Big data for smart cities with KNIME a real experience in the SmartSantander testbed

Abstract

Summary Big data techniques are conceived as the powerful tool to exploit all the potential of the Internet of Things and the smart cities. A new dimension of understanding about the human behaviours is expected to be reached through all the gathered data in the emerging smart environment. The described potential, so‐called Human Dynamics, pursues to describe in real‐time the human behaviours and activities. This work presents our experiences for big data analytics in smart cities, in terms of sensors data management, data fusion and knowledge discovery from the data. The data used is from the European Project SmartSantander, where the traffic behaviour has been correlated with respect to the temperature in the Santander City. The evolution of both flows present a similar behaviour, in detail, a fine grain correlation is discovered. On the one hand, the traffic distribution, aggregated by temperature bins, follows up a Poisson distribution model. The Poisson modelling allows to interpolate and predict complex behaviours based on simple measures such as the temperature. At the same time, on the other hand, the isolated traffic density distribution, without taking into account the temperature‐based aggregation has been analysed. The traffic distribution has presented a burst behaviour, which presents a closer model to the human dynamics. Therefore, this work presents as the smart cities data can be modelled as Poisson or Human Dynamics (burst models). Finally, reference data analytics process, data sets and models are offered for the Open Source Data analytics platform Konstanz Information Miner (KNIME). Copyright © 2014 John Wiley & Sons, Ltd.