Observatorio de I+D+i UPM

Descripción
In this work we have presented a genetic algorithm approach for implementing rules deployed to detect various types of attacks. We have used the features identified as the most important ones for each attack type to form the rules in order to get a high detection rate. These features were identified by deploying principal component analysis and multi expression programming. Thenceforth we have used serial combination of the rules implemented for detection of user-to-root attacks with the rules implemented for detection of other types of attacks in order to decrease the false-positive rate. The model was verified on KDD99 intrusion detection dataset, demonstrating higher detection rates than those reported by the stateof-the-art, while using small subset of features from the original set.
Internacional	Si
Nombre congreso	International Conference on Emerging Security Information, Systems and Technologies, Securware'07
Tipo de participación	960
Lugar del congreso	Valencia (España)
Revisores	Si
ISBN o ISSN	0-7695-2989-5
DOI
Fecha inicio congreso	14/10/2007
Fecha fin congreso	20/10/2007
Desde la página
Hasta la página
Título de las actas

Esta actividad pertenece a memorias de investigación

• Autor: Atta Baddi
• Autor: Zorana Bankovic . (UPM)
• Autor: Octavio Nieto-Taladriz Garcia (UPM)
• Autor: Slobodan Bojanic Antonijevic (UPM)

• Creador: Grupo de Investigación: Laboratorio de Sistemas Integrados (LSI)
• Departamento: Ingeniería Electrónica