Traffic prediction method based on graph transfer learning

Abstract

The invention belongs to the field of computer data analysis, and provides a traffic prediction method based on graph transfer learning, which can efficiently transfer knowledge from a source domain with sufficient training samples to a target domain with insufficient training samples, and improve the prediction performance of a traffic prediction model based on a graph convolutional neural network under the condition of small samples. The method provided by the invention can help to improve the prediction accuracy of a traffic prediction model based on graph data under the condition of small samples. According to the method, spatial clustering and time clustering regular terms are introduced, spatial-temporal pattern matching and prediction accuracy are balanced, and matching of node spatial-temporal pattern similarity in a target domain and a source domain is completed in a data driving mode, so that the effect of reducing negative migration is achieved, and the calculation overhead of the conventional method in the space-time mode matching process is greatly reduced.

Description

technical field [0001] The invention belongs to the field of computer data analysis, in particular to a method for transferring specific time and space knowledge across fields. Background technique [0002] Traffic forecasting is very important to the management of urban traffic. It can help the traffic management department to predict the traffic flow, driving speed and occupancy rate of urban roads, and conduct real-time guidance to reduce the occurrence of urban road congestion. In recent years, deep learning methods have been widely used in the field of traffic forecasting, and compared with traditional time series forecasting models, the accuracy of forecasting has been greatly improved. Currently, traffic prediction models based on deep learning mainly model the problem from two aspects, namely time dependence and space dependence. Time-dependent capture mainly uses recurrent neural network (RNN) or temporal convolutional network (TCN), while space-dependent capture m...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the technical problem that the existing traffic prediction method based on graph convolutional neural network has low prediction accuracy in the case of small samples, the present invention designs a traffic knowledge transfer method based on node spatio-temporal pattern clustering, which can learn from training Efficiently transfer knowledge from the source domain with sufficient samples to the target domain with insufficient training samples, and improve the prediction performance of the traffic prediction model based on graph convolutional neural network in the case of small samples

Images