Semantic based orthographic with prepositional phrase for English-Tamil translation

Abstract

Machine translation quality has improved substantially in recent years. Many researchers have contributed and developed quite good algorithms, frameworks and models for various languages. Processing of natural language involves various levels, complexities, ambiguities arise at each of those levels. Some pragmatic and semantic approaches can be used to tackle these issues. Generally, prepositions are plays sound role in meaningful translation for any languages. While translating from English to Tamil, preposition in English sentences should be translated to postposition to have meaningful sentences. Thus the prepositional phase errors and orthographic errors are the major issue in machine translation. The main goal of this paper is to improve the translation quality. To achieve the goal, we use some semantic rule to correct the prepositional and orthographic errors in this English-Tamil translation. The proposed approach takes as input an English sentence containing a preposition and the correct postposition and correct spelling in Tamil for that particular sentence context as output. The rules are evaluated with corpus data and the performance is good. The outcomes of the results are compared with Google Translate.