JSE_V1_N2_A3 T. Jenitha Jeba Sheeli V. Joseph Raj Journal on Software Engineering 2230 – 7168 1 2 27 33 Parsing, Genetics, Neural network, Natural language, Syntactic structure Natural Language Understanding Interface allows the user to interact with the computer in a flexible and friendly manner. This paper describes a neural network model of syntactic parser for a Natural language understanding interface. This intelligent parser performs the parsing in two steps: Parts-of-speech tagging is carried out by a Back Propagation Network that takes an English word as input and generates as output the associated parts-of-speech tags; parse structure generation is done by a similar intelligent network that uses the output of the parts-of-speech tagging network and grammar rules. The proposed Back Propagation Network uses genetic algorithm for weight determination. Being a robust and optimization technique, Genetics Algorithm (GA) outperforms the gradient based conventional training algorithm, giving solution fairly accurately and quickly in less number of iterations. October - December 2006 Copyright © 2006 i-manager publications. All rights reserved. i-manager Publications http://www.imanagerpublications.com/Article.aspx?ArticleId=769