Statistical Learning Based Framework for Random Networks Knowledge Extraction Applied in Smart Cities

Abstract

Smart Cities are a future reality that emerged recently. They became a wide research field around the world. These cities will combine the power of ubiquitous communication networks and wireless sensors with the efficient management systems to solve daily challenges and create exciting services. In this work, we involve the power of artificial intelligence to solve one of the serious challenges in big cities. This concerns the traffic management and prediction. This work proposes a statistical model serving the analysis of a random graph that represents, in reality, roads on map. Using those models and collected data from sensors or human agents, we can extract useful hidden knowledge for the best decision making. To prove the reliability of the approach, a Monte Carlo simulation algorithm is designed and results confirms the added-value of the approach.