Interactive discovery of association rules over data streams

Se Jung Shin, Won Suk Lee

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

To trace the changes of association rules over an online data stream efficiently, this paper proposes two different methods of generating association rules directly over the changing set of currently frequent itemsets. These methods can avoid the drawbacks of the conventional two-step approach and provide an efficient way. The prefix tree itself can be utilized as an index structure for finding the current support of an association rule. While all of the currently frequent itemsets are monitored by the prefix tree, a traversal stack is employed to efficiently enumerate all association rules. In the on-line environment, a user may be interested in finding those association rules whose antecedents or consequents are fixed as a specific itemset. For this purpose, two additional methods, namely Assoc-X and Assoc-Y, are introduced. Finally, the proposed methods are compared by a series of experiments to identify its various characteristics.

Original languageEnglish
Pages (from-to)341-352
Number of pages12
JournalComputer Systems Science and Engineering
Volume29
Issue number5
Publication statusPublished - 2014 Sep 1

Fingerprint

Association rules
Association Rules
Data Streams
Frequent Itemsets
Prefix
Trace
Series
Experiment
Experiments

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Theoretical Computer Science
  • Computer Science(all)

Cite this

@article{85f457d08da44c7584f7f9d4577a0f7a,
title = "Interactive discovery of association rules over data streams",
abstract = "To trace the changes of association rules over an online data stream efficiently, this paper proposes two different methods of generating association rules directly over the changing set of currently frequent itemsets. These methods can avoid the drawbacks of the conventional two-step approach and provide an efficient way. The prefix tree itself can be utilized as an index structure for finding the current support of an association rule. While all of the currently frequent itemsets are monitored by the prefix tree, a traversal stack is employed to efficiently enumerate all association rules. In the on-line environment, a user may be interested in finding those association rules whose antecedents or consequents are fixed as a specific itemset. For this purpose, two additional methods, namely Assoc-X and Assoc-Y, are introduced. Finally, the proposed methods are compared by a series of experiments to identify its various characteristics.",
author = "Shin, {Se Jung} and Lee, {Won Suk}",
year = "2014",
month = "9",
day = "1",
language = "English",
volume = "29",
pages = "341--352",
journal = "Computer Systems Science and Engineering",
issn = "0267-6192",
publisher = "CRL Publishing",
number = "5",

}

Interactive discovery of association rules over data streams. / Shin, Se Jung; Lee, Won Suk.

In: Computer Systems Science and Engineering, Vol. 29, No. 5, 01.09.2014, p. 341-352.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Interactive discovery of association rules over data streams

AU - Shin, Se Jung

AU - Lee, Won Suk

PY - 2014/9/1

Y1 - 2014/9/1

N2 - To trace the changes of association rules over an online data stream efficiently, this paper proposes two different methods of generating association rules directly over the changing set of currently frequent itemsets. These methods can avoid the drawbacks of the conventional two-step approach and provide an efficient way. The prefix tree itself can be utilized as an index structure for finding the current support of an association rule. While all of the currently frequent itemsets are monitored by the prefix tree, a traversal stack is employed to efficiently enumerate all association rules. In the on-line environment, a user may be interested in finding those association rules whose antecedents or consequents are fixed as a specific itemset. For this purpose, two additional methods, namely Assoc-X and Assoc-Y, are introduced. Finally, the proposed methods are compared by a series of experiments to identify its various characteristics.

AB - To trace the changes of association rules over an online data stream efficiently, this paper proposes two different methods of generating association rules directly over the changing set of currently frequent itemsets. These methods can avoid the drawbacks of the conventional two-step approach and provide an efficient way. The prefix tree itself can be utilized as an index structure for finding the current support of an association rule. While all of the currently frequent itemsets are monitored by the prefix tree, a traversal stack is employed to efficiently enumerate all association rules. In the on-line environment, a user may be interested in finding those association rules whose antecedents or consequents are fixed as a specific itemset. For this purpose, two additional methods, namely Assoc-X and Assoc-Y, are introduced. Finally, the proposed methods are compared by a series of experiments to identify its various characteristics.

UR - http://www.scopus.com/inward/record.url?scp=84913555930&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84913555930&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84913555930

VL - 29

SP - 341

EP - 352

JO - Computer Systems Science and Engineering

JF - Computer Systems Science and Engineering

SN - 0267-6192

IS - 5

ER -