Numeric expressions (such as nineteen hundred sixty three or three quarters) and date expressions (such as last weekend or the third quarter of 2004) are widespread in natural language texts and their recognition is both important and relatively easy. Correctly identifying such expressions in texts can greatly reduce the complexity of further processing, such as parsing, and contribute to the computation of the text meaning. In this project you will design and implement a grammar for such expressions in Urdu. The result should be a program whose input is a Urdu text and whose output is the same text, where numeric and date expressions are properly annotated. The grammar will be developed using finite-state technology.
Emoticon Emoticon
Note: only a member of this blog may post a comment.