Advanced SIKS/LOIS Course on Process Mining and Data Mining (PMDM) The SIKS/LOIS Course on Process Mining and Data Mining (PMDM) will take place on October 26-27 on the campus of the Technische Universiteit Eindhoven (TU/e). This course is intended for people who want to learn more about process and data mining and the interplay between both. The course will be followed by the LOIS workshop "Process Mining meets Data Mining" that takes place on Wednesday October 28 and the Benelux Conference on Artificial Intelligence (BNAIC) that takes place on October 29-30. These events are organized by Eindhoven University of Technology (TU/e) under the auspices of the Belgium-Netherlands Association for Artificial Intelligence (BNVKI), the Dutch Research School for Information and Knowledge Systems (SIKS), and the IEEE Task Force on Process Mining. Although people are stimulated to participate in all three co-located events, it is possible to take the SIKS/LOIS course separately. No prior knowledge of data and/or process mining is assumed from the participants. The SIKS/LOIS course is part of the Advanced Components Stage of SIKS educational program. Therefore, SIKS-PhD-students working on the SIKS foci Enterprise Information Systems and Computational Intelligence are strongly encouraged to participate. PROGRAM Day 1 (Monday October 26) On the first day the primary focus is on data mining. Data mining is a thriving research discipline aimed at developing automatic tools for extracting information from huge data collections. The need for data mining emerged due to newly developed technologies for gathering and storing data. Whereas only few decades ago, shop owners kept track of sales manually, nowadays supermarkets scan and record every single purchase, collecting huge databases. As most of the existing analysis techniques did not scale up to these unprecedented amounts of data, the need for new computational analysis techniques arose and data mining emerged as a discipline. During the first day of this course, we will provide an overview of the most popular data mining techniques, including: clustering, classification, and pattern mining. For all data mining tasks, special attention will be paid to the following aspects: "Which problems do they solve," "What are the major computational issues and how are they addressed algorithmically?" and "What are the strengths and limitations of the different techniques? 09.00 Coffee 09.30 Overview of the course and an introduction to the data mining field [Toon Calders] 11.00 Coffee 11.30 Association Rules and Pattern Mining [Toon Calders] 12.30 Lunch 13.30 Clustering [Mykola Pechenizkiy] 15.00 Coffee 15.30 Classification [Mykola Pechenizkiy] 16.30 From Data Mining to Process Mining [Ton Weijters] 17.00 Closing of first day Day 2 (Tuesday October 27) The second day is devoted to process mining. Process mining addresses the problem that most people have very limited information about what is actually happening in their organization. In practice, there is often a significant gap between what is prescribed or supposed to happen, and what actually happens. Only a concise assessment of the organizational reality, which process mining strives to deliver, can help in verifying process models, and ultimately be used in a process redesign effort or BPMS implementation. During the second day an overview is given of contemporary process mining techniques. After explaining the basic ideas and introducing three types of process mining (process discovery, conformance checking, and model extension), three approaches are presented in more detail. First, the so-called Alpha algorithm is explained. This is a simple algorithm that is able to discover all kinds of control-flow structures, but that cannot deal with noise and infrequent behavior. After introducing the basics, a heuristics driven process discovery is presented (Heuristics Miner). Finally, the attention shifts from discovery to conformance checking and performance measurements. It is shown that event logs can be replayed on models to assess their quality and to enrich them. All process mining approaches are illustrated by applying the process mining tool ProM to concrete event logs. There will also be hands-on exercises. 08.30 Coffee 09.00 Overview of process mining (discovery, conformance, and extension) [Wil van der Aalst] 10.30 Coffee 11.00 Process discovery and the Alpha algorithm [Wil van der Aalst] 12.30 Lunch 13.30 Mining less structured process models: The Heuristics Miner [Ton Weijters] 15.00 Coffee 15.30 Replay analysis for conformance checking and performance visualization [Boudewijn van Dongen] 17.00 Closing of course ORGANIZATION This course is organized by: - Wil van der Aalst (http://wwwis.win.tue.nl/~wvdaalst/), - Paul de Bra (http://wwwis.win.tue.nl/~debra/), - Toon Calders (http://wwwis.win.tue.nl/~tcalders), - Boudewijn van Dongen (www.processmining.org), - Mykola Pechenizkiy (http://www.win.tue.nl/~mpechen/), and - Ton Weijters (http://is.tm.tue.nl/staff/aweijters/). For more information about the content of the course contact one of the persons above. To register send an e-mail to Ine van der Ligt -van de Moosdijk (wsinfsys@tue.nl). Registration is required. Lunches are included but there is no registration fee. The course will take place in De Zwarte Doos (www.dezwartedoos.nl) at the TU/e campus (http://w3.tue.nl/en/the_university/route_and_map/).