The relationship of DBSCAN to matrix factorization and spectral clustering

Erich Schubert; Sibylle Hess; Katharina Morik

The relationship of DBSCAN to matrix factorization and spectral clustering

Erich Schubert
, Sibylle Hess
, Katharina Morik

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

15 Citaten (Scopus)

349 Downloads (Pure)

Samenvatting

DBSCAN is a popular approach for density-based clustering.
In this short ``work in progress'' paper, we want to present an interpretation of
DBSCAN as a matrix factorization problem, which introduces
a theoretical connection (but not an equivalence)
between DBSCAN and Spectral Clustering (SC).

While this does not yield a faster algorithm for DBSCAN,
establishing this relationship is a step towards a more unified
view of clustering, by identifying further relationships between
some of the most popular clustering algorithms.

Originele taal-2	Engels
Titel	LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018
Subtitel	Proceedings of the conference "Lernen, Wissen, Daten Analysen
Plaats van productie	Aachen
Uitgeverij	RWTH Aachen
Pagina's	330-334
Aantal pagina's	5
Status	Gepubliceerd - 2018
Extern gepubliceerd	Ja
Evenement	LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018 - Mannheim, Duitsland Duur: 22 aug 2018 → 24 aug 2018 https://bibliographie.ub.rub.de/retrieve/Conference/7fee8151-004e-44e5-90fd-7e9023e167e7/

Publicatie series

Naam	CEUR workshop proceedings
Volume	2191

Congres

Congres	LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018
Verkorte titel	LWDA2019
Land/Regio	Duitsland
Stad	Mannheim
Periode	22/08/18 → 24/08/18
Internet adres	https://bibliographie.ub.rub.de/retrieve/Conference/7fee8151-004e-44e5-90fd-7e9023e167e7/

Toegang tot document

Author versionFinale auteursversie , 353 KB

Andere bestanden en links

Link to publication in Scopus

Citeer dit

@inproceedings{8fe3d9b35f73410c8f25ea1612e4187c,

title = "The relationship of DBSCAN to matrix factorization and spectral clustering",

abstract = "DBSCAN is a popular approach for density-based clustering. In this short ``work in progress'' paper, we want to present an interpretation of DBSCAN as a matrix factorization problem, which introduces a theoretical connection (but not an equivalence) between DBSCAN and Spectral Clustering (SC). While this does not yield a faster algorithm for DBSCAN, establishing this relationship is a step towards a more unified view of clustering, by identifying further relationships between some of the most popular clustering algorithms.",

author = "Erich Schubert and Sibylle Hess and Katharina Morik",

year = "2018",

language = "English",

series = "CEUR workshop proceedings",

publisher = "RWTH Aachen",

pages = "330--334",

booktitle = "LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018",

note = "LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018, LWDA2019 ; Conference date: 22-08-2018 Through 24-08-2018",

url = "https://bibliographie.ub.rub.de/retrieve/Conference/7fee8151-004e-44e5-90fd-7e9023e167e7/",

}

Schubert, E, Hess, S & Morik, K 2018, The relationship of DBSCAN to matrix factorization and spectral clustering. in LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018: Proceedings of the conference "Lernen, Wissen, Daten Analysen. CEUR workshop proceedings, vol. 2191, RWTH Aachen, Aachen, blz. 330-334, LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018, Mannheim, Duitsland, 22/08/18.

The relationship of DBSCAN to matrix factorization and spectral clustering. / Schubert, Erich; Hess, Sibylle; Morik, Katharina.
LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018: Proceedings of the conference "Lernen, Wissen, Daten Analysen. Aachen: RWTH Aachen, 2018. blz. 330-334 (CEUR workshop proceedings; Vol. 2191).

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

TY - GEN

T1 - The relationship of DBSCAN to matrix factorization and spectral clustering

AU - Schubert, Erich

AU - Hess, Sibylle

AU - Morik, Katharina

PY - 2018

Y1 - 2018

N2 - DBSCAN is a popular approach for density-based clustering. In this short ``work in progress'' paper, we want to present an interpretation of DBSCAN as a matrix factorization problem, which introduces a theoretical connection (but not an equivalence) between DBSCAN and Spectral Clustering (SC). While this does not yield a faster algorithm for DBSCAN, establishing this relationship is a step towards a more unified view of clustering, by identifying further relationships between some of the most popular clustering algorithms.

AB - DBSCAN is a popular approach for density-based clustering. In this short ``work in progress'' paper, we want to present an interpretation of DBSCAN as a matrix factorization problem, which introduces a theoretical connection (but not an equivalence) between DBSCAN and Spectral Clustering (SC). While this does not yield a faster algorithm for DBSCAN, establishing this relationship is a step towards a more unified view of clustering, by identifying further relationships between some of the most popular clustering algorithms.

UR - https://www.scopus.com/pages/publications/85053612124

M3 - Conference contribution

T3 - CEUR workshop proceedings

SP - 330

EP - 334

BT - LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018

PB - RWTH Aachen

CY - Aachen

T2 - LWDA 2018 - Lernen, Wissen, Daten, Analysen 2018

Y2 - 22 August 2018 through 24 August 2018

ER -

The relationship of DBSCAN to matrix factorization and spectral clustering

Samenvatting

Publicatie series

Congres

Toegang tot document

Andere bestanden en links

Vingerafdruk

Citeer dit