A Multilevel Graph Representation for Big Data Interpretation in Real Scenarios

Nowadays with the increase of smart objects, which make our world ever smart, it can be possible to observe a rapidly growing up of a large amount of data produced from a various sources. Even if there are numerous approaches, automatic and manual, present in the literature that try to interpret data by extracting information, these data become overwhelmed with a mass of information that is difficult to understand. In this context, we have to analyse and understand the data in order to have a new knowledge starting from this information. These data, if correctly managed, could help us for Big Data analysis and it has helpful for Smart City application. The main aim of this paper is to provide an approach for data interpretation, which take advance of three graphs (Ontologies, Context Dimension Tree and Bayesian Networks). This approach, through graph approaches abovementioned, is able to represent the real scenario both from the point of view of the sensors involved and of the services and events connected to the data.