Unsupervised Dataset Cleaning Framework for Encrypted Traffic Classification

Kun Qiu 1 , Ying Wang 2 , Baoqian Li 2 , Wenjun Zhu 1 · 2025

Read Unsupervised Dataset Cleaning Framework for Encrypted Traffic Classification by Kun Qiu 1 , Ying Wang 2 , Baoqian Li 2 , Wenjun Zhu 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-00701.txt.

Description

Kun Qiu, with three co-authors, Ying Wang, Baoqian Li, and Wenjun Zhu, presents an unsupervised framework for cleaning encrypted traffic data. Classifying network flows into categories helps enterprise and carrier networks, but encryption defeats deep packet inspection, and machine learning methods need clean training data without useless flows.

Existing cleaning relies on manual inspection of every packet, which is costly and slow. Qiu and his co-authors automate the removal of irrelevant protocols, background activity, control messages, and long-lived sessions from encrypted mobile traffic, without labeled examples. On real-world datasets, their cleaning costs only about two to two and a half percent in classification accuracy compared with manual cleaning, making it an efficient preprocessing step.

Catalog record

CREATORKun Qiu 1 , Ying Wang 2 , Baoqian Li 2 , Wenjun Zhu 1
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE8 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-00701

Browse the full library · About the Tower of Babel