LEFRA: Learning from Associations

Ray R. Hashemi, Louis LeBlanc, Bart J. Westgeest, Alexander A. Tyler

Research output: Contribution to book or proceedingConference articlepeer-review

1 Scopus citations

Abstract

In data mining, a Multi-level Association Analysis (MAA) produces a set of association rules. These rules mainly identify those values of multiple attributes that arc associated to each other. In this paper, we introduce a new learning paradigm based on association rules called "Learning from Association (LEFRA)" which is used as a part of a predictive system to predict the effect of a number of carcinogens on liver. The validity of the proposed learning paradigm is established by comparing its performance with the performance of logistic regression which has been applied on the same datasct.

Original languageEnglish
Title of host publicationSoft Computing with Industrial Applications - International Symposium on Soft Computing for Industry, ISSCI - Proceedings of the Sixth Biannual World Automation Congress, WAC 2004
EditorsM. Jamshidi, M. Reuter, D. Andina, J.S. Jamshidi
Pages549-554
Number of pages6
StatePublished - 2004
EventSoft Computing with Industrial Applications - International Symposium on Soft Computing for Industry, ISSCI - Sixth Biannual World Automation Congress, WAC 2004 - Sevilla, Spain
Duration: Jun 28 2004Jul 1 2004

Publication series

NameSoft Computing with Industrial Applications - Proceedings of the Sixth Biannual World Automation Congress

Conference

ConferenceSoft Computing with Industrial Applications - International Symposium on Soft Computing for Industry, ISSCI - Sixth Biannual World Automation Congress, WAC 2004
Country/TerritorySpain
CitySevilla
Period06/28/0407/1/04

Scopus Subject Areas

  • General Engineering

Keywords

  • Association Analysis
  • Data Mining
  • Learning from Association
  • Predictive Systems

Fingerprint

Dive into the research topics of 'LEFRA: Learning from Associations'. Together they form a unique fingerprint.

Cite this