An analysis of clustering objectives for feature selection applied to encrypted traffic identification

Keyphrases

• Feature Selection 100%
• Encrypted Traffic Classification 100%
• Multi-objective Genetic Algorithm 100%
• Logarithmic Transformation 50%
• Flow-based 25%
• Detection Rate 25%
• Selected Features 25%
• False Positive Rate 25%
• Number of Clusters 25%
• Performance Gain 25%
• Data Prior 25%
• Feature Clustering 25%

Computer Science

• multi-objective evolutionary algorithm 100%
• Feature Selection 100%
• Gold Standard 50%
• False Positive Rate 25%
• Detection Rate 25%
• Performance Gain 25%

Engineering

• Feature Extraction 100%
• Multiobjective Genetic Algorithm 100%