A multi-level approach for forecasting critical events in smart cities

Nowadays, the development of new technological innovations and the concept of the Smart City leads us to new approaches to improve the quality of life. The aim of this paper is to investigate the possibility of providing a multi-level approach capable of monitoring critical events. A system based on this methodology, able to gain the behaviors of people when a critical event occurs, may be fundamental to decision-makers and the authorities for rapid response and monitoring during such scenarios. Typically, it can be possible modelling the context according to various data, such as meteorological, traffic, social ones and many others, in light of this will be introduced a methodology which tries to merge Context Aware techniques with probabilistic approach based on Bayesian Network. In this paper, the ability of this methodology, capable of providing performances comparable to standard forecasting systems, are shown by applying it to a case of study in the area of London.