Automated Concept Extraction From Plain Text

We describe a system for extracting concepts from un-structured text. We do this by identifying relation-ships between words in the text based on a lexical database and identifying groups of these words which form closely tied conceptual groups. The word rela-tionships are used to create a directed graph, called a Semantic Relationship Graph (SRG). This SRG a robust representation of the relationships between word senses which can be used to identify the individ-ual concepts which occur in the text. We demonstrate the usefulness of this technique by creating a classifier based on SRGs which is considerably more accurate than a Naive Bayes text classifier