Traffic accident prediction method based on space-time diagram convolutional network

Abstract

The invention relates to a traffic accident prediction method based on a space-time diagram convolutional network, which mainly combines the diagram convolutional network and a long-short-term memory network to respectively obtain space information feature extraction and time sequence information feature extraction of an actual traffic road network, fully considers traffic information data such as time, weather and interest points and constructs a topological structure of a road network by adopting a region division method; firstly, a regional road network is constructed into a graph structure, a graph convolutional network is used for learning a complex road network topological structure to obtain spatial correlation in a traffic state, and then a long-short-term memory network is used for learning dynamic changes of traffic accident data to obtain temporal correlation in the traffic state; and finally, the two networks are combined to construct a prediction model, and the traffic accident is predicted based on the prediction model so as to improve the prediction performance of the traffic accident.

Description

Technical field [0001] The present invention relates to the field of data prediction techniques, and more particularly to a traffic accident prediction method based on a time-space graph convolutionary network. Background technique [0002] Road Traffic Safety is an important area of ​​common concern in the world. According to the World Health Organization's "Global Road Safety Status Report", the world is about 1350,000 people die in traffic accidents every year, and the economy is rapid Development, the situation of road traffic safety has become more serious, so raising the predictive performance of traffic accidents is an urgent and important research task, which helps provide timely warning to road traffic abnormalities, reduce accident hazards. , Reduce accident loss. [0003] Some people apply the Bayesian network and the logistic regression model to traffic accident prediction research; some people use the method of genetic algorithm, mode search and artificial neural net...

AI Technical Summary

Problems solved by technology

However, convolutional neural networks are usually used in Euclidean spaces, such as images, regular grids, etc., such models generally cannot work in the context of urban road networks with complex topologies, and thus cannot describe spatial dependencies in nature

Images