Paper title: Self-attentive deep learning method for online traffic classification and its interpretability
Accepted by Elsevier Computer Networks (https://doi.org/10.1016/j.comnet.2021.108267)
NetAI20 version https://github.com/xgr19/SAM-for-Traffic-Classification/tree/SAM-before-NetAI
Run the files as follow:
- python3 preprocess.py
- python3 tool.py
- python3 train.py
The dataset is available at http://mawi.wide.ad.jp/mawi/samplepoint-G/2020/202006101400.html
Other datasets (the code of this repo is reuseful): ISCX：https://www.unb.ca/cic/datasets/vpn.html UNIBS：http://netweb.ing.unibs.it/~ntw/tools/traces/
We have new repositories about DL algorithms on traffic classification: xgr19/Loong, xgr19/Soter, xgr19/SOTERIA, xgr19/Mousika! Especially, Mousika is about using P4 to reach nanosecond-level classification.