Semi-Automatic Document Processing in An Experimental Model of information Retrieval System

Pačovski, Veno and Kon-Popovska, Margita (2003) Semi-Automatic Document Processing in An Experimental Model of information Retrieval System. In: Proceedings of the Fourth Conference on Informatics and Information Technology. Institute of Informatics, Faculty of Natural Sciences and Mathematics, Ss. Cyril and Methodius University in Skopje, Macedonia, Skopje, Macedonia, pp. 138-144. ISBN 9989-668-45-0


Download (568kB) | Preview
Official URL:


The paper presents an approach to document processing in an information retrieval system. Namely, when documents are entered in the Information retrieval system, some kind of language processing has to be applied. Considering the diversity of grammar rules of different natural languages, this processing can be performed by partial expert intervention while entering the documents. In the paper, certain characteristics of documents written in Macedonian language (grammar) and Cyrillic alphabet are discussed and a kind of semi-automatic system approach to document processing and related database structures in order to store parts of the expert knowledge is proposed. The aim is to enable the system to perform automatic language processing with some degree of confidence and use expert interventions, to avoid automatic context analysis. The approach is continually tested on a test-collection containing over 4.000 rather short documents, common news, growing by 10-15 on daily bases. The work represents a part of research in Information retrieval systems, encountering specifics of documents in native language and alphabet, aiming to improve the quality of retrieval in national information retrieval systems.

Item Type: Book Section
Uncontrolled Keywords: retrieval, database, processing, information retrieval, database systems, document processing
Subjects: International Conference on Informatics and Information Technologies > Information systems
Depositing User: Vangel Ajanovski
Date Deposited: 28 Oct 2016 00:15
Last Modified: 28 Oct 2016 00:15

Actions (login required)

View Item View Item