Department of Computer, Control and Management Engineering

During the last years, a number of studies have experimented with applying process discovery techniques with the goal of automatically modelling human routines as if they were business processes. However, while promising results have already been achieved, applying process-oriented techniques to smart spaces data comes with its own set of challenges, due to the nature of smart spaces data and characteristics of human behaviour. This paper surveys existing approaches that apply process discovery to smart spaces data and analyses how they deal with the following challenges identified in the literature: choosing a suitable modelling formalism for human behaviour; bridging the abstraction gap between sensor and event logs; segmenting logs in traces; handling multi-user environments; and addressing human behaviour evolution. The main contribution of this article is two-fold: (i) providing the research community with an analysis of the existing applications of process discovery to smart spaces and how they address the above challenges, and (ii) assisting further research efforts by outlining opportunities for future work.