A framework for automated association mining over multiple databases

Esma Nur Çinicioǧlu, Gürdal Ertek, Deniz Demirer, Hasan Ersin Yörük

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

Literature on association mining, the data mining methodology that investigates associations between items, has primarily focused on efficiently mining larger databases. The motivation for association mining is to use the rules obtained from historical data to influence future transactions. However, associations in transactional processes change significantly over time, implying that rules extracted for a given time interval may not be applicable for a later time interval. Hence, an analysis framework is necessary to identify how associations change over time. This paper presents such a framework, reports the implementation of the framework as a tool, and demonstrates the applicability of and the necessity for the framework through a case study in the domain of finance.

Original languageEnglish
Title of host publicationINISTA 2011 - 2011 International Symposium on INnovations in Intelligent SysTems and Applications
Pages79-85
Number of pages7
DOIs
Publication statusPublished - 2011
Externally publishedYes
Event2011 International Symposium on INnovations in Intelligent SysTems and Applications, INISTA 2011 - Istanbul-Kadikoy, Turkey
Duration: Jun 15 2011Jun 18 2011

Publication series

NameINISTA 2011 - 2011 International Symposium on INnovations in Intelligent SysTems and Applications

Conference

Conference2011 International Symposium on INnovations in Intelligent SysTems and Applications, INISTA 2011
Country/TerritoryTurkey
CityIstanbul-Kadikoy
Period6/15/116/18/11

Keywords

  • association mining
  • association mining over multiple databases
  • association mining visualization
  • data mining
  • graph visualization

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Information Systems

Fingerprint

Dive into the research topics of 'A framework for automated association mining over multiple databases'. Together they form a unique fingerprint.

Cite this