This paper, for the purpose of meeting challenges of fewer resources of storage and calculation in the detection of ICS intrusion as well as real-time requirements, has particularly designed an online hybrid kernel le...
详细信息
ISBN:
(纸本)9789811996962;9789811996979
This paper, for the purpose of meeting challenges of fewer resources of storage and calculation in the detection of ICS intrusion as well as real-time requirements, has particularly designed an online hybrid kernel learning machine with dynamic forgetting mechanism. First, on the basis of online kernel limit learning machine, a dynamic forgetting mechanism is designed to dynamically adjust the amount of forgetting data according to the current block error, which reduces the system burden and improves the detection accuracy. Then, it replaces the former single kernel function with a hybrid kernel function, which successfully advances the accuracy rate and generalized performance. Finally, a hybrid noise-reducing autoencoder is created to perform dimensional reduction of industrial data with huge dimensions, resulting in the improvement of algorithm and efficiency. The validity and superiority of the proposed online hybrid kernel learning machine with dynamic forgetting mechanism are verified through simulation experiments.
Network security problems bring many imperceptible threats to the integrity of data and the reliability of device services,so proposing a network intrusion detection model with high reliability is of great research si...
详细信息
Network security problems bring many imperceptible threats to the integrity of data and the reliability of device services,so proposing a network intrusion detection model with high reliability is of great research significance for network *** to the strong generalization of invalid features during training process,it is more difficult for single autoencoder intrusion detection model to obtain effective results.A network intrusion detection model based on the Ensemble of Denoising Adversarial autoencoder(EDAAE)was proposed,which had higher accuracy and reliability compared to the traditional anomaly detection *** the adversarial learning idea of Adversarial autoencoder(AAE),the discriminator module was added to the original model,and the encoder part was used as the *** distribution of the hidden space of the data generated by the encoder matched with the distribution of the original *** generalization of the model to the invalid features was also reduced to improve the detection *** the same time,the denoising autoencoder and integrated operation was introduced to prevent overfitting in the adversarial learning *** on the CICIDS2018 traffic dataset showed that the proposed intrusion detection model achieves an Accuracy of 95.23%,which out performs traditional self-encoders and other existing intrusion detection models methods in terms of overall performance.
暂无评论