Short-term Traffic Flow Prediction Based on Deep Learning

Abstract

The accuracy of short-term traffic flow prediction is affected by two factors: one is the accuracy of data collection, and another is model selection. During data collection, aiming at the high cost and low precision of the traditional use of fixed equipment to collect traffic flow data, this paper proposes an algorithm model combining convolutional neural network and support vector regression, and designs an input matrix considering the influence degree of the road segment. The example is proved that the proposed algorithm model is better than the ARIMA and SVR model, and it is an effective traffic flow prediction method.