Artificial Neural Network for Intrusion Detection
         Lixin Wang

Synopsis
An intrusion detection system mainly consists of two parts: anomaly detection 
and misuse detection. Anomaly detection is based on the assumption that 
observed activities deviate from expected normal system use. Misuse detection 
compares data against predefined patterns usually collected by an IDS signature 
database. One of the main shortcomings of misuse detection is that future 
attacks cannot be predicted or detected without predefining them in the IDS 
signature database. In this paper, we will focus on the study of the 
application of Neural Networks in the areas on misuse detection in order to 
detect known attacks and even variations of known attacks.
Reference:
[1] A study in using neural networks for anomaly and misuse detection. Anup K. 
Ghosh & Aaron Schwartzbard
[2] Artifical Neural Networks for Misuse Detection. James Cannady 
[3] A Survey and Analysis of Neural Network Approaches to Intrusion Detection. 
Hussam O. Mousa