Revealing daily human activity pattern using process mining approach

Abstract

In the last few years, with the emergence of ambient assisted living, the study of human behavioral pattern took a wide interest from research communities around the world. In many literatures, pattern recognition was widely adopted approach to implements in human behavior study from computing perspective. Pattern recognition brings a promising results in terms of accuracy for modeling human behavior. But the problem with this approach is the complexity of knowledge representation which formulated in mathematical model. In turns, a correction by the experts is hardly conducted. In another hand, gathering a graphical insight is not a trivial task. This paper investigate the use of process mining technology to gives an alternative to such problems. Process mining is data-driven approach to infer a graphical representation of any kind of process. In terms of human behavior, process can be defined as sequences of activities performed by human on daily basis. From the conducted experiments process mining was shown a potential use to infer a human daily activity pattern in a graphical representation.