Real-time probing of control-flow and data-flow in event logs

Paolo Ceravolo, Ernesto Damiani, Emilio Francesco Schepis, Gabriel Marques Tavares

Research output: Contribution to journalConference articlepeer-review

2 Scopus citations

Abstract

Traditional Process Mining offers batch analysis of business processes but does not transpose smoothly into online environments due to specific design constraints. Techniques adapted to support online analysis require peculiar adjustments that inherently restrict their focus to a single task. In this work, we extend the Concept Drift in Event Stream Framework (CDESF) tool to handle multiple attributes simultaneously. Our extension promotes CDESF to analyze both control-flow and data-flow characteristics in online event streams. Experiments used real and synthetic data for concept drift and anomaly detections. Results show that additional perspectives should be considered as they contain valuable information about processes.

Original languageBritish English
Pages (from-to)751-758
Number of pages8
JournalProcedia Computer Science
Volume197
DOIs
StatePublished - Jan 2022
Event6th Information Systems International Conference, ISICO 2021 - Virtual, Online, Italy
Duration: 7 Aug 20218 Aug 2021

Keywords

  • Anomaly detection
  • Clustering
  • Concept drift detection
  • Event stream
  • Online process mining

Fingerprint

Dive into the research topics of 'Real-time probing of control-flow and data-flow in event logs'. Together they form a unique fingerprint.

Cite this