Online Network traffic anomaly detection method combining OS-ELM and SADE

Network traffic anomaly detection methods can detect that is significantly different from normal by analyzing network traffic, and are seen as an effective means to unknown new attacks because they do not rely on static feature codes. However, most of the current have low accuracy high false alarm rate. In this paper, OS-ELM (Online Sequential Extreme Learning Machine) method proposed. First, we use SMOTE balance data samples improve classification performance. order reduce dimensionality between features eliminate redundancy, Stacked Denoising Autoencoder (SDAE) adopt dimension vectors. Finally, test real analyze effect model structure external noise performance, experimental results verify correctness our scheme. Compared with other based reconstruction, proposed has higher better