Abstract
Novelty detection refers to the task of finding observations that are new or unusual when compared to the ‘known’ behavior. Its practical and challenging nature has been proven in many application domains while in process mining field has very limited researched. In this paper we propose a framework for the multi-modal analysis of novel behavior in business processes. The framework exploits the potential of representation learning, and allows to look at the process from different perspectives besides that of the control flow. Experiments on a real-world dataset confirm the quality of our proposal.
Original language | American English |
---|---|
Title of host publication | Intelligent Data Engineering and Automated Learning - IDEAL 2020: 21st International Conference, Guimaraes, Portugal, November 4-6, 2020, Proceedings |
State | Published - 1 Jan 2020 |
Keywords
- multi-modality
- novelty detection
- process mining
- trace embedding
EGS Disciplines
- Computer Sciences