Resolution of the noun attachment problem in Urdu machine translation

This work addresses the problem of resolving the noun attachment problem in Urdu machine translation. This is basically an ambiguity issue which arises when the source language is to be translated into the target language. In our approach, the input to the system is a sentence in Urdu, and its output determines if the nouns that occur together in the sentence are attached or are completely separate. We have developed a conceptual model for the resolution of this problem which revolves around prepositions present in simple declarative sentence of Urdu. In this paper we present the readers with the algorithms developed and experimental results of the proposed procedure. In the end, some enhancement strategies of the methodology are also presented.