Bayesian classifcation of events for task labeling using workfow models

Download
  1. (PDF, 391 KB)
  2. Get@NRC: Bayesian classifcation of events for task labeling using workfow models (Opens in a new window)
AuthorSearch for: ; Search for:
TypeArticle
Proceedings titleBusiness Process Management Workshops
Series titleLNBIP (Lecture Notes in Business Information Processing); Volume 17
ConferenceThe 4th Workshop on Business Process Intelligence (BPI 08) in conjunction with Business Process Management (BPM 2008), 1 September 2008, Milan, Italy
ISSN1865-1348
ISBN978-3-642-00327-1
Pages97108; # of pages: 12
Subjectworkflow; process mining; task labeling; Bayesian classification
AbstractWe investigate a method designed to improve accuracy of work°ow mining in the case that the identi¯cation of task labels for log events are uncertain. Here we consider how the accuracy of an indepen- dent task identi¯er, such as a classi¯cation or clustering engine, can be improved by examining work°ow. After brie°y introducing the notion of iterative work°ow mining, where the mined work°ow is used to help improve the true task labelings which, when re-mined, will produce a more accurate work°ow model, we demonstrate a Bayesian updating ap- proach to determining posterior probabilities for each label for a given event, by considering the probabilities from the previous step as well as information as to the beliefs of the labels that can be gained by exam- ining the work°ow model. Experiments show that labeling accuracy can be increased signi¯cantly, resulting in more accurate work°ow models.
Publication date
PublisherSpringer Verlag
LanguageEnglish
AffiliationNational Research Council Canada; NRC Institute for Information Technology
Peer reviewedYes
NRC number50389
NPARC number5764119
Export citationExport as RIS
Report a correctionReport a correction
Record identifier9f83c326-6d42-4eb3-956d-6218d1385ddf
Record created2009-03-29
Record modified2016-05-09
Bookmark and share
  • Share this page with Facebook (Opens in a new window)
  • Share this page with Twitter (Opens in a new window)
  • Share this page with Google+ (Opens in a new window)
  • Share this page with Delicious (Opens in a new window)