Multiple Views Extraction from Semantic Trajectories.

The interest in exploiting crowd-sourced location information has recently emerged as it can bring many valuable benefits. This is particularly the case where multi-dimensional semantic information represent human trajectories and contextual information arising in indoor and outdoor spaces. Users have different interests when interpreting and analysing trajectories. While some users just want to visualise the data, others require either higher-level information or aggregated knowledge. This paper introduces a modelling approach and data manipulation mechanisms that extract from generic semantic trajectories multiple views at different levels of abstraction to produce hybrid spatial representation for mobility patterns. This approach considers a multi-layered graph representation for trajectories according to some given spatio-temporal, contextual and user-defined criteria. The approach has been experimented with real data and implemented within a graph database that illustrates its potential.