Automated boolean matrix data representation scheme through AVL tree for mining association rules

GhassanSaleh Al-Dharhani, Zulaiha Ali Othman, Azuraliza Abu Bakar

Research output: Contribution to journalArticle

Abstract

A main issue in associate rule mining is to extract quality rules with faster times. The most popular method is improving the data representation method. This study aims to propose an automated Boolean Matrix data transactional representation scheme that improves the existing Apriori algorithm for mining association rules. It contains two stages. Firstly, a data representation scheme through AVL-Tree which is a container of balanced actual data to represent the data from transaction database to Boolean Matrix automatically, and secondly, N-frequent itemsets generated from the Apriori algorithm is made based on the proposed representation. The improvement deals with data representation since n-frequent itemsets with association rules are generated. The result show that the propose data representation has improved the computation time using AVL-Tree compare with Apriori algorithm.

Original languageEnglish
Pages (from-to)420-429
Number of pages10
JournalAmerican-Eurasian Journal of Sustainable Agriculture
Volume7
Issue number4
Publication statusPublished - Jul 2013

Fingerprint

containers
Databases
extracts
methodology

Keywords

  • Apriori
  • Association rules mining
  • AVL-tree
  • Boolean matrix
  • Data representation

ASJC Scopus subject areas

  • Agronomy and Crop Science
  • Food Science
  • Horticulture

Cite this

@article{e8a19e4f500e4c5ea87d747d2e9a19b0,
title = "Automated boolean matrix data representation scheme through AVL tree for mining association rules",
abstract = "A main issue in associate rule mining is to extract quality rules with faster times. The most popular method is improving the data representation method. This study aims to propose an automated Boolean Matrix data transactional representation scheme that improves the existing Apriori algorithm for mining association rules. It contains two stages. Firstly, a data representation scheme through AVL-Tree which is a container of balanced actual data to represent the data from transaction database to Boolean Matrix automatically, and secondly, N-frequent itemsets generated from the Apriori algorithm is made based on the proposed representation. The improvement deals with data representation since n-frequent itemsets with association rules are generated. The result show that the propose data representation has improved the computation time using AVL-Tree compare with Apriori algorithm.",
keywords = "Apriori, Association rules mining, AVL-tree, Boolean matrix, Data representation",
author = "GhassanSaleh Al-Dharhani and {Ali Othman}, Zulaiha and {Abu Bakar}, Azuraliza",
year = "2013",
month = "7",
language = "English",
volume = "7",
pages = "420--429",
journal = "American-Eurasian Journal of Sustainable Agriculture",
issn = "1995-0748",
publisher = "American-Eurasian Network for Scientific Information",
number = "4",

}

TY - JOUR

T1 - Automated boolean matrix data representation scheme through AVL tree for mining association rules

AU - Al-Dharhani, GhassanSaleh

AU - Ali Othman, Zulaiha

AU - Abu Bakar, Azuraliza

PY - 2013/7

Y1 - 2013/7

N2 - A main issue in associate rule mining is to extract quality rules with faster times. The most popular method is improving the data representation method. This study aims to propose an automated Boolean Matrix data transactional representation scheme that improves the existing Apriori algorithm for mining association rules. It contains two stages. Firstly, a data representation scheme through AVL-Tree which is a container of balanced actual data to represent the data from transaction database to Boolean Matrix automatically, and secondly, N-frequent itemsets generated from the Apriori algorithm is made based on the proposed representation. The improvement deals with data representation since n-frequent itemsets with association rules are generated. The result show that the propose data representation has improved the computation time using AVL-Tree compare with Apriori algorithm.

AB - A main issue in associate rule mining is to extract quality rules with faster times. The most popular method is improving the data representation method. This study aims to propose an automated Boolean Matrix data transactional representation scheme that improves the existing Apriori algorithm for mining association rules. It contains two stages. Firstly, a data representation scheme through AVL-Tree which is a container of balanced actual data to represent the data from transaction database to Boolean Matrix automatically, and secondly, N-frequent itemsets generated from the Apriori algorithm is made based on the proposed representation. The improvement deals with data representation since n-frequent itemsets with association rules are generated. The result show that the propose data representation has improved the computation time using AVL-Tree compare with Apriori algorithm.

KW - Apriori

KW - Association rules mining

KW - AVL-tree

KW - Boolean matrix

KW - Data representation

UR - http://www.scopus.com/inward/record.url?scp=84885203744&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885203744&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84885203744

VL - 7

SP - 420

EP - 429

JO - American-Eurasian Journal of Sustainable Agriculture

JF - American-Eurasian Journal of Sustainable Agriculture

SN - 1995-0748

IS - 4

ER -