A Proposal for A Matching Function for An information Retrieval System

Pačovski, Veno and Kon-Popovska, Margita (2002) A Proposal for A Matching Function for An information Retrieval System. In: Proceedings of the Third 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. 61-69. ISBN 9989-668-36-1


Download (715kB) | Preview
Official URL: http://ciit.finki.ukim.mk


The paper considers the idea of proposing a matching function between document and the query, based, on a objective (statistical) parameters. The function also allows additional qualitative ranking of a subset of retrieved documents, which contain all the words of a query. The function is tested on a test-collection containing over 1000 rather short documents (common news) which is continually growing. Paper represents a part of own continuous research in Information retrieval and as a starting point, we presume previous results: the database of documents is a welldefined structure and computed quotients of relevance, as a kind of measurement for the strength of a document-word relation as well as word-subject relation are available. The further research, will discuss the possibilities of interaction with the user, from the point of implementing the weighted retrieval and enabling additional intervention upon the set of retrieved documents, using user's judgement (the relevance feedback) to improve the quality of retrieval.

Item Type: Book Section
Uncontrolled Keywords: relevancy, matching, retrieval, ranking, evaluation, matching function, information retrieval
Subjects: International Conference on Informatics and Information Technologies > Artificial Intelligence
International Conference on Informatics and Information Technologies > Intelligent Systems
Depositing User: Vangel Ajanovski
Date Deposited: 28 Oct 2016 00:15
Last Modified: 28 Oct 2016 00:15
URI: http://eprints.finki.ukim.mk/id/eprint/11315

Actions (login required)

View Item View Item