Mining Unlabeled Event Log

Most of the organization perform their business process in an unmanaged environment, in which the information systems perform the activities with the lack of knowledge about the business process. Leading to the inability of correlating running activity to their process instances due to the missing case identifier within the recorded events. The uncorrelated events are so-called unlabeled event log. The book addresses the problem of correlating the unlabeled events and generate all the possible corresponding labeled event logs, especially when the event logs are generated from cyclic business processes. The book proposes a new event correlation approach to correlate the unlabeled events from the cyclic and acyclic business processes, which composed of main technique which is Deducing Case ID (DCI) and a preprocessing step to generate Relationship Matrix. DCI correlates the events based on the knowledge of (a) the relation between the activities, and (b) the execution behavior of the activities. The output of the event correlation approach is the labeled logs that can be used by the process mining techniques to evaluate and improve the business process

Autorentext

Dina Bayomie interested in BPM and process mining fields. she got her B.S and M.Sc. from Faculty of Computers and Information, Cairo University. Her master research focuses on the uncorrelated events problem which is discussed in this book. Currently, she is continuing her research as a Ph.D. student in Insitute for information Business, WU Vienna.