Robust Pseudo-Hierarchical Support Vector Clustering

Michael Sass Hansen; Karl Sjöstrand; Hildur Olafsdóttir; Henrik B. W. Larsson; Mikkel Bille Stegmann; Rasmus Larsen; Bjarne Kjær Ersbøll; Janne Heikkilä; Ivar Austvoll; Ingela Nyström

Robust Pseudo-Hierarchical Support Vector Clustering

Michael Sass Hansen, Karl Sjöstrand, Hildur Olafsdóttir, Henrik B. W. Larsson, Mikkel Bille Stegmann, Rasmus Larsen, Bjarne Kjær Ersbøll (Editor), Janne Heikkilä (Editor), Ivar Austvoll (Editor), Ingela Nyström (Editor)

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

Support vector clustering (SVC) has proven an efficient algorithm for clustering of noisy and high-dimensional data sets, with applications within many fields of research. An inherent problem, however, has been setting the parameters of the SVC algorithm. Using the recent emergence of a method for calculating the entire regularization path of the support vector domain description, we propose a fast method for robust pseudo-hierarchical support vector clustering (HSVC). The method is demonstrated to work well on generated data, as well as for detecting ischemic segments from multidimensional myocardial perfusion magnetic resonance imaging data, giving robust results while drastically reducing the need for parameter estimation.

Original language	English
Title of host publication	Scandinavian Conference on Image Analysis 2007
Publisher	Springer
Publication date	2007
Publication status	Published - 2007
Event	15th Scandinavian Conference on Image Analysis - Aalborg, Denmark Duration: 10 Jun 2007 → 24 Jun 2007 Conference number: 15

Conference

Conference	15th Scandinavian Conference on Image Analysis
Number	15
Country/Territory	Denmark
City	Aalborg
Period	10/06/2007 → 24/06/2007

Keywords

Hierarchical support vector clustering

OpenUrl availability

Full text

Cite this

@inproceedings{46586389ee7c4c1ab8f21e52a26c3692,

title = "Robust Pseudo-Hierarchical Support Vector Clustering",

abstract = "Support vector clustering (SVC) has proven an efficient algorithm for clustering of noisy and high-dimensional data sets, with applications within many fields of research. An inherent problem, however, has been setting the parameters of the SVC algorithm. Using the recent emergence of a method for calculating the entire regularization path of the support vector domain description, we propose a fast method for robust pseudo-hierarchical support vector clustering (HSVC). The method is demonstrated to work well on generated data, as well as for detecting ischemic segments from multidimensional myocardial perfusion magnetic resonance imaging data, giving robust results while drastically reducing the need for parameter estimation.",

keywords = "Hierarchical support vector clustering",

author = "Hansen, {Michael Sass} and Karl Sj{\"o}strand and Hildur Olafsd{\'o}ttir and Larsson, {Henrik B. W.} and Stegmann, {Mikkel Bille} and Rasmus Larsen and Ersb{\o}ll, {Bjarne Kj{\ae}r} and Janne Heikkil{\"a} and Ivar Austvoll and Ingela Nystr{\"o}m",

year = "2007",

language = "English",

booktitle = "Scandinavian Conference on Image Analysis 2007",

publisher = "Springer",

note = "15th Scandinavian Conference on Image Analysis, SCIA ; Conference date: 10-06-2007 Through 24-06-2007",

}

TY - GEN

T1 - Robust Pseudo-Hierarchical Support Vector Clustering

AU - Hansen, Michael Sass

AU - Sjöstrand, Karl

AU - Olafsdóttir, Hildur

AU - Larsson, Henrik B. W.

AU - Stegmann, Mikkel Bille

AU - Larsen, Rasmus

A2 - Ersbøll, Bjarne Kjær

A2 - Heikkilä, Janne

A2 - Austvoll, Ivar

A2 - Nyström, Ingela

N1 - Conference code: 15

PY - 2007

Y1 - 2007

N2 - Support vector clustering (SVC) has proven an efficient algorithm for clustering of noisy and high-dimensional data sets, with applications within many fields of research. An inherent problem, however, has been setting the parameters of the SVC algorithm. Using the recent emergence of a method for calculating the entire regularization path of the support vector domain description, we propose a fast method for robust pseudo-hierarchical support vector clustering (HSVC). The method is demonstrated to work well on generated data, as well as for detecting ischemic segments from multidimensional myocardial perfusion magnetic resonance imaging data, giving robust results while drastically reducing the need for parameter estimation.

AB - Support vector clustering (SVC) has proven an efficient algorithm for clustering of noisy and high-dimensional data sets, with applications within many fields of research. An inherent problem, however, has been setting the parameters of the SVC algorithm. Using the recent emergence of a method for calculating the entire regularization path of the support vector domain description, we propose a fast method for robust pseudo-hierarchical support vector clustering (HSVC). The method is demonstrated to work well on generated data, as well as for detecting ischemic segments from multidimensional myocardial perfusion magnetic resonance imaging data, giving robust results while drastically reducing the need for parameter estimation.

KW - Hierarchical support vector clustering

M3 - Article in proceedings

BT - Scandinavian Conference on Image Analysis 2007

PB - Springer

T2 - 15th Scandinavian Conference on Image Analysis

Y2 - 10 June 2007 through 24 June 2007

ER -

Robust Pseudo-Hierarchical Support Vector Clustering

Abstract

Conference

Keywords

OpenUrl availability

Fingerprint

Cite this