A Review on Traffic Prediction Methods for Intelligent Transportation System in Smart Cities

Traffic flow prediction is one of the key problems in traffic control and guidance system as well as the important functions of intelligent transportation system (ITS). The fast expansion in machine learning new methods and in the appearance of new data sources makes it possible to evaluate and forecast traffic conditions in smart cities more quickly and accurately. Traffic estimation and prediction system has the ability to reduce traffic congestion and improve road capacity effectively. In this paper, the existing traffic prediction methods for smart cities are provided in detail and the problems and challenges of the prediction models are analyzed in depth. Based on the analysis of the existing short-term traffic flow forecasting methods, the possible development trend of short-term traffic flow predicting approaches in the future is pointed out.