Analysis of the Contextual Behaviour of Mobile Subscribers

In this paper, contextual behavior of mobile subscribers is studied with data collected straight from smartphones. The paper develops an approach to study how people use mobile devices in different contexts, by proposing an algorithm that works with device-based sensor data. This approach consists of context detection and data analysis. The context detection algorithm analyses cellular network radio logs in modeling the location of people. This paper then analyses usage patterns over different contexts. Demonstration of the contextual modeling with a sample of Finnish smartphone users proves that the applications of the approach are numerous.