ANTA: Machine Learning vs. Deep Learning for Network Traffic Analysis
ANTA (Accelerated Network Traffic Analytics) was ADAL’s SCARP 2020 project, A Performance Study between Machine Learning and Deep Learning Methods on Network Traffic Analysis.
The team studied the problem of detecting TLS-encrypted malware using flow metadata and TLS features. They benchmarked a suite of widely used machine learning and deep learning models on two malware flow datasets, measuring classification accuracy as well as run-time performance (throughput, CPU and RAM utilization), and accelerated the detectors with Intel DAAL and OpenVINO.
Team
- Matthew Grohotolski and Connor DiLeo, student researchers
- Dr. Peilong Li, faculty mentor
- Collaborators: Dr. Yan Luo and Onur Barut (University of Massachusetts Lowell) and Tong Zhang (Intel)