Reproduce “From Ethernet to Wi-Fi: Flow Feature Selection and Construction for Traffic Classification” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Francisco Rau, Petia Georgieva, Carlos Herranz, Iñaki Val, Joaquin Perez
The authors select and construct flow features to improve Ethernet-trained application classification on Wi-Fi without training the classifier on Wi-Fi. Using labelled Wi-Fi development data for feature design, they reduce 29 inputs to seven. On held-out simulated Wi-Fi scenario S7, XGBoost accuracy increases from 88.44% ± 0.27 to 99.92% ± 0.18; this result is limited to the tested simulation settings.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.