Skip to Main content Skip to Navigation
Conference papers

port2dist: Semantic Port Distances for Network Analytics

Abstract : Traffic analysis is a predominant task to support multiple types of management operations. When shifting from manually built signatures to machine learning techniques, a problem resides in the model to represent traffic features. The most notable examples are the TCP and UDP ports, near port numbers in the numerical space is not representative of a close semantic from an operational point of view. We have thus developed a technique to learn meaningful metrics between ports from scanning strategies followed by attackers. In this demonstration, we propose the port2dist tool, allowing to get, seek and retrieve semantic dissimilarities between port numbers.
Document type :
Conference papers
Complete list of metadata

Cited literature [6 references]  Display  Hide  Download

https://hal.inria.fr/hal-02345491
Contributor : Jérôme François <>
Submitted on : Monday, November 4, 2019 - 3:19:54 PM
Last modification on : Monday, November 30, 2020 - 10:26:03 PM
Long-term archiving on: : Wednesday, February 5, 2020 - 9:30:55 PM

File

im2019.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02345491, version 1

Collections

Citation

Laurent Evrard, Jérôme François, Jean-Noël Colin, Frédéric Beck. port2dist: Semantic Port Distances for Network Analytics. IM 2019 - The 16th IFIP/IEEE Symposium on Integrated Network and Service Management - Demo session, Apr 2019, Washington DC, United States. ⟨hal-02345491⟩

Share

Metrics

Record views

80

Files downloads

436