 |
Paris Descartes University Seminar Series on Data Analytics Invited Seminar Talk |
 |
|
Prof. Matthias Weidlich, Humboldt University of Berlin (Germany) when: 14 March 2018, 11am where: room Turing Reunion, 7th floor, Paris Descartes University, 45 Rue Des Saints Peres, Paris 75006 Abstract In domains such as e-commerce, logistics, or healthcare, the conduct of service processes is widely supported by information systems and event data is generated continuously during process execution. Such event data constitutes a valuable source of information. It serves as the basis for techniques that target monitoring and improvement of service processes and their supporting information systems. In this talk, we outline a pipeline for event-driven analysis of service processes. As a first step, this pipeline comprises a method to learn abstractions of event data suitable for process analysis. Taking pre-processed event data as a starting point, we then present an approach to balance overfitting and underfitting in the discovery of analysis models. Finally, we turn to online monitoring of processes based on these analysis models, and show how to optimise the evaluation of event queries based on insights into the behaviour of a service process. Short Bio Matthias Weidlich leads the Process-Driven Architectures group at the Department of Computer Science at Humboldt-Universitäzu Berlin (HU), which is funded by the German Research Foundation in the Emmy-Noether Programme. Earlier, he held positions at Imperial College London, UK, and the Technion, Israel. He holds a PhD in Computer Science from the Hasso-Plattner-Institute, University of Potsdam, Germany. His research focuses on process-oriented and event-driven information systems, including methods for their specification, their data-driven analysis, and optimisations of their run-time behaviour. He is a Junior-Fellow of the German Informatics Society and recipient of the Berlin Young Researcher Award 2016. Hosted by: Themis Palpanas List of past seminars |