An effective convolutional neural network based on SMOTE and Gaussian mixture model for intrusion detection in imbalanced dataset

Keyphrases

• Binary Classification 33%
• CICIDS2017 66%
• Class Classification 33%
• Class Imbalance 100%
• Classification Algorithms 33%
• Classification Performance 33%
• Cluster-based 33%
• Convolution Kernel 33%
• Convolutional Neural Network 100%
• Detection Rate 100%
• Effective Solutions 33%
• Flow-based Intrusion Detection 33%
• Gaussian Mixture Model 100%
• Imbalance Processing 33%
• Imbalanced Data 100%
• Intrusion Datasets 33%
• Intrusion Detection 100%
• Intrusion Detection System 66%
• Key Security 33%
• Large-scale Dataset 33%
• Learning Rate 33%
• Malicious Behavior 33%
• Minority Class 66%
• Model Performance 33%
• Modern Network 33%
• Multi-class Classification 33%
• Network Intrusion Detection System 33%
• Neural Network 100%
• Novel Class 33%
• Oversampling Technique 100%
• Processing Methods 33%
• Processing Technology 33%
• Security Device 33%
• Synthetic Minority 100%
• Undersampling 33%
• UNSW-NB15 Dataset 66%

Computer Science

• Binary Classification 20%
• Class Classification 20%
• Class Imbalance 20%
• Classification Algorithm 20%
• Convolution Kernel 20%
• Convolutional Neural Network 100%
• Detection Method 20%
• Detection Rate 60%
• Effective Solution 20%
• Experimental Result 20%
• Gaussian Mixture Model 100%
• Intrusion Detection 100%
• Learning Rate 20%
• Malicious Activity 20%
• Minority Class 40%
• Multiclass Classification 20%
• Network Intrusion Detection System 20%
• Performance Model 20%
• Processing Method 20%
• Sampling Technique 100%
• Security Device 20%