Conversion of the Vietnammese Grammar into Sign Language Structure using the Example-Based Machine Translation Algorithm

This study has been trying to find the existence of correlation Vietnamese and Vietnamese sign language (VSL) based on kinds of words. The conversion of the Vietnamese sentence structure into the shortened structure of the VSL is normally without rules. In order to solve that problem, a method of machine translation has been applied. The correlation between these two languages will be analyzed by word processing tools and synthesized from the VSL experts’ corrections to develop training datasets. The study has compared the use of words or types of words to construct the training datasets. It has been also evaluated using the example-based machine translation algorithm (EBMT) and the Translation Error Rate (TER). The results indicate that the accuracy of conversion is 97.42% which can use for converting of the Vietnamese text into VSL.