SEMANTIC AND AUTOMATIC INDEXING: study of subject analysis of thesis and dissertations

Abstract

This study aims at evaluating the specific contributions of automatic indexation techniques for the semantic representation process of dissertations’ and thesis’ contents. It describes the manual and automatic indexation processes and approaches the application of the semantic-syntactical criteria used in the automatic extraction of relevant terms in the representation of academic documents. It discusses theoretical references drawn from the Semantics and Computational Linguistics. To implement the automatic indexation process the parser Tropes for automatic extraction of the terms and the Taxonomy of Information Science are presented (Hawkins, Larson and Caton, 2003) as a semantic context built in the software.