CRACS starts developing new research project with IGC

A group of researchers from CRACS - INESC Porto LA (Centre for Research in Advanced Computing Systems) has recently started developing a new project together with the theoretical immunology group from the Gulbenkian Institute for Science (IGC).

The project's goal is to create a system for computer networks which detects intruders. For that, this system will use a learning model inspired by the vertebrate animals' immune system.

This model is based on a recent theory concerning the functioning of the immune system. It suggests that the cells' activation (or response) starting point is dynamically adjusted throughout time, taking in consideration the history of its iterations with other cells which transport foreign agents into our body.

This is the stage in which there is a conclusion of an algorithm simulator. At this stage it is possible to take conclusions about the system's dynamic behaviour throughout time, including about the evolution of the cell's activation starting point when exposed to foreign agents (not necessarily "dangerous").

Doing this, the intention is to simulate the system using data from the network traffic so that its classification potentialities in normal or non-normal traffic may be explored.

Manuel Eduardo Correia and Mário Antunes are the researchers from CRACS working on this project.