AGNEsMiner is a declarative machine learning technique that represents the problem of process discovery as a multi-relational classification problem on event logs supplemented with Artificially Generated Negative Events (AGNEs).About ----- Process discovery is the automated construction of structured process models from information system event logs. Such event logs often contain positive examples only. Without negative examples, it is a challenge to strike the right balance between recall and precision, while dealing with problems such as expressiveness, noise, and incomplete event logs. AGNEsMiner is a declarative machine learning technique that deals with these challenges by representing process discovery as a multi-relational classification problem on event logs supplemented with Artificially Generated Negative Events (AGNEs). The AGNEs technique has been implemented as a mining plugin in the ProM Framework of the TU/Eindhoven. The implementation below is a rework of the original plugin by dr. Stijn Goedertier (for ProM 5), now available for ProM 6. The revised edition allows to induce artifial negative events using the original (slower) induction strategy, or the faster induction strategy by dr. vanden Broucke ([see also this page](http://www.processmining.be/neconformance/)). References ---------- If you use AGNEsMiner in your work, please cite the following references: * Goedertier, S., Martens, D., Vanthienen, J., Baesens, B. (2009). Robust process discovery with artificial negative events. The Journal of Machine Learning Research, 10, 1305-1340. * vanden Broucke, S., De Weerdt, J., Vanthienen, J., Baesens, B. (2014). Determining process model precision and generalization with weighted artificial negative events. IEEE Transactions on Knowledge and Data Engineering, 26 (8), 1877-1889. * De Weerdt, J., Vanthienen, J., Baesens, B. (2014). Determining process model precision and generalization with weighted artificial negative events. Knowledge and Data Engineering, IEEE Transactions on, 26(8), 1877-1889. Implementation -------------- AGNEsMiner has been re-implemented as a [ProM 6](//www.promtools.org) plugin. The following JAR file contains the plugin: * [Version of 2020-11-18](downloads/agnesminer-20201117.jar) (bug fix for newer SWI-Prolog versions) * [Version of 2017-11-07](downloads/agnesminer-20171107.jar) (improved handling for non-standard logs) The plugin depends on “SWI-Prolog”. Download and install the *latest stable* release from [here](http://www.swi-prolog.org/download/stable). The 64-bit version is recommended, if your system supports it. Note that you will need to specify the full path of the SWI-Prolog executable in the wizard screen of the plugin, e.g. `C:\Program Files\swipl\bin\swipl.exe`. The plugin also depends on the “ACE” inductive logic programming data mining environment, which can be obtained [here](https://dtai.cs.kuleuven.be/ACE/). We have tested the plugin with version `ACE-1.2.8-b1` only. Contact us (see below) should you fail to obtain or install ACE. Note that you will need to specify the full path of the ACE executable in the wizard screen of the plugin, e.g. `C:\Program Files\ACE-1.2.8-b1\windows\bin\ACE.exe`. 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): java.exe -Xmx4G -da -classpath "./dist/*;./lib/*;./plugins/*" -Djava.library.path=.//lib -Djava.util.Arrays.useLegacyMergeSort=true org.processmining.contexts.uitopia.UI The legacy version of AGNEsMiner (ProM 5) is available for download [here](http://perswww.kuleuven.be/~u0041863/AGNEs.php). Contact ------- Contact the authors at: * [Seppe vanden Broucke](mailto:seppe.vandenbroucke@kuleuven.be) (corresponding author)
Department of Decision Sciences and Information Management, KU Leuven
Naamsestraat 69, B-3000 Leuven, Belgium Screenshots ----------- [
](#i01)
[
](#i02)
[
](#i03)
[
](#i05)
[
](#i06)
[
](#i04)