Clicks

An effective algorithm for mining subspace clusters in categorical datasets

Mohammed J. Zaki, Markus Peters, Ira Assent, Thomas Seidl

Research output: Contribution to journalArticle

50 Citations (Scopus)

Abstract

We present a novel algorithm called Clicks, that finds clusters in categorical datasets based on a search for k-partite maximal cliques. Unlike previous methods, Clicks mines subspace clusters. It uses a selective vertical method to guarantee complete search. Clicks outperforms previous approaches by over an order of magnitude and scales better than any of the existing method for high-dimensional datasets. These results are demonstrated in a comprehensive performance study on real and synthetic datasets.

Original languageEnglish
Pages (from-to)51-70
Number of pages20
JournalData and Knowledge Engineering
Volume60
Issue number1
DOIs
Publication statusPublished - 1 Jan 2007
Externally publishedYes

Fingerprint

Guarantee
Clique

Keywords

  • Categorical data
  • Clustering
  • k-Partite graph
  • Maximal cliques

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Clicks : An effective algorithm for mining subspace clusters in categorical datasets. / Zaki, Mohammed J.; Peters, Markus; Assent, Ira; Seidl, Thomas.

In: Data and Knowledge Engineering, Vol. 60, No. 1, 01.01.2007, p. 51-70.

Research output: Contribution to journalArticle

Zaki, Mohammed J. ; Peters, Markus ; Assent, Ira ; Seidl, Thomas. / Clicks : An effective algorithm for mining subspace clusters in categorical datasets. In: Data and Knowledge Engineering. 2007 ; Vol. 60, No. 1. pp. 51-70.
@article{d2559dfc670542b7b5ab88cd453a9201,
title = "Clicks: An effective algorithm for mining subspace clusters in categorical datasets",
abstract = "We present a novel algorithm called Clicks, that finds clusters in categorical datasets based on a search for k-partite maximal cliques. Unlike previous methods, Clicks mines subspace clusters. It uses a selective vertical method to guarantee complete search. Clicks outperforms previous approaches by over an order of magnitude and scales better than any of the existing method for high-dimensional datasets. These results are demonstrated in a comprehensive performance study on real and synthetic datasets.",
keywords = "Categorical data, Clustering, k-Partite graph, Maximal cliques",
author = "Zaki, {Mohammed J.} and Markus Peters and Ira Assent and Thomas Seidl",
year = "2007",
month = "1",
day = "1",
doi = "10.1016/j.datak.2006.01.005",
language = "English",
volume = "60",
pages = "51--70",
journal = "Data and Knowledge Engineering",
issn = "0169-023X",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - Clicks

T2 - An effective algorithm for mining subspace clusters in categorical datasets

AU - Zaki, Mohammed J.

AU - Peters, Markus

AU - Assent, Ira

AU - Seidl, Thomas

PY - 2007/1/1

Y1 - 2007/1/1

N2 - We present a novel algorithm called Clicks, that finds clusters in categorical datasets based on a search for k-partite maximal cliques. Unlike previous methods, Clicks mines subspace clusters. It uses a selective vertical method to guarantee complete search. Clicks outperforms previous approaches by over an order of magnitude and scales better than any of the existing method for high-dimensional datasets. These results are demonstrated in a comprehensive performance study on real and synthetic datasets.

AB - We present a novel algorithm called Clicks, that finds clusters in categorical datasets based on a search for k-partite maximal cliques. Unlike previous methods, Clicks mines subspace clusters. It uses a selective vertical method to guarantee complete search. Clicks outperforms previous approaches by over an order of magnitude and scales better than any of the existing method for high-dimensional datasets. These results are demonstrated in a comprehensive performance study on real and synthetic datasets.

KW - Categorical data

KW - Clustering

KW - k-Partite graph

KW - Maximal cliques

UR - http://www.scopus.com/inward/record.url?scp=33845981111&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33845981111&partnerID=8YFLogxK

U2 - 10.1016/j.datak.2006.01.005

DO - 10.1016/j.datak.2006.01.005

M3 - Article

VL - 60

SP - 51

EP - 70

JO - Data and Knowledge Engineering

JF - Data and Knowledge Engineering

SN - 0169-023X

IS - 1

ER -