estWin: Online data stream mining of recent frequent itemsets by sliding window method

Joong Hyuk Chang, Won Suk Lee

Research output: Contribution to journalArticle

57 Citations (Scopus)

Abstract

Knowledge embedded in a data stream is likely to be changed as time goes by. Identifying the recent change of the knowledge quickly can provide valuable information for the analysis of the data stream. However, most mining algorithms over a data stream are not able to extract the recent change of knowledge in a data stream adaptively. This is because the obsolete information of old data elements which may be no longer useful or possibly invalid at present is regarded as being as important as that of recent data elements. This paper proposes a sliding window method that finds recently frequent itemsets over a transactional online data stream adaptively. The size of a sliding window defines the desired life-time of information in a newly generated transaction. Consequently, only recently generated transactions in the range of the window are considered to find the recently frequent itemsets of a data stream.

Original languageEnglish
Pages (from-to)76-90
Number of pages15
JournalJournal of Information Science
Volume31
Issue number2
DOIs
Publication statusPublished - 2005 Dec 1

Fingerprint

transaction
present
time

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Library and Information Sciences

Cite this

@article{9f1053ac74e045b59d9e00bace661caa,
title = "estWin: Online data stream mining of recent frequent itemsets by sliding window method",
abstract = "Knowledge embedded in a data stream is likely to be changed as time goes by. Identifying the recent change of the knowledge quickly can provide valuable information for the analysis of the data stream. However, most mining algorithms over a data stream are not able to extract the recent change of knowledge in a data stream adaptively. This is because the obsolete information of old data elements which may be no longer useful or possibly invalid at present is regarded as being as important as that of recent data elements. This paper proposes a sliding window method that finds recently frequent itemsets over a transactional online data stream adaptively. The size of a sliding window defines the desired life-time of information in a newly generated transaction. Consequently, only recently generated transactions in the range of the window are considered to find the recently frequent itemsets of a data stream.",
author = "Chang, {Joong Hyuk} and Lee, {Won Suk}",
year = "2005",
month = "12",
day = "1",
doi = "10.1177/0165551505050785",
language = "English",
volume = "31",
pages = "76--90",
journal = "Journal of Information Science",
issn = "0165-5515",
publisher = "SAGE Publications Ltd",
number = "2",

}

estWin : Online data stream mining of recent frequent itemsets by sliding window method. / Chang, Joong Hyuk; Lee, Won Suk.

In: Journal of Information Science, Vol. 31, No. 2, 01.12.2005, p. 76-90.

Research output: Contribution to journalArticle

TY - JOUR

T1 - estWin

T2 - Online data stream mining of recent frequent itemsets by sliding window method

AU - Chang, Joong Hyuk

AU - Lee, Won Suk

PY - 2005/12/1

Y1 - 2005/12/1

N2 - Knowledge embedded in a data stream is likely to be changed as time goes by. Identifying the recent change of the knowledge quickly can provide valuable information for the analysis of the data stream. However, most mining algorithms over a data stream are not able to extract the recent change of knowledge in a data stream adaptively. This is because the obsolete information of old data elements which may be no longer useful or possibly invalid at present is regarded as being as important as that of recent data elements. This paper proposes a sliding window method that finds recently frequent itemsets over a transactional online data stream adaptively. The size of a sliding window defines the desired life-time of information in a newly generated transaction. Consequently, only recently generated transactions in the range of the window are considered to find the recently frequent itemsets of a data stream.

AB - Knowledge embedded in a data stream is likely to be changed as time goes by. Identifying the recent change of the knowledge quickly can provide valuable information for the analysis of the data stream. However, most mining algorithms over a data stream are not able to extract the recent change of knowledge in a data stream adaptively. This is because the obsolete information of old data elements which may be no longer useful or possibly invalid at present is regarded as being as important as that of recent data elements. This paper proposes a sliding window method that finds recently frequent itemsets over a transactional online data stream adaptively. The size of a sliding window defines the desired life-time of information in a newly generated transaction. Consequently, only recently generated transactions in the range of the window are considered to find the recently frequent itemsets of a data stream.

UR - http://www.scopus.com/inward/record.url?scp=30344444271&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=30344444271&partnerID=8YFLogxK

U2 - 10.1177/0165551505050785

DO - 10.1177/0165551505050785

M3 - Article

AN - SCOPUS:30344444271

VL - 31

SP - 76

EP - 90

JO - Journal of Information Science

JF - Journal of Information Science

SN - 0165-5515

IS - 2

ER -