Hybrid technique for frequent pattern extraction from Sequential Database

Rajalakshmi Selvaraj, Venu Madhav Kuthadi, Tshilidzi Marwala

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

Abstract

Data mining has became a familiar tool for mining stored value from the large scale databases that are known as Sequential Database. These databases has large number of itemsets that can arrive frequently and sequentially, it can also predict the users behaviors. The evaluation of user behavior is done by using Sequential pattern mining where the frequent patterns extracted with several limitations. Even the previous sequential pattern techniques used some limitations to extract those frequent patterns but these techniques does not generated the more reliable patterns .Thus, it is very complex to the decision makers for evaluation of user behavior. In this paper, to solve this problem a technique called hybrid pattern is used which has both time based limitation and space limitation and it is used to extract more feasible pattern from sequential database. Initially, the space limitation is applied to break the sequential database using the maximum and minimum threshold values. To this end, the time based limitation is applied to extract more feasible patterns where a bury-time arrival rate is computed to extract the reliable patterns.

Original languageEnglish
Title of host publicationProceedings of the 3rd International Conference on Frontiers of Intelligent Computing
Subtitle of host publicationTheory and Applications, FICTA 2014
EditorsSiba K. Udgata, Bhabendra Narayan Biswal, Jyotsna Kumar Mandal, Suresh Chandra Satapathy
PublisherSpringer Verlag
Pages265-275
Number of pages11
ISBN (Electronic)9783319119328
DOIs
Publication statusPublished - 2014
Event3rd International Conference on Frontiers in Intelligent Computing: Theory and Applications, FICTA-2014 - Reykjavík, Iceland
Duration: 14 Nov 201415 Nov 2014

Publication series

NameAdvances in Intelligent Systems and Computing
Volume327
ISSN (Print)2194-5357

Conference

Conference3rd International Conference on Frontiers in Intelligent Computing: Theory and Applications, FICTA-2014
Country/TerritoryIceland
CityReykjavík
Period14/11/1415/11/14

Keywords

  • Data mining
  • Hybrid pattern mining
  • Sequential pattern

ASJC Scopus subject areas

  • Control and Systems Engineering
  • General Computer Science

Fingerprint

Dive into the research topics of 'Hybrid technique for frequent pattern extraction from Sequential Database'. Together they form a unique fingerprint.

Cite this