K-AP: Generating specified K clusters by efficient Affinity Propagation

Handle URI:
http://hdl.handle.net/10754/564322
Title:
K-AP: Generating specified K clusters by efficient Affinity Propagation
Authors:
Zhang, Xiangliang ( 0000-0002-3574-5665 ) ; Wang, Wei; Nørvåg, Kjetil; Sebag, Michèle
Abstract:
The Affinity Propagation (AP) clustering algorithm proposed by Frey and Dueck (2007) provides an understandable, nearly optimal summary of a data set. However, it suffers two major shortcomings: i) the number of clusters is vague with the user-defined parameter called self-confidence, and ii) the quadratic computational complexity. When aiming at a given number of clusters due to prior knowledge, AP has to be launched many times until an appropriate setting of self-confidence is found. The re-launched AP increases the computational cost by one order of magnitude. In this paper, we propose an algorithm, called K-AP, to exploit the immediate results of K clusters by introducing a constraint in the process of message passing. Through theoretical analysis and experimental validation, K-AP was shown to be able to directly generate K clusters as user defined, with a negligible increase of computational cost compared to AP. In the meanwhile, K-AP preserves the clustering quality as AP in terms of the distortion. K-AP is more effective than k-medoids w.r.t. the distortion minimization and higher clustering purity. © 2010 IEEE.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Computer Science Program; Machine Intelligence & kNowledge Engineering Lab
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
2010 IEEE International Conference on Data Mining
Conference/Event name:
10th IEEE International Conference on Data Mining, ICDM 2010
Issue Date:
Dec-2010
DOI:
10.1109/ICDM.2010.107
Type:
Conference Paper
ISSN:
15504786
ISBN:
9780769542560
Appears in Collections:
Conference Papers; Computer Science Program; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorZhang, Xiangliangen
dc.contributor.authorWang, Weien
dc.contributor.authorNørvåg, Kjetilen
dc.contributor.authorSebag, Michèleen
dc.date.accessioned2015-08-04T06:23:38Zen
dc.date.available2015-08-04T06:23:38Zen
dc.date.issued2010-12en
dc.identifier.isbn9780769542560en
dc.identifier.issn15504786en
dc.identifier.doi10.1109/ICDM.2010.107en
dc.identifier.urihttp://hdl.handle.net/10754/564322en
dc.description.abstractThe Affinity Propagation (AP) clustering algorithm proposed by Frey and Dueck (2007) provides an understandable, nearly optimal summary of a data set. However, it suffers two major shortcomings: i) the number of clusters is vague with the user-defined parameter called self-confidence, and ii) the quadratic computational complexity. When aiming at a given number of clusters due to prior knowledge, AP has to be launched many times until an appropriate setting of self-confidence is found. The re-launched AP increases the computational cost by one order of magnitude. In this paper, we propose an algorithm, called K-AP, to exploit the immediate results of K clusters by introducing a constraint in the process of message passing. Through theoretical analysis and experimental validation, K-AP was shown to be able to directly generate K clusters as user defined, with a negligible increase of computational cost compared to AP. In the meanwhile, K-AP preserves the clustering quality as AP in terms of the distortion. K-AP is more effective than k-medoids w.r.t. the distortion minimization and higher clustering purity. © 2010 IEEE.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.subjectAffinity Propagationen
dc.subjectClusteringen
dc.subjectK-medoidsen
dc.titleK-AP: Generating specified K clusters by efficient Affinity Propagationen
dc.typeConference Paperen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentComputer Science Programen
dc.contributor.departmentMachine Intelligence & kNowledge Engineering Laben
dc.identifier.journal2010 IEEE International Conference on Data Miningen
dc.conference.date14 December 2010 through 17 December 2010en
dc.conference.name10th IEEE International Conference on Data Mining, ICDM 2010en
dc.conference.locationSydney, NSWen
dc.contributor.institutionInterdisciplinary Centre for Security, Reliability and Trust (SnT Centre), University of Luxembourg, Luxembourg, Luxembourgen
dc.contributor.institutionDepartment of Computer and Information Science, Norwegian University of Science and Technology (NTNU), Norwayen
dc.contributor.institutionTAO - LRI, CNRS, INRIA, Université Paris-Sud 11, Franceen
kaust.authorZhang, Xiangliangen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.