Design and Implementation of Big Data Forecasting System Based on Intelligent Transportation

The rapid increase of the number of vehicles worldwide has led to serious traffic congestion and long waiting time for traffic lights, which has become a major difficulty for all major cities. As a new generation of technical means to solve the above issues, intelligent transportation provides new possibilities for traffic system dispatching. This paper designs a big data prediction system based on intelligent transportation, with which a user-friendly UI interface is implemented based on Python and the Tkinter Python library. Specifically, we introduce PyMySql library to implement the Python user UI interface connection to the database and MySQL database to realize the massive traffic data storage and management and Pandas, Numpy, Sklearn, Tensorflow and Keras are utilized to build traffic data prediction model. In order to solve the critical issues such as incomplete data and network topology structure, the system is designed to support three different modes, namely no sequence data forecasting model, single sequence data forecast model and road network topological structure prediction model.