Performance Tradeoff in ML-Based Intrusion Detection Systems: Efficacy vs. Resource Usage
- Resource Type
- Conference
- Authors
- Dai, Nathan; Uludag, Suleyman
- Source
- 2024 IEEE 21st Consumer Communications & Networking Conference (CCNC) Consumer Communications & Networking Conference (CCNC), 2024 IEEE 21st. :1030-1031 Jan, 2024
- Subject
- Communication, Networking and Broadcast Technologies
Computing and Processing
Robotics and Control Systems
Performance evaluation
Analytical models
Intrusion detection
Machine learning
Network security
Internet of Things
Decision trees
Resource consumption
IoT
Cybersecurity
- Language
- ISSN
- 2331-9860
The rise in cyber attacks demands improved network security for resource-constrained IoT devices. We assess machine learning models on the CICIDS-2017 dataset, emphasizing the need for a balance between performance and resource efficiency. Principal component analysis (PCA) aids in creating efficient models, with XGBoost and Decision Tree as suitable options, providing decent performance without excessive resource requirements. Rebalancing techniques can enhance recall but may lower precision.