Exploring Text Semantics to Extract Key-Fragments for Model Answers

Abstract

In context with the recent developments in the understanding of text semantics at machine level, this paper is an attempt to extract some of the most crucial fragments that play a key role as semantic units in natural language text. The context is intuitively extracted from typed dependency structures basically depicting dependency relations using the relevant Part-Of-Speech tagged representation of the text. These relations imply deep, fine grained, labeled dependencies that encode long-distance relations and passive information. The present work focuses on extracting the key noun phrases participating both in subject and object roles that are intended to be subsequently used in framing sentential components for model answers in any selected working domain.