Elman short-term traffic flow prediction model based on association rules

摘要

The traditional Elman neural network has the problem of easily falling into a local minimum or non-convergence in short-term traffic flow prediction, which leads to lower accuracy of traffic flow prediction. The traffic flow shows a clear periodic pattern in the time series. Based on this, time series factors are introduced, and an Elman short-term traffic flow prediction model based on association rule modification is proposed. The variable self-connection feedback gain factor is used to improve the prediction model in real time. The model uses the Apriori algorithm to mine association rules for traffic flow and time series factors, and corrects and compensates traffic flow prediction values according to the rules. At the same time, it reduces the size of the neural network solution space and breaks local convergence. The experimental results show that the prediction effect of the proposed prediction model is superior to the traditional model, and it is verified that the introduction of time series factors combined with association rules can reduce the error of traffic flow prediction.