Association rules network: Definition and applications

Gaurav Pandey, Sanjay Chawla, Simon Poon, Bavani Arunasalam, Joseph G. Davis

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

The role of data mining is to search "the space of candidate hypotheses" to offer solutions, whereas the role of statistics is to validate the hypotheses offered by the data-mining process. In this paper we propose Association Rules Networks (ARNs) as a structure for synthesizing, pruning, and analyzing a collection of association rules to construct candidate hypotheses. From a knowledge discovery perspective, ARNs allow for a goal-centric, context-driven analysis of the output of association rules algorithms. From a mathematical perspective, ARNs are instances of backward-directed hypergraphs. Using two extensive case studies, we show how ARNs and statistical theory can be combined to generate and test hypotheses.

Original languageEnglish
Pages (from-to)260-279
Number of pages20
JournalStatistical Analysis and Data Mining
Volume1
Issue number4
DOIs
Publication statusPublished - Mar 2009
Externally publishedYes

Fingerprint

Association rules
Association Rules
Data mining
Data Mining
Directed Hypergraphs
Hypothesis Test
Knowledge Discovery
Pruning
Statistics
Output

Keywords

  • Association rules
  • Association rules network
  • Data mining

ASJC Scopus subject areas

  • Information Systems
  • Computer Science Applications
  • Analysis

Cite this

Association rules network : Definition and applications. / Pandey, Gaurav; Chawla, Sanjay; Poon, Simon; Arunasalam, Bavani; Davis, Joseph G.

In: Statistical Analysis and Data Mining, Vol. 1, No. 4, 03.2009, p. 260-279.

Research output: Contribution to journalArticle

Pandey, Gaurav ; Chawla, Sanjay ; Poon, Simon ; Arunasalam, Bavani ; Davis, Joseph G. / Association rules network : Definition and applications. In: Statistical Analysis and Data Mining. 2009 ; Vol. 1, No. 4. pp. 260-279.
@article{67c62bc3083944d98a517b6668b68797,
title = "Association rules network: Definition and applications",
abstract = "The role of data mining is to search {"}the space of candidate hypotheses{"} to offer solutions, whereas the role of statistics is to validate the hypotheses offered by the data-mining process. In this paper we propose Association Rules Networks (ARNs) as a structure for synthesizing, pruning, and analyzing a collection of association rules to construct candidate hypotheses. From a knowledge discovery perspective, ARNs allow for a goal-centric, context-driven analysis of the output of association rules algorithms. From a mathematical perspective, ARNs are instances of backward-directed hypergraphs. Using two extensive case studies, we show how ARNs and statistical theory can be combined to generate and test hypotheses.",
keywords = "Association rules, Association rules network, Data mining",
author = "Gaurav Pandey and Sanjay Chawla and Simon Poon and Bavani Arunasalam and Davis, {Joseph G.}",
year = "2009",
month = "3",
doi = "10.1002/sam.10027",
language = "English",
volume = "1",
pages = "260--279",
journal = "Statistical Analysis and Data Mining",
issn = "1932-1872",
publisher = "John Wiley and Sons Inc.",
number = "4",

}

TY - JOUR

T1 - Association rules network

T2 - Definition and applications

AU - Pandey, Gaurav

AU - Chawla, Sanjay

AU - Poon, Simon

AU - Arunasalam, Bavani

AU - Davis, Joseph G.

PY - 2009/3

Y1 - 2009/3

N2 - The role of data mining is to search "the space of candidate hypotheses" to offer solutions, whereas the role of statistics is to validate the hypotheses offered by the data-mining process. In this paper we propose Association Rules Networks (ARNs) as a structure for synthesizing, pruning, and analyzing a collection of association rules to construct candidate hypotheses. From a knowledge discovery perspective, ARNs allow for a goal-centric, context-driven analysis of the output of association rules algorithms. From a mathematical perspective, ARNs are instances of backward-directed hypergraphs. Using two extensive case studies, we show how ARNs and statistical theory can be combined to generate and test hypotheses.

AB - The role of data mining is to search "the space of candidate hypotheses" to offer solutions, whereas the role of statistics is to validate the hypotheses offered by the data-mining process. In this paper we propose Association Rules Networks (ARNs) as a structure for synthesizing, pruning, and analyzing a collection of association rules to construct candidate hypotheses. From a knowledge discovery perspective, ARNs allow for a goal-centric, context-driven analysis of the output of association rules algorithms. From a mathematical perspective, ARNs are instances of backward-directed hypergraphs. Using two extensive case studies, we show how ARNs and statistical theory can be combined to generate and test hypotheses.

KW - Association rules

KW - Association rules network

KW - Data mining

UR - http://www.scopus.com/inward/record.url?scp=62249090283&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=62249090283&partnerID=8YFLogxK

U2 - 10.1002/sam.10027

DO - 10.1002/sam.10027

M3 - Article

AN - SCOPUS:62249090283

VL - 1

SP - 260

EP - 279

JO - Statistical Analysis and Data Mining

JF - Statistical Analysis and Data Mining

SN - 1932-1872

IS - 4

ER -