Plug-in enhancement framework: Breaking through performance bottleneck of pre-trained models for encrypted traffic classification

Key Points

• Encrypted traffic classification achieves improved accuracy with enhanced pre-trained models and plug-in framework.
• A performance boost of 20% in classification accuracy demonstrates the effectiveness of the proposed approach.
• The assessment utilizes a novel enhancement framework for machine learning models to address performance bottlenecks.
• These findings support the need for advanced techniques in machine learning to optimize encrypted data handling.