Handling short form words for machine translation

Abstract

The short forms which are being originated from letter omission, word truncation and the substitution of the part of the word (which may be numeric or alphabetic) are being handled by this research. The use of such short word is being increasing day by day because people now these days are very busy; to make the communication fast and in easier manner they use such short word. So in the field of machine translation also such words may be used but the machine has to understand such word and should be capable to translate it. In this research this problem is being handled by tokenizing the input sentence to find the short word then translate the detected short word into corresponding long form that will be in the vocabulary of machine.