Anomaly Detection (AD)

Result Type

Scientific or Technological R&D Result including ICT Hardware (TRL 3-5)

Lead Developer

Software Imagination & Vision SRL (SIMAVI)


Network Traffic, Anomaly Detection

Anomaly Detection component is based on both machine learning algorithms and statistics-based algorithms. The AD thus performs a fast and efficient analysis of packets in a network and detects unknown attacks, in part due to the rapid development of malware.

Result Description

Anomaly detection uses the following types of algorithms:

k-means-clustering algorithm for analysing HTTP and DNS traffic

Statistical algorithms for issues like Alien accessing too many hosts, P2P communication, Media streaming client, C&C BotNet communication.

Target Business Sectors / EU Policy Areas

  • Business and industry
  • Energy
  • Public health

Contribution to UN Sustainable Development Goals