Sripath Mishra

Sripath Mishra

CS PhD candidate at Purdue. Anomaly and intrusion detection for edge and distributed systems — explainable AI, low-latency inference, federated learning, retrieval, GNNs. Seeking Applied Scientist / ML Engineer roles, 2027 start.

Enriching Network Traffic Traces for Increasing Classification Accuracy

Published in IEEE International Performance, Computing, and Communications Conference (IPCCC), 2025

Traffic classifiers are usually evaluated on traces that mix many sources of variation at once, which makes it impossible to say which factor a given augmentation is compensating for. This work builds a testbed that isolates one network factor at a time — ten Dockerized services across five application classes, from video streaming and conferencing down to email and FTP — and uses it to measure augmentation effects per classifier family.

The result is that augmentation is method-specific: the identical augmentation that improves packet-size-sequence classifiers measurably degrades timestamp-based ones. “Does augmentation help?” is therefore the wrong question — the answer depends on which signal the classifier reads.

Authors: Sripath Mishra, Akhil Prasad, Sonia Fahmy
Venue: IEEE IPCCC 2025, Austin, TX

Paper on IEEE Xplore

Recommended citation: Sripath Mishra, Akhil Prasad, Sonia Fahmy. "Enriching Network Traffic Traces for Increasing Classification Accuracy." IEEE International Performance, Computing, and Communications Conference (IPCCC), Austin, TX, November 2025.
Download Paper