RE: Suggestions

["(infor) urko zurutuza" <uzurutuza () eps mondragon edu>]

Hi all,

  As Steffano (il intrusso italiano) says, you must be very careful about that. There have been a lot of reserch 
projects involving IA and anomaly detection, so you should start reading amount of research papers. 

For example, some algorithms used:

- neural nets
- fuzzy logic
- inmune algorithms
- genetic algorithms
- decision trees
- data mining...
- and much more

A good starting point could be: 

http://www-rnks.informatik.tu-cottbus.de/en/security/idsbody.html

Anyway, I think your question on that research is a little bit strange, as ther is a good research group in Brazil, 
ACME! Computer Security Research (http://www.acme.ibilce.unesp.br/), using neural nets for discovering new attacks.

Un saludo,

__________________________________________________
MONDRAGON UNIBERTSITATEA
Urko Zurutuza
Dpto. Informática
Loramendi 4 - Aptdo.23
20500 Arrasate-Modragon
Tel. +34 943 739636 // +34 943 794700 Ext.297
www.eps.mondragon.edu
uzurutuza () eps mondragon edu

> -----Mensaje original-----
> De: Stefano Zanero [mailto:stefano.zanero () ieee org]
> Enviado el: miércoles, 26 de mayo de 2004 21:34
> Para: Thiago dos Santos Guzella; focus-ids () securityfocus com
> Asunto: Re: Suggestions
>
> Thiago dos Santos Guzella wrote:
>
> > Hi there,
> >
> > I am taking part in a research project on artificial inteligence, and my
> > objective is to create a IDS (possibly hybrid), capable of detecting
> attacks
> > never seeing before (by using some artificial inteligence algorithms).
>
> Thiago, do a GOOD search of the usual places (portal.acm.org and
> www.ieeexplore.ieee.org) before losing time in defining things already
> defined or proving things already proven.
>
> There's much room for improvement in the field, but the concept has
> already been drawn out many times and perhaps you don't need to reinvent
> all of the wheel...
>
> --
> Cordialmente,
> Stefano Zanero
>
> Computer World Italia (www.cwi.it)
>
>
> --------------------------------------------------------------------------
> -
>
> --------------------------------------------------------------------------
> -



---------------------------------------------------------------------------

---------------------------------------------------------------------------