Process-mining based dynamic software architecture reconstruction

Abstract

Dynamic architecture reconstruction approaches aim to reconstruct the run-time architecture of a software system. Process mining is an emerging field combining process analytics and data science techniques. In this paper, we present an approach that creates interactive architecture visualizations without requiring any knowledge of the source code of the system under study. The approach is implemented in the tool AJPOLog. with two case studies, we show that this approach creates reliable results with relatively little effort. Though the studies also show that more research is needed to apply process mining techniques in the field of architecture reconstruction.