The 2nd Conformance Checking Challenge (CCC 2020) is a co-located event of the 2nd International Conference on Process Mining 2020. The Challenge provides participants with artifacts stemmed from real processes and invites them to analyze the conformance between the observed (event logs) and modelled (process models) behaviors. The Challenge requests the participants to submit efficient, useful and understandable conformance analytics for the provided models and logs in terms of a report. We further encourage participants to use any tools or techniques at their disposal, including open-source or proprietary ones, as well as those implemented specifically for this Challenge.
Abstract Submission Deadline: 23 August 2020
Report Submission Deadline: 31 August 2020
- Abel Armas-Cervantes, The University of Melbourne
- Artem Polyvyanyy, The University of Melbourne
- Gert Janssenswillen, Hasselt University
- Luciano García-Bañuelos, Tecnológico de Monterrey
Challenge & Problem Statement
Given a dataset consisting of several process models and event logs, describe an “interesting” process conformance analytics that summarizes commonalities and discrepancies between the processes described in the provided process models and event logs. Using your analytics, for each process model, suggest an event log from which it was most likely discovered.
Your report must address the assessment criteria listed below.
The reports will be evaluated by a group of experts in process mining. To determine the winner(s), each submitted report will be assessed against the criteria listed below. The organizers may decide to nominate several winners for different categories, e.g., the student, academic and industry winner, depending on the number and quality of the received submissions.
Demonstrate the benefits of applying your process conformance analytics, both for the proposed dataset and in general, including:
- benefits of using your analytics for analyzing the proposed dataset;
- confirmed and potential benefits of using your analytics in organizations;
- data science, business analytics, and business intelligence use cases your analytics can support/enable;
- business decisions your analytics can informate and/or automate; and
- evidence that your analytics is insightful for the defined stakeholders.
Discuss how your process conformance analytics is different from that already presented in the literature, including:
- contributions to gaps in research on conformance checking; and
- novelty of the insights on commonalities and discrepancies between the analyzed processes your analytics delivers.
Discuss how efficient and, hence, feasible in practice are the algorithms that construct your process conformance analytics, including:
- theoretical aspects of the computational complexity of the algorithms;
- practical aspects of the computational complexity of the algorithms; and
- trade-offs of computational resources for constructing your analytics.
Discuss aspects related to the understandability of your process conformance analytics by prospective stakeholders, including:
- fit of your analytics for the purpose;
- clarity of your analytics for domain experts;
- clarity of your analytics for non-domain experts; and
- cognitive load associated with interpreting your analytics.
Apply your process conformance analytics to correctly relate process models to event logs from which they were constructed.
The dataset provided with this Challenge includes five event logs and five process models. The five event logs are the filtered, anonymized and manipulated versions of the five event logs in the CoSeLoG project dataset (https://doi.org/10.4121/uuid:26aba40d-8b2d-435b-b5af-6d4bfbd7a270). The five process models were discovered from the preprocessed event logs using a state-of-the-art process discovery technique.
The dataset of the Challenge is made publicly available at the challenge repository (https://github.com/ccchallenge/datasets).
For more information on the CoSeLoG project, refer to: https://doi.org/10.4121/uuid:26aba40d-8b2d-435b-b5af-6d4bfbd7a270.
A submission should contain a PDF Report of at most 20 pages, including figures, tables and appendices, in the LNCS/LNBIP format (http://www.springer.com/computer/lncs?SGWID=0-164-6-791344-0) specified by Springer (available for both LaTeX and MS Word). Submissions must be made through EasyChair at https://easychair.org/conferences/?conf=icpm2020, where you must indicate that your submission is to the Conformance Checking Challenge.
The winner team will be selected by a jury and invited to present the results at the conference. They will also receive a certificate, and a small token of appreciation.
Direct your questions related to the Challenge to the organizers via firstname.lastname@example.org