Presentation 2007-07-03
Proposal on Popular Music Clustering Method Focused the Chord Progression Pattern
Makiko NAGASAWA, Chiemi WATANABE, Takayuki ITO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) There are various music in the world, classical music, jazz, and popular music. Target of our study is mining of relativity among popular music and their attributes, such as composer, raking of hit charts, date of release, and among features of listener's favorite music, focusing on the popular music. For these achievements, we perform clustering focusing on the chord progression that based on music and invariance is high. In the previous work, we proposed similarity measure of chord progression of music data and we classified several Japanese pop music data. In this paper, we reports two experiments using the similarity measure. First, we visualize the relation clustering results to analyze the relation between the chord progression and their song's metadata such as title, performer and producer. Second, we improve the similarity measure of chord progression using typical chord progression pattern. In our proposed measure, each block of chord progression is represented as combination of typical chord patterns, a distance of two chord progressions is calculated based on vector space model.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Clustering / Mining / Music information processing / Visualization
Paper # DE2007-84
Date of Issue

Conference Information
Committee DE
Conference Date 2007/6/25(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Data Engineering (DE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Proposal on Popular Music Clustering Method Focused the Chord Progression Pattern
Sub Title (in English)
Keyword(1) Clustering
Keyword(2) Mining
Keyword(3) Music information processing
Keyword(4) Visualization
1st Author's Name Makiko NAGASAWA
1st Author's Affiliation Graduate Division of Mathematics and Computer Science , Ochanomizu University()
2nd Author's Name Chiemi WATANABE
2nd Author's Affiliation Department of Natural and Advanced Science, Ochanomizu University
3rd Author's Name Takayuki ITO
3rd Author's Affiliation Department of Natural and Advanced Science, Ochanomizu University
Date 2007-07-03
Paper # DE2007-84
Volume (vol) vol.107
Number (no) 131
Page pp.pp.-
#Pages 6
Date of Issue