Presentation 2003/3/8
On Characteristics of Dissimilarity Measures for Multiscale Matching
Shoji HIRANO, Shusaku TSUMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents some properties of the dissimilarity measures used in the multiscale matching. Multiscale matching is a method to compare planar curves by partly changing observation scales. For the multiscale comparison of one-dimensional temporal sequences, we proposed the dissimilarity measure of subsequences consists of four components, that are, rotation angle, length, phase and gradient. On the synthetic data, we empirically examined contribution of the four components to the resultant dissimilarity between subsequences.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) multiscale matching / temporal data mining / proximity measure
Paper # AI2002-90
Date of Issue

Conference Information
Committee AI
Conference Date 2003/3/8(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 Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) On Characteristics of Dissimilarity Measures for Multiscale Matching
Sub Title (in English)
Keyword(1) multiscale matching
Keyword(2) temporal data mining
Keyword(3) proximity measure
1st Author's Name Shoji HIRANO
1st Author's Affiliation Department of Medical Informatics, Shimane Medical University, School of Medicine()
2nd Author's Name Shusaku TSUMOTO
2nd Author's Affiliation Department of Medical Informatics, Shimane Medical University, School of Medicine
Date 2003/3/8
Paper # AI2002-90
Volume (vol) vol.102
Number (no) 711
Page pp.pp.-
#Pages 4
Date of Issue