Author(s)

Sen Wu

Corresponding Author

Sen Wu

ABSTRACT

The purpose of this article is to study the algorithm model of highway traffic flow prediction based on big data analysis, so as to provide accurate traffic flow prediction information and support the decision optimization of traffic management departments. In order to achieve this goal, this article adopts LSTM (Long-term and short-term memory network) as a forecasting model, and uses its advantage of capturing the long-term dependence of sequence data to forecast highway traffic flow. During the experiment, a representative traffic flow data set is selected, which is used for model training and testing after preprocessing and feature engineering. Through continuous training and optimization, we get a stable and accurate LSTM prediction model. The experimental results show that LSTM model has obvious advantages in highway traffic flow forecasting, which can capture the time series changes of traffic flow and has good forecasting ability for both short-term and long-term traffic flow changes. Especially in peak hours and complex road sections, the prediction accuracy of the model is relatively high.

KEYWORDS

Highway traffic flow forecast; Big data analysis; LSTM model; Accuracy of prediction; Time series change