Social network analysis and role hierarchy mining through MXML-based event logs

Abstract

This paper is divided into three main parts. In the first part of the study, we initially collected a process event log -from an information system -during an Academic Writing (English) training course in Thailand. The original data was consisted of 330 process instances and 3,326 events in total. Being aware of the fact that ProM process mining framework accepts and supports only XES and MXML logs, the data was subsequently configured and converted into the fitting format. In the second part of the study, we applied role hierarchy mining technique in order to evaluate the role of every individual throughout the course/training program. Moreover, the organizational mining technique enabled us to better understand the functional structure of the training organization (based on models) and ultimately improve (or benchmark) the underlying processes. In the thirst part of the study, we used social network mining as a pre-requisite for social network analysis to discover the interaction patterns between the originators during the teaching process. Accordingly, the handover of work metric -of the social network mining technique-made possible to not only investigate direct relationships but also indirect relationships as well.