In this paper, a distinctive approach of Machine Translation (MT) from Bangla language to Universal Networking Language (UNL) is proffered. This approach corroborates analyzing Bangla sentences more precisely. The analysis churns out a semantic net like structure expressed by means of UNL. The UNL system comprises two major components, namely, EnConverter (used for converting the text from a Native language to UNL) and DeConverter (used for converting the text from UNL to a Native language). This paper discusses the framework for designing EnConverter for Bangla language with a particular attention on generating UNL attributes and relations from Bangla Sentence input. The structural constitution of Bangla EnConverter, algorithm for understanding the Bangla sentence input and resolution of UNL relations and attributes are also conferred in this paper. The paper highlights the EnConversion analyzing rules for the EnConverter and indicates its usage in generating UNL expressions. This paper also covers the results of implementing Bangla EnConverter and compares these with the system available in a Language Server located in Russia.
|Number of pages||6|
|Journal||ARPN Journal of Engineering and Applied Sciences|
|Publication status||Published - 22 Aug 2015|
- machine translation
- knowledge base
- natural language parsing
- universal networking language