Method and device for predicting network user behavior

Abstract

The invention discloses a method and device for predicting network user behavior. The method comprises the following steps: determining a corresponding parameter set of a prediction model; determining a network user behavior training data set; establishing a hierarchical structure of a neural network prediction model according to the determined parameter set, and establishing an input weight matrix, a feedback weight matrix and a dynamic pool internal connection weight matrix; and training the prediction model by virtue of the training data set, calculating the output weight matrix of the prediction model, adjusting the input of the prediction mode, and finally providing a corresponding prediction result by the prediction model.

Description

technical field [0001] The invention relates to the field of communication network business management and monitoring, in particular to a method and device for network user behavior prediction. Background technique [0002] The network user behavior prediction method in the prior art uses a method based on BP (Back Propagation) neural network to predict user behavior. However, the training method of this BP neural network method is relatively complicated, which is essentially a gradient descent method, and the algorithm is easy to fall into local extremum, and the efficiency is not high. There is a contradiction between the predictive ability (generalization ability) and the training ability (approximation ability) of the BP network, and "overfitting phenomenon" will appear. Therefore, this user behavior prediction method based on BP neural network cannot fully learn the law of network user behavior distribution, and there is a certain one-sidedness in network user behavior...

AI Technical Summary

Problems solved by technology

Therefore, this user behavior prediction method based on BP neural network cannot fully learn the law of network user behavior distribution, and there is a certain one-sidedness in network user behavior prediction.

On the other hand, the performance indicators of the learning algorithm of the BP neural network are not as good as the echo state neural network prediction method based on the complex network, and are not suitable for direct application in the prediction of network user behavior.

Images