Traffic flow prediction method based on feature reconstruction error

Abstract

The invention discloses a traffic flow prediction method based on a feature reconstruction error, and belongs to the technical field of machine learning. The method comprises the steps of: (1) selecting a target machine learning network, and initializing the parameters of the target machine learning network; (2) constructing a training data set of traffic flow, and initializing parameters of a feature correction weight matrix; (3) training the feature correction weight matrix by using the training data set, and using a random gradient descent algorithm and a feature reconstruction error loss function in the training process; (4) fixing feature correction weight matrix parameters, training a target machine learning network, and using a random gradient descent algorithm in the training process; (5) repeating the step (3) and the step (4) until the loss function converges or reaches a maximum training step number; and (6) after the training is finished, inputting the traffic flow data tobe predicted into the trained network model to obtain the predicted traffic flow. According to the invention, the stability of the model in traffic flow prediction can be enhanced.

Description

technical field [0001] The invention belongs to the technical field of machine learning, and in particular relates to a traffic flow prediction method based on feature reconstruction errors. Background technique [0002] The training process of the traffic flow prediction model based on neural network requires that the data samples follow the independent and identical distribution assumption (i.i.d assumption), that is, the training and testing data sets are sampled from the same data distribution. Under the assumption of i.i.d., the trained model can be directly applied to the test data set, and can obtain the same effect as the training data set. Although this approach has proven to be very successful in many research public datasets, it is flawed in practical applications. The reason is the data selection bias that is ubiquitous in practical applications, that is, it is impossible to guarantee that the training or testing samples are completely randomly sampled. Therefo...

AI Technical Summary

Problems solved by technology

These works by changing the weight of training samples need to maintain a weight matrix with the same size as the number of samples, which makes its calculation and space complexity in large-scale industrial applications very large. Therefore, these works are important in industrial applications of traffic flow forecasting. less feasible

Images