Access
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The development of Statistical Machine Translation (SMT) is a breakthrough that affects the accessibility of multilingual education by helping to automatically translate academic material in a variety of languages. Its systematic, clear strategy renders it applicable to education settings that have limited resources. But currently prevailing SMT and neuralbased systems fail on low resource languages, inconsistent terminologies, and errors in domain-specific translations, that are barriers to learning. To solve these problems, it suggests an SMT framework that uses Minimum Bayes Risk (MBR) Decoding, which produces better translation output by minimizing expected error between the translation output and a set of candidate hypothesis. The strategy guarantees increased accuracy, reliability and domain congruency in the educational content. The suggested approach will allow to correctly translate textbooks, assessments and teaching materials, having an effect on the provision of fair learning opportunities. The results suggest that MBR-based SMT is very effective in promoting the quality of translation, preserving the terminological consistency, and promoting access to education by underrepresented language communities.
Fields of Study & Subjects
Natural Language Processing Techniques
Translation Studies and Practices
Speech Recognition and Synthesis
Publication details
- Journal / source
- International Conference on Intelligent Systems for Communication, IoT and Security (ICISCoIS)
- Pages
- 1–6
- Publication date
- 8 January 2026
- Language
- EN