The spread of mobile devices with several sensors, together with mobile communication, provides huge volumes of real-time data (big data) about users’ mobility habits, which should be correctly analysed to extract useful knowledge. In our research we explore a data mining approach based on a Naïve Bayes (NB) classifier applied to different sources of big data. To achieve this goal, we propose a methodology based on four processes that collects data and merges different data sources into pre-defined data classes. We can apply this methodology to different big data sources and extract a diversity of knowledge that can be applied to the development of dedicated applications and decision processes in the area of intelligent transportation systems, such as route advice, CO2 emissions reduction through fuel savings, and provision of smart advice for public transportation usage.

--

Big data,Data mining,Naive bayes,Mobile device,Sensor information

• Physical Sciences - Natural Sciences