1 / 21

Alignment-based Precision Checking

Tallinn, 3 September 2012. Alignment-based Precision Checking. A. Adriansyah 1 , J. Munoz Gamma 2 , J. Carmona 2 , B.F. van Dongen 1 , W.M.P. van der Aalst 1. Department of Mathematics and Computer Science, Eindhoven University of Technology, The Netherlands

ludlow
Télécharger la présentation

Alignment-based Precision Checking

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Tallinn, 3 September 2012 Alignment-based Precision Checking A. Adriansyah1, J. Munoz Gamma2, J. Carmona2, B.F. van Dongen1, W.M.P. van der Aalst1 Department of Mathematics and Computer Science, Eindhoven University of Technology, The Netherlands Software Department, UniversitatPolitècnicade Catalunya, Spain

  2. Model needs to be precise Event Log How to measure precision for unfitting event logs? aabd abd acd aabd c a d d a b b a Unfitting a b d end end c c d Process Models

  3. Overview Unfitting Log aabbd adabd a d … Alignment Automaton Alignment Precision Metrics Fitting Traces aabd aabd acd … Process Model Experiments

  4. Optimal alignment: least #deviations Trace adab aabd Reference: W.M.P. van der Aalst, A. Adriansyah, B.F. van Dongen(2012). Replaying history on process models for conformance checking and performance analysis. WIREs Data Mining and Knowledge Discovery, 2(2), 182-192. Deviation Log Moves a a d b Alignment a a b d Fitting trace, closest to the original Model Moves Deviation b a Process Model a b d end c

  5. Optimal alignments can be many Trace ad abd or acd Alignment 1 Alignment 2 a d a d d d a c a b b a Both alignments are optimal Process Model a b d end c

  6. Overview Unfitting Log aabbd adabd a d … Alignment Automaton Alignment Precision Metrics Fitting Traces aabd aabd acd … Process Model Experiments

  7. 1-Alignment automaton (Prefix-based) Fitting Traces 1-Align Automaton aabd abd abd or acd aabd aabd abd abd aabd b d 2 a 2 2 2 2 a b d 4 4 b a a b d end c Process Model

  8. 1-Alignment automaton (Prefix-based) Fitting Traces 1-Align Automaton aabd abd abd aabd b d 2 a 2 2 2 2 … a b d 4 4 c 0 b a Imprecision a b d end … c Process Model

  9. All-Alignment Automaton (Prefix-based) Fitting Traces All-Alignments Automaton aabd abd abd or acd aabd 1 1 1 b d 2 a 2 2 1 1.5 1.5 1 2 3 3 2 a b d 4 4 0.5 0.5 c d b a a b d end c Process Model

  10. Comparison of log automata Without alignment Fitting Traces Event Log aabd abd abd/acd aabd aabd abd ad adab 1-Align Prefix-based Log Automata b a All-Align a b d end Process Model c

  11. Overview Unfitting Log aabbd adabd a d … Alignment Automaton Alignment Precision Metrics Fitting Traces aabd aabd acd … Process Model Use calculation Experiments Reference: J. Muñoz-Gama, J. Carmona (2010). A Fresh Look at Precision in Process Conformance. 8th International Conference of Business Process Management (BPM): 211-226.

  12. Overview Unfitting Log aabbd adabd a d … Alignment Automaton Alignment Precision Metrics Fitting Traces aabd aabd acd … Process Model Experiments

  13. Implementation in ProM 6 Package: “ETConformance”

  14. Experiment settings Event Log Duplicate Log Combined Log Flower (F) Flower’ (F’) Precise-Precise (PP’) Precise-Flower (PF’) Precise (P) Precise’ (P’) Flower -Precise (FP’) Flower-Flower (FF’)

  15. Experiment 1: Measuring precision stability Non-fitting logs and models Perfectly fitting logs and models P FP’ PF’ PP’ F P FF’ FP’ PF’ PP’ Values of change significantly, but not and

  16. Experiment 2: Sensitivity to unfitting trace Model P Model PF’ Precision Precision #Removed events #Removed events Model FP’ Model PP’ Precision Precision #Removed events #Removed events fluctuates when traces are unfitting

  17. Experiment 3: Real life cases Using 5 logs and models from municipalities in the Netherlands Computation time of is much shorter than , yet provide close estimation

  18. Overview Unfitting Log aabbd adabd a d … Alignment Automaton Alignment Precision Metrics Fitting Traces aabd aabd acd … Process Model Experiments

  19. Conclusion and Future Work Pre-alignment provides more accurate precision measurements in cases where logs are non-fitting Applicable to models with duplicate tasks/unlogged activities gives close estimation to with lower computation time Log-Model evaluation needs to consider multiple dimension of conformance Potential use of the technique: auditing, conformance diagnosis, process repair

  20. Questions

  21. Thanks!

More Related