In the new study by the Malware Lab we present NetFlowMeter, a flow analyser developed entirely by one of our Artificial Intelligence teams during their research activities.
Designed to process raw network traffic (pcap files) and extract a set of features for use with machine learning algorithms, the software is now available as an open source project on Tinexta Defence's GitHub account, which we officially launch.
The software was created as an advanced reimplementation of CICFlowMeterone of the most widely used tools in the scientific community, with the aim of optimising its performance and correcting calculation and labelling bugs that can compromise the quality of the datasets produced.
Validation of the tool was conducted using both public datasets such as CICIDS2017, commonly used in academia as a benchmark to test machine learning algorithms applied to network intrusion detection (IDS), both proprietary datasets generated through automatic pipelines developed by our team, fed by real traffic collected from networked exposed systems.
The release of this software is part of Tinexta Defence's long-term strategy to foster transparency, interoperability and collaboration in the scientific community by strengthening investment in research and development of cybersecurity applications and tools.
If you wish to learn more, here is the link to our comprehensive study.
In addition, you can subscribe to the specific mailing list Cyber Studios by Tinexta Defence, to receive updates on upcoming research:
