The core of smart city is to build intelligent transportation system.. An intelligent transportation system can analyze the traffic data with time and space characteristics in the city and acquire rich and valuable knowledge, and it is of great significance to realize intelligent traffic scheduling and urban planning. This article specifically introduces the extensive application of urban transportation infrastructure data in the construction and development of smart cities. This article first explains the related concepts of big data and intelligent transportation systems and uses big data to illustrate the operation of intelligent transportation systems in the construction of smart cities. Based on the machine learning and deep learning method, this paper is aimed at the passenger flow and traffic flow in the smart city transportation system. This paper deeply excavates the time, space, and other hidden features. In this paper, the traffic volume of the random sections in the city is predicted by using the graph convolutional neural network (GCNN) model, and the data are compared with the other five models (VAR, FNN, GCGRU, STGCN, and DGCNN). The experimental results show that compared with the other 4 models, the GCNN model has an increase of 8% to 10% accuracy and 15% fault tolerance. In forecasting morning and evening peak traffic flow, the accuracy of the GCNN model is higher than that of other models, and its trend is basically consistent with the actual traffic volume, the predicted results can reflect the actual traffic flow data well. Aimed at the application of intelligent transportation in an intelligent city, this paper proposes a machine learning prediction model based on big data, and this is of great significance for studying the mechanical learning of such problems. Therefore, the research of this paper has a good implementation prospect and academic value.
Heterogeneous Data Aggregation Schemes to Determine Traffic Flow Parameters in Regional Intelligent Transportation Systems
