Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Journal of Intelligent Information Systems, 20:3, 255283, 2003 c 2003 Kluwer Academic Publishers. Manufactured in The Netherlands.
 

Summary: Journal of Intelligent Information Systems, 20:3, 255283, 2003
c 2003 Kluwer Academic Publishers. Manufactured in The Netherlands.
A Statistical Theory for Quantitative
Association Rules
YONATAN AUMANN aumann@cs.biu.ac.il
Department of Computer Science, Bar-Ilan University, Ramat Gan, Israel 52900
YEHUDA LINDELL, lindell@us.ibm.com
IBM T.J. Watson Research, 19 Skyline Drive, Hawthorne, New York 10532, USA
Received December 31, 2001; Revised November 21, 2002; Accepted November 22, 2002
Abstract. Association rules are a key data-mining tool and as such have been well researched. So far, this
research has focused predominantly on databases containing categorical data only. However, many real-world
databases contain quantitative attributes and current solutions for this case are so far inadequate. In this paper we
introduce a new definition of quantitative association rules based on statistical inference theory. Our definition
reflects the intuition that the goal of association rules is to find extraordinary and therefore interesting phenomena in
databases. We also introduce the concept of sub-rules which can be applied to any type of association rule. Rigorous
experimental evaluation on real-world datasets is presented, demonstrating the usefulness and characteristics of
rules mined according to our definition.
Keywords: data mining, knowledge discovery in data bases, quantitative association rules, statistical inference
theory
1. Introduction

  

Source: Aumann, Yonatan - Computer Science Department, Bar Ilan University

 

Collections: Computer Technologies and Information Sciences