BROWSE

Related Researcher

Author's Photo

Song, Minseok
Business Process Intelligence(BPI) Lab
Research Interests
  • Process mining
  • Social network analysis/Organizational structure analysis
  • Process knowledge management
  • Business process analysis, simulation, and optimization
  • Data mining/business intelligence

ITEM VIEW & DOWNLOAD

A comparative study of dimensionality reduction techniques to enhance trace clustering performances

DC Field Value Language
dc.contributor.author Song, Minseok ko
dc.contributor.author Yang, H. ko
dc.contributor.author Siadat, S. H. ko
dc.contributor.author Pechenizkiy, M. ko
dc.date.available 2014-04-10T01:52:15Z -
dc.date.created 2013-06-28 ko
dc.date.issued 2013-07 -
dc.identifier.citation EXPERT SYSTEMS WITH APPLICATIONS, v.40, no.9, pp.3722 - ko
dc.identifier.issn 0957-4174 ko
dc.identifier.uri https://scholarworks.unist.ac.kr/handle/201301/3361 -
dc.identifier.uri http://www.scopus.com/inward/record.url?partnerID=HzOxMe3b&scp=84874651427 ko
dc.description.abstract Process mining techniques have been used to analyze event logs from information systems in order to derive useful patterns. However, in the big data era, real-life event logs are huge, unstructured, and complex so that traditional process mining techniques have difficulties in the analysis of big logs. To reduce the complexity during the analysis, trace clustering can be used to group similar traces together and to mine more structured and simpler process models for each of the clusters locally. However, a high dimensionality of the feature space in which all the traces are presented poses different problems to trace clustering. In this paper, we study the effect of applying dimensionality reduction (preprocessing) techniques on the performance of trace clustering. In our experimental study we use three popular feature transformation techniques; singular value decomposition (SVD), random projection (RP), and principal components analysis (PCA), and the state-of-the art trace clustering in process mining. The experimental results on the dataset constructed from a real event log recorded from patient treatment processes in a Dutch hospital show that dimensionality reduction can improve trace clustering performance with respect to the computation time and average fitness of the mined local process models. ko
dc.description.statementofresponsibility close -
dc.language ENG ko
dc.publisher PERGAMON-ELSEVIER SCIENCE LTD ko
dc.subject Big datum ko
dc.subject Comparative studies ko
dc.subject Computation time ko
dc.subject Dimensionality reduction ko
dc.subject Dimensionality reduction techniques ko
dc.subject Dutch hospitals ko
dc.subject Event logs ko
dc.subject Experimental studies ko
dc.subject Feature space ko
dc.subject Feature transformations ko
dc.subject High dimensionality ko
dc.subject In-process ko
dc.subject PCA ko
dc.subject Principal components analysis ko
dc.subject Process mining ko
dc.subject Process model ko
dc.subject Random projections ko
dc.subject State of the art ko
dc.subject Trace clustering ko
dc.subject Treatment process ko
dc.subject Useful patterns ko
dc.title A comparative study of dimensionality reduction techniques to enhance trace clustering performances ko
dc.type ARTICLE ko
dc.identifier.scopusid 2-s2.0-84874651427 ko
dc.identifier.wosid 000316581300036 ko
dc.type.rims ART ko
dc.description.scopustc 4 *
dc.date.scptcdate 2014-07-12 *
dc.identifier.doi 10.1016/j.eswa.2012.12.078 ko
Appears in Collections:
SBA_Journal Papers

find_unist can give you direct access to the published full text of this article. (UNISTARs only)

Show simple item record

qrcode

  • mendeley

    citeulike

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

MENU