Accelerating Process Mining using Relational Databases

Given the abundance of event data, the challenge of process mining today is to enable process mining in the large. This research aims to address scalability problem in terms of memory use and time consumption. To this end, we use relational databases as the framework to both store event data and do process mining analysis. We conduct a pre-computation of intermediate structures during insertion time of the data. Finally, we implement the existing process mining algorithms to be compatible with relational database settings.