Detection and Suggestion English Grammar Errors for MyanmarEnglish Machine Translation System

Abstract

This paper presents a system for detection and suggestion English grammar errors in Myanmar-English statistical machine translation system. There are very few spelling errors in the translation output, because all words come from Myanmar-English Bilingual corpus. However, the translated sentences might be incomplete in grammar because the syntactic structures of Myanmar and English language are totally different. Therefore, we propose a chunk-based grammar checker by using Naïve Bayesian theory and rule-based model. Context free grammar (CFG) is also applied to make chunkbased sentence pattern. The syntactic chunk structure of a sentence is used to recognize grammatical relations of chunks. This system evaluated the performance of detection and suggestion for simple, compound and complex sentence types. The effectiveness of the system can be confirmed through the experimental results.