Network fault model training method and device, network fault diagnosis and prediction method and device and electronic equipment

Abstract

The embodiment of the invention relates to the field of network fault detection, and discloses a network fault model training method and device, a network fault diagnosis and prediction method and device and electronic equipment. The method comprises the following steps: acquiring equipment data in a target network and user behavior data in a network using process of a user; acquiring user complaint data and user experience data in the process of using the target network by the user; taking the equipment data and the user behavior data for marking the network faults as training samples, and performing classification training on the network faults by adopting a GBDT model to obtain a GBDT network fault classification model; taking the user complaint data and the user experience data as training samples, and performing prediction training on the network fault by adopting a DNN model to obtain a DNN network fault prediction model; and taking output results of the GBDT network fault classification model and the DNN network fault prediction model as training samples, and performing training by adopting an FM model to obtain an FM network fault diagnosis prediction model. Through the constructed model, the network fault can be accurately and efficiently diagnosed and predicted.

Description

Technical field [0001] The embodiments of the present invention relate to the field of network fault detection, and in particular to a method, device, and electronic equipment for network fault model training, diagnosis and prediction. Background technique [0002] Network fault diagnosis and prediction is to use technology to determine whether the network will fail and predict the probability of a certain type of failure. Currently, there are two popular active network fault diagnosis and prediction technologies. One is to use Bayesian network model to diagnose network faults. This method relies on the network topology and the alarm information of each node. Another active network fault prediction method is based on the gradient boosting tree model (GBDT). This method first collects network failure symptom data and failure data, and then uses a gradient boosting tree classifier for classification training, and predicts network failures based on the classifier. [0003] However,...

AI Technical Summary

Problems solved by technology

[0004] First of all, the existing network fault prediction models usually only use fault alarm data and fault type data, without considering the impact of different network usage habits of users on network quality, and the impact of different quality requirements on user satisfaction, so that the prediction results are accurate low degree

[0005] Secondly, the Bayesian network model requires a comprehensive understanding of the network topology and finds out all network links for each user, which is not efficient

Moreover, this method relies on equipment alarm information, and can only determine whether a network failure occurs after some equipment sends out a failure alarm, and cannot issue an alarm before a network failure occurs

[0006] Again, the use of gradient boosting tree classifiers is prone to overfitting, and the model cannot predict cases outside the sample

Images