Intrusion detection system using optimum-path forest

• Autor(a) principal: Pereira, Clayton [UNESP]
• Data de Publicação: 2011
• Outros Autores: Nakamura, Rodrigo [UNESP], Papa, João Paulo [UNESP], Costa, Kelton
• Tipo de documento: Artigo de conferência
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1109/LCN.2011.6115182; http://hdl.handle.net/11449/72855
• topic: Artificial intelligence techniques; Data sets; Intrusion Detection Systems; Pattern classifier; Pattern recognition techniques; Real time; Training patterns; Computer crime; Forestry; Neural networks; Pattern recognition; Telecommunication networks; Intrusion detection; Algorithms; Artificial Intelligence; Neural Networks; Pattern Recognition; Telecommunications

Intrusion detection systems that make use of artificial intelligence techniques in order to improve effectiveness have been actively pursued in the last decade. Neural networks and Support Vector Machines have been also extensively applied to this task. However, their complexity to learn new attacks has become very expensive, making them inviable for a real time retraining. In this research, we introduce a new pattern classifier named Optimum-Path Forest (OPF) to this task, which has demonstrated to be similar to the state-of-the-art pattern recognition techniques, but extremely more efficient for training patterns. Experiments on public datasets showed that OPF classifier may be a suitable tool to detect intrusions on computer networks, as well as allow the algorithm to learn new attacks faster than the other techniques. © 2011 IEEE.