Polygot: An approach towards reliable translation by name identification and memory optimization using semantic analysis

We present a study of improving the efficiency, complexity and performance of language translation process by a translator. The goal of this research is to develop an efficient translation system for any language by optimizing the memory consumption and also identifying the names as nouns efficiently. Although a number of researches can be found on natural language processing in different areas, those were performed keeping English as the only target language mostly. However, a good number of languages remain nearly unexplored in the research fields yet. This study basically focuses on Bengali Language as an example of the unexplored languages. Some noticeable studies on Bengali language are on Bangla keyboard layout design, English to Bangla translator, etc. so far. However, very few researches have been done to translate Bengali text to English till now. To develop an efficient translation system is very complex and expensive as it requires huge amount of time and resources. In all the languages, there are many words having multiple meanings and multiple forms and also some sentences having multiple grammatical structures to express the same meaning. Besides, the names of people may not be easily identified due to the vast diversity of names and also the tags (prefix/ suffix) attached to emphasize names. Therefore, it remains a great challenge to recognize a sentence of a particular language with accurate semantic analysis. However it is very important to have a generalised translation system which can compute various possible outputs in reasonable time and space. In this paper we focus on the correct interpretation of the names as nouns in a sentence and also the optimization of space requirement using semantic analysis.