Novel techniques and an efficient algorithm for closed pattern mining

András Király, Asta Laiho, J. Abonyi, Attila Gyenesei

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

In this paper we show that frequent closed itemset mining and biclustering, the two most prominent application fields in pattern discovery, can be reduced to the same problem when dealing with binary (0-1) data. FCPMiner, a new powerful pattern mining method, is then introduced to mine such data efficiently. The uniqueness of the proposed method is its extendibility to non-binary data. The mining method is coupled with a novel visualization technique and a pattern aggregation method to detect the most meaningful, non-overlapping patterns. The proposed methods are rigorously tested on both synthetic and real data sets.

Original languageEnglish
Pages (from-to)5105-5114
Number of pages10
JournalExpert Systems with Applications
Volume41
Issue number11
DOIs
Publication statusPublished - Sep 1 2014

Fingerprint

Agglomeration
Visualization

Keywords

  • Biclustering
  • Closed frequent itemset mining
  • Clustering visualization
  • Data mining algorithm
  • Pattern detection

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Engineering(all)

Cite this

Novel techniques and an efficient algorithm for closed pattern mining. / Király, András; Laiho, Asta; Abonyi, J.; Gyenesei, Attila.

In: Expert Systems with Applications, Vol. 41, No. 11, 01.09.2014, p. 5105-5114.

Research output: Contribution to journalArticle

Király, András ; Laiho, Asta ; Abonyi, J. ; Gyenesei, Attila. / Novel techniques and an efficient algorithm for closed pattern mining. In: Expert Systems with Applications. 2014 ; Vol. 41, No. 11. pp. 5105-5114.
@article{eca432a0704743bf8fb32e85f70169f0,
title = "Novel techniques and an efficient algorithm for closed pattern mining",
abstract = "In this paper we show that frequent closed itemset mining and biclustering, the two most prominent application fields in pattern discovery, can be reduced to the same problem when dealing with binary (0-1) data. FCPMiner, a new powerful pattern mining method, is then introduced to mine such data efficiently. The uniqueness of the proposed method is its extendibility to non-binary data. The mining method is coupled with a novel visualization technique and a pattern aggregation method to detect the most meaningful, non-overlapping patterns. The proposed methods are rigorously tested on both synthetic and real data sets.",
keywords = "Biclustering, Closed frequent itemset mining, Clustering visualization, Data mining algorithm, Pattern detection",
author = "Andr{\'a}s Kir{\'a}ly and Asta Laiho and J. Abonyi and Attila Gyenesei",
year = "2014",
month = "9",
day = "1",
doi = "10.1016/j.eswa.2014.02.029",
language = "English",
volume = "41",
pages = "5105--5114",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",
number = "11",

}

TY - JOUR

T1 - Novel techniques and an efficient algorithm for closed pattern mining

AU - Király, András

AU - Laiho, Asta

AU - Abonyi, J.

AU - Gyenesei, Attila

PY - 2014/9/1

Y1 - 2014/9/1

N2 - In this paper we show that frequent closed itemset mining and biclustering, the two most prominent application fields in pattern discovery, can be reduced to the same problem when dealing with binary (0-1) data. FCPMiner, a new powerful pattern mining method, is then introduced to mine such data efficiently. The uniqueness of the proposed method is its extendibility to non-binary data. The mining method is coupled with a novel visualization technique and a pattern aggregation method to detect the most meaningful, non-overlapping patterns. The proposed methods are rigorously tested on both synthetic and real data sets.

AB - In this paper we show that frequent closed itemset mining and biclustering, the two most prominent application fields in pattern discovery, can be reduced to the same problem when dealing with binary (0-1) data. FCPMiner, a new powerful pattern mining method, is then introduced to mine such data efficiently. The uniqueness of the proposed method is its extendibility to non-binary data. The mining method is coupled with a novel visualization technique and a pattern aggregation method to detect the most meaningful, non-overlapping patterns. The proposed methods are rigorously tested on both synthetic and real data sets.

KW - Biclustering

KW - Closed frequent itemset mining

KW - Clustering visualization

KW - Data mining algorithm

KW - Pattern detection

UR - http://www.scopus.com/inward/record.url?scp=84897987528&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897987528&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2014.02.029

DO - 10.1016/j.eswa.2014.02.029

M3 - Article

AN - SCOPUS:84897987528

VL - 41

SP - 5105

EP - 5114

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 11

ER -