Research On Unbalanced Traffic Intrusion Detection Technology Based On Deep Learning

With the exponential growth of global computer network equipment,network threats continue to rise,and ensuring the security of the network environment has become an urgent problem to be solved.The existing network intrusion detection model does not fully learn timing features,resulting in low accuracy.Moreover,there are malicious traffic with low attack frequency in the intrusion detection data set.If it cannot be intercepted in time,it will cause immeasurable losses to network equipment and data.Therefore,the improvement of network intrusion detection model and the processing of unbalanced data sets have become the research hotspots of network intrusion detection.The following are the primary research and advancements of this paper:(1)Aiming at the problems of low classification accuracy of intrusion malicious traffic and incomplete timing feature learning of current network intrusion detection models,a network intrusion detection model CNN-BIGRU-Attention is proposed that integrates CNNBIGRU(bidirectional gated recurrent unit)and attention mechanism.The network intrusion detection model was tested on the CIC-IDS-2017 dataset.The experimental results showed that the overall prediction accuracy of malicious traffic types reached 98.25%.The(F1-Score)value has a certain improvement compared with the comparison model.(2)Aiming at the problem that the current network intrusion detection model has a low recognition rate for malicious traffic types with a small number of samples in the unbalanced data set,a CNN-BIGRU-Attention intrusion detection model integrating SMOTE-ENN mixed sampling is proposed,and the experimental results It shows that the model helps to identify malicious traffic types with a small sample size,and the overall prediction accuracy rate reaches99.29%.The network intrusion detection model significantly improves the detection rate.(3)Using the CNN-BIGRU-Attention intrusion detection model fused with SMOTE-ENN mixed sampling proposed in this paper,a network intrusion detection system based on deep learning is designed and implemented,and the intrusion detection results are visualized.