Paper Info
Title
An impression-based retrieval system of music collection
Abstract
Recently we can easily create multimedia contents without creating each medium data composing it. In this case, we should often need to select a medium data which is close to the imaged impression from the data which we do not know at all. We propose a retrieval method for music works using impression words. The characteristic of the proposed method is using the affective value of music work that is based on the results of psychological research on music. In order to develop a creation method of the affective value, we examine relation between the affective value and the music structure by applying the machine learning technology. And we show results of the examination and discuss a creation method of the affective value.
Year
DOI
Venue
2000
10.1109/KES.2000.884181
KES'2000: FOURTH INTERNATIONAL CONFERENCE ON KNOWLEDGE-BASED INTELLIGENT ENGINEERING SYSTEMS & ALLIED TECHNOLOGIES, VOLS 1 AND 2, PROCEEDINGS
Keywords
Field
DocType
computer networks,musical structure,testing,psychology,multiple signal classification,psychological research,learning artificial intelligence,databases,machine learning,music,information retrieval
Music information retrieval,Multiple signal classification,Information retrieval,Musical,Impression,Computer science,Musicality,Psychological research,Affect (psychology),Musical form
Conference
Citations 
PageRank 
References 
8
1.12
4
Authors
3
Name
Order
Citations
PageRank
A. Sato1182.95
J. Ogawa281.12
H. Kitakami39449.68