Opens in a new tab

AES E-Library

← Back to search

Convention Paper

Music Archive Metadata Processing Based on Flow Graphs

Authors: Kostek, Bozena; Czyzewski, Andrzej

AES Convention 116 · Paper 6138 · May 2004

Abstract

The paper addresses the capabilities that should be expected from intelligent Web search tools in order to respond properly to user's music information retrieval needs. An advanced query algorithm was engineered employing a concept of inference rule derivation from flow graphs with regard to semantic data processing. This concept, introduced recently by Pawlak, is used for mining knowledge in databases. The created database searching engine utilizes knowledge acquired in advance and stored in flow graphs in order to enable searching in musical repositories. Results obtained show that employing the implemented method the resulting search matches are ranked optimally, thus metadata related to recorded sound can be retrieved efficiently with the use of this algorithm.

Details

Published in
AES Convention 116
AES Convention
116
Paper number
6138
Publication date
May 6, 2004
Session subject
Audio Recording and Reproduction; Archiving and Content Management
Affiliation
Gdansk University of Technology, Multimedia Systems Department, Gdansk, PL (See document for exact affiliation information.)
Type
Convention Paper