Rozinat's Conformance Framework Port for ProM 6
Dr. Anne Rozinat's conformance checker framework remains one of the most well-known and seminal conformance checking frameworks in the process mining community. Sadly, its implementation—a ProM 5 plugin—has never been ported to ProM 6. Until now.
[Dr. Anne Rozinat's](http://www.fluxicon.com/team/) conformance checker framework remains one of the most famous and seminal conformance checking frameworks in the process mining community, introducing the token based fitness, successful execution, proper completion, behavioral appropriateness and structural appropriateness metrics.
Its implementation, however, has become increasingly more difficult to use by end-users and researchers, as it remains implemented as a [ProM 5 plugin](http://www.processmining.org/online/conformance_checker) only.
![Screenshot of ProM 5 plugin](images/screenshot.png)
Thanks to our previous efforts put into our [CoBeFra](/cobefra) benchmarking suite, we are able to present a port, making Rozinat's conformance checking plugin finally available for use within ProM 6.
* [Rozinat's original conformance checker documentation](http://www.processmining.org/online/conformance_checker)
* A. Rozinat and W.M.P. van der Aalst. Conformance Checking of Processes Based on Monitoring Real Behavior. Information Systems, Volume 33, Issue 1, Pages 64-95.
We would appreciate that you cite [CoBeFra](/cobefra) when you use this port:
* vanden Broucke, S., De Weerdt, J., Vanthienen, J., Baesens, B. (2013). A comprehensive benchmarking framework (CoBeFra) for conformance analysis between procedural process models and event logs in ProM. Proceedings of the IEEE Symposium on Computational Intelligence and Data Mining, CIDM 2013, part of the IEEE Symposium Series on Computational Intelligence 2013, SSCI 2013. IEEE Symposium on Computational Intelligence and Data Mining (CIDM 2013). Singapore, 16-19 April 2013 (pp. 254-261). New York, USA: IEEE.
This port is made available as a [ProM 6](//www.promtools.org) plugin. The following JAR file contains the plugin:
* [Version of 2018-07-26](downloads/rozinatconformance-20180726.jar)
Source code is made available as well:
* [Source code archive](downloads/rozinatconformance-src.zip)
The plugin depends on the `Graphviz` ProM 6 package, which is normally installed by default.
You will need to make sure that ProM can find the downloaded JAR in its classpath. To do so, you can create a folder `plugins` in the ProM installation directory, place the downloaded JAR file in this directory, and start ProM with the following command (Windows example for ProM 6.7, adjust accordingly):
java -classpath ./plugins/*;./ProM67_lib/*;./ProM67_dist/*;./* -Djava.util.Arrays.useLegacyMergeSort=true -Djava.library.path=./ProM66_lib -ea -Xmx2g -XX:MaxPermSize=512m -XX:+UseCompressedOops org.processmining.contexts.uitopia.UI
*Important:* note that the plugin was ported with a minimal number of modifications. Any errors or scalability issues present in the original version are thus still present in this port. That said, it should be useful for researchers and instructors. The following modifications have been performed:
* Dependency on Grappa was removed completely
* Drawing of Petri nets is still done by calling `dot`, but output is now shown as an image. This has the downside that nodes and edges are not selectable, but this was not used in the original plugin anyway (power users can inspect `stdout` to figure out the location of the image files, which can then be inserted in their documents). The upside is that a huge amount of code could be streamlined
* Petri net-related classes were modified to fix problematic issues related to `clone()`
* Various unused classes and methods were removed
* Presence of GUI aspects stemming from ProM 5 were removed
* It is possible to set a timeout for various steps in the algorithm
* It is possible to just use a loaded log and Petri net from within ProM 6. These are converted to ProM 5 representations as best as possible, but errors might occur in case of strange naming schemes (e.g. transitions which strange characters)
Contact the authors at:
* [Seppe vanden Broucke](mailto:email@example.com) (corresponding author)
Department of Decision Sciences and Information Management, KU Leuven
Naamsestraat 69, B-3000 Leuven, Belgium