Andrea Burattin, Ph.D.

Assistant Professor

Toward an Anonymous Process Mining

Abstract

Process mining is a modern family of techniques applied to datasets generated from business processes run in organizations, in order to improve and obtain useful insights and performance measurements on the processes themselves (with clear societal and economical benefits). While these techniques are very promising in understanding business processes, their complete and efficient implementation inside the organizations is often not possible. Hence, in a way similar to what is done for most non core activities, and in particular for most ICT services, companies evaluate the possibility of outsourcing such task. However, the confidentiality of the dataset related to the business processes are often key assets for most of modern companies. Then, in order to avoid threats that might come from disclosing such information, most companies decide not to benefit from these process mining techniques.

In this work, we propose a possible approach toward a complete solution which allows outsourcing of Process Mining without thwarting the confidentiality of the dataset and processes. Furthermore, we provide a prototype implementation of our proposed approach and run several experiments that confirmed the feasibility of our approach. We believe the one highlighted in this paper is an important direction to work on, in order to remove the obstacles that prevent companies to fully benefit from outsourcing process mining.

Paper Information and Files

Published in Proceedings of Future Internet of Things and Cloud (FiCloud 2015); Rome, Italy; August 24-26 2015.

DOI: 10.1109/FiCloud.2015.9