Sphere packing for clustering sets of vectors in feature space

Darío García-García; Raúl Santos-Rodríguez

doi:10.1109/ICASSP.2011.5946738

Sphere packing for clustering sets of vectors in feature space

Darío García-García^*, Raúl Santos-Rodríguez

^*Corresponding author for this work

Department of Engineering Mathematics

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

2 Citations (Scopus)

Abstract

We propose a method for clustering sets of vectors by packing spheres learnt to represent the support of the different sets. The algorithm can work efficiently in a kernel-induced feature space by using the kernel trick. Experimental results on synthetic and real-world datasets show that the proposal is competitive with the state of the art.

Original language	English
Title of host publication	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages	2092-2095
Number of pages	4
DOIs	https://doi.org/10.1109/ICASSP.2011.5946738
Publication status	Published - 18 Aug 2011
Event	2011 (36th) IEEE International Conference on Acoustics, Speech, and Signal Processing - Prague, Czech Republic Duration: 22 May 2011 → 27 May 2011

Conference

Conference	2011 (36th) IEEE International Conference on Acoustics, Speech, and Signal Processing
Abbreviated title	ICASSP '11
Country/Territory	Czech Republic
City	Prague
Period	22/05/11 → 27/05/11

Keywords

Clustering
kernel methods
sequences

Access to Document

10.1109/ICASSP.2011.5946738

Handle.net

Persistent link

Cite this

@inproceedings{bf23bf50dbca4e56b1a33575c7cd3bdd,

title = "Sphere packing for clustering sets of vectors in feature space",

abstract = "We propose a method for clustering sets of vectors by packing spheres learnt to represent the support of the different sets. The algorithm can work efficiently in a kernel-induced feature space by using the kernel trick. Experimental results on synthetic and real-world datasets show that the proposal is competitive with the state of the art.",

keywords = "Clustering, kernel methods, sequences",

author = "Dar{\'i}o Garc{\'i}a-Garc{\'i}a and Ra{\'u}l Santos-Rodr{\'i}guez",

year = "2011",

month = aug,

day = "18",

doi = "10.1109/ICASSP.2011.5946738",

language = "English",

isbn = "9781457705397",

pages = "2092--2095",

booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

note = "2011 (36th) IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP '11 ; Conference date: 22-05-2011 Through 27-05-2011",

}

García-García, D & Santos-Rodríguez, R 2011, Sphere packing for clustering sets of vectors in feature space. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 5946738, pp. 2092-2095, 2011 (36th) IEEE International Conference on Acoustics, Speech, and Signal Processing, Prague, Czech Republic, 22/05/11. https://doi.org/10.1109/ICASSP.2011.5946738

TY - GEN

T1 - Sphere packing for clustering sets of vectors in feature space

AU - García-García, Darío

AU - Santos-Rodríguez, Raúl

PY - 2011/8/18

Y1 - 2011/8/18

N2 - We propose a method for clustering sets of vectors by packing spheres learnt to represent the support of the different sets. The algorithm can work efficiently in a kernel-induced feature space by using the kernel trick. Experimental results on synthetic and real-world datasets show that the proposal is competitive with the state of the art.

AB - We propose a method for clustering sets of vectors by packing spheres learnt to represent the support of the different sets. The algorithm can work efficiently in a kernel-induced feature space by using the kernel trick. Experimental results on synthetic and real-world datasets show that the proposal is competitive with the state of the art.

KW - Clustering

KW - kernel methods

KW - sequences

UR - http://www.scopus.com/inward/record.url?scp=80051611813&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2011.5946738

DO - 10.1109/ICASSP.2011.5946738

M3 - Conference Contribution (Conference Proceeding)

AN - SCOPUS:80051611813

SN - 9781457705397

SP - 2092

EP - 2095

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

T2 - 2011 (36th) IEEE International Conference on Acoustics, Speech, and Signal Processing

Y2 - 22 May 2011 through 27 May 2011

ER -

Sphere packing for clustering sets of vectors in feature space

Abstract

Conference

Keywords

Access to Document

Handle.net

Other files and links

Fingerprint

Cite this