Read out

Guest Talk: Evidence-Based Conformance Checking for Event Logs with Coarse-grained Timestamps

Dr. Henrik Leopold

Date/Time: 07.05.2018, 11:00

Location: D2.2.094 

Abstract 

Conformance-checking techniques are an efficient means to detect undesired behavior. They provide the possibility to automatically detect differences between actual behavior and desired behavior by comparing an event log with a process model. However, a key assumption of traditional conformance-checking techniques is that all events in the event log are labeled with timestamps that allow to infer the event order. Unfortunately, this is not always the case in practice. One of the most common issues is that timestamps are too course-grained. They might, for instance, reveal the day but not the exact time of an event. For such event logs, traditional conformance-checking techniques are not able to produce useful results. Recognizing this limitation, we propose an approach that uses information from the entire event log to derive the most likely execution sequence for a given case without a clear event order. We use a large event log from the healthcare domain to illustrate that our approach indeed allows conducting more accurate conformance checking and provides additional analytic insights.

Bio 

Dr. Henrik Leopold is an Assistant Professor at the Department of Computer Science at the Vrije Universiteit Amsterdam (VU). Before joining the VU in February 2015, he held positions as an Assistant Professor at the WU Vienna and as a postdoctoral research fellow at the Humboldt University of Berlin. In July 2013, he obtained a PhD degree in Information Systems from the Humboldt University of Berlin. His doctoral thesis received the German TARGION Award 2014 for the best dissertation in the field of strategic information management. His research is mainly concerned with the interplay between information systems and business processes. He is particularly interested in how to leverage technology from the field of artificial intelligence (such as machine learning and natural language processing) to analyze and support the execution of business processes. He has published over 60 research papers and articles, among others, in Data & Knowledge Engineering, Decision Support Systems, IEEE Transactions on Software Engineering, and Information Systems.



Back to overview