Efficient mining of frequent patterns based on a condensed tree structure

Byung Joon Park, Sang Young Kim, Lynn Choi

Research output: Contribution to journalArticle

Abstract

Frequent pattern mining has attracted a great deal of interests during the recent surge in Web mining research because it is the basis of many applications such as customer behavior analysis and trend prediction. Researchers have proposed various data structures and algorithms to discover frequently occurring patterns from a given data set. In particular, tree structures have been popular since they can effectively represent the input data set for efficient pattern discovery. In this paper, we propose an efficient tree structure and its associated algorithm that provides a considerable performance improvement over CATS, one of the fastest frequent pattern mining algorithms known to date, in terms of memory usage and processing time. We demonstrate the effectiveness of our algorithm and performance improvement over me existing approach by a series of experiments.

Original languageEnglish
Pages (from-to)1665-1670
Number of pages6
JournalInformation (Japan)
Volume17
Issue number5
Publication statusPublished - 2014 Jan 1

Fingerprint

Data structures
Data storage equipment
Processing
Experiments

ASJC Scopus subject areas

  • General

Cite this

Efficient mining of frequent patterns based on a condensed tree structure. / Park, Byung Joon; Kim, Sang Young; Choi, Lynn.

In: Information (Japan), Vol. 17, No. 5, 01.01.2014, p. 1665-1670.

Research output: Contribution to journalArticle

Park, Byung Joon ; Kim, Sang Young ; Choi, Lynn. / Efficient mining of frequent patterns based on a condensed tree structure. In: Information (Japan). 2014 ; Vol. 17, No. 5. pp. 1665-1670.
@article{f299b447b4f44e299739e10fda61b143,
title = "Efficient mining of frequent patterns based on a condensed tree structure",
abstract = "Frequent pattern mining has attracted a great deal of interests during the recent surge in Web mining research because it is the basis of many applications such as customer behavior analysis and trend prediction. Researchers have proposed various data structures and algorithms to discover frequently occurring patterns from a given data set. In particular, tree structures have been popular since they can effectively represent the input data set for efficient pattern discovery. In this paper, we propose an efficient tree structure and its associated algorithm that provides a considerable performance improvement over CATS, one of the fastest frequent pattern mining algorithms known to date, in terms of memory usage and processing time. We demonstrate the effectiveness of our algorithm and performance improvement over me existing approach by a series of experiments.",
keywords = "Frequent pattern, Pattern discovery, Tree structure",
author = "Park, {Byung Joon} and Kim, {Sang Young} and Lynn Choi",
year = "2014",
month = "1",
day = "1",
language = "English",
volume = "17",
pages = "1665--1670",
journal = "Information (Japan)",
issn = "1343-4500",
publisher = "International Information Institute",
number = "5",

}

TY - JOUR

T1 - Efficient mining of frequent patterns based on a condensed tree structure

AU - Park, Byung Joon

AU - Kim, Sang Young

AU - Choi, Lynn

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Frequent pattern mining has attracted a great deal of interests during the recent surge in Web mining research because it is the basis of many applications such as customer behavior analysis and trend prediction. Researchers have proposed various data structures and algorithms to discover frequently occurring patterns from a given data set. In particular, tree structures have been popular since they can effectively represent the input data set for efficient pattern discovery. In this paper, we propose an efficient tree structure and its associated algorithm that provides a considerable performance improvement over CATS, one of the fastest frequent pattern mining algorithms known to date, in terms of memory usage and processing time. We demonstrate the effectiveness of our algorithm and performance improvement over me existing approach by a series of experiments.

AB - Frequent pattern mining has attracted a great deal of interests during the recent surge in Web mining research because it is the basis of many applications such as customer behavior analysis and trend prediction. Researchers have proposed various data structures and algorithms to discover frequently occurring patterns from a given data set. In particular, tree structures have been popular since they can effectively represent the input data set for efficient pattern discovery. In this paper, we propose an efficient tree structure and its associated algorithm that provides a considerable performance improvement over CATS, one of the fastest frequent pattern mining algorithms known to date, in terms of memory usage and processing time. We demonstrate the effectiveness of our algorithm and performance improvement over me existing approach by a series of experiments.

KW - Frequent pattern

KW - Pattern discovery

KW - Tree structure

UR - http://www.scopus.com/inward/record.url?scp=84903957328&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84903957328&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84903957328

VL - 17

SP - 1665

EP - 1670

JO - Information (Japan)

JF - Information (Japan)

SN - 1343-4500

IS - 5

ER -