Mining top-K association rules

Philippe Fournier-Viger*, Cheng Wei Wu, Vincent S. Tseng

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

43 Scopus citations


Mining association rules is a fundamental data mining task. However, depending on the choice of the parameters (the minimum confidence and minimum support), current algorithms can become very slow and generate an extremely large amount of results or generate too few results, omitting valuable information.This is a serious problem because in practice users have limited resources for analyzing the results and thus are often only interested in discovering a certain amount of results, and fine tuning the parameters is time-consuming.To address this problem, we propose an algorithm to mine the top-k association rules, where k is the number of association rules to be found and is set by the user. The algorithm utilizes a new approach for generating association rules named rule expansions and includes several optimizations. Experimental results show that the algorithm has excellent performance and scalability, and that it is an advantageous alternative to classical association rule mining algorithms when the user want to control the number of rules generated.

Original languageEnglish
Title of host publicationAdvances in Artificial Intelligence - 25th Canadian Conference on Artificial Intelligence, Canadian AI 2012, Proceedings
Number of pages13
StatePublished - 6 Jun 2012
Event25th Canadian Conference on Artificial Intelligence, AI 2012 - Toronto, ON, Canada
Duration: 28 May 201230 May 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7310 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference25th Canadian Conference on Artificial Intelligence, AI 2012
CityToronto, ON


  • association rule mining
  • rule expansion
  • support
  • top-k rules

Fingerprint Dive into the research topics of 'Mining top-K association rules'. Together they form a unique fingerprint.

Cite this