English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Encrypted traffic classification requires discriminative and robust trafficrepresentation captured from content-invisible and imbalanced traffic data foraccurate classification, which is challenging but indispensable to achievenetwork security and network management. The major limitation of existingsolutions is that they highly rely on the deep features, which are overlydependent on data size and hard to generalize on unseen data. How to leveragethe open-domain unlabeled traffic data to learn representation with stronggeneralization ability remains a key challenge. In this paper,we propose a newtraffic representation model called Encrypted Traffic Bidirectional EncoderRepresentations from Transformer (ET-BERT), which pre-trains deepcontextualized datagram-level representation from large-scale unlabeled data.The pre-trained model can be fine-tuned on a small number of task-specificlabeled data and achieves state-of-the-art performance across five encryptedtraffic classification tasks, remarkably pushing the F1 of ISCX-Tor to 99.2%(4.4% absolute improvement), ISCX-VPN-Service to 98.9% (5.2% absoluteimprovement), Cross-Platform (Android) to 92.5% (5.4% absolute improvement),CSTNET-TLS 1.3 to 97.4% (10.0% absolute improvement). Notably, we provideexplanation of the empirically powerful pre-training model by analyzing therandomness of ciphers. It gives us insights in understanding the boundary ofclassification ability over encrypted traffic. The code is available at:https://github.com/linwhitehat/ET-BERT.

ET-BERT: A Contextualized Datagram Representation with Pre-training Transformers for Encrypted Traffic Classification