"Interval rule matrix and Taylor methods for network intrusion detection"

Chenyi Hu
Computer Science Department
University of Central Arkansas

Knowledge discovery and acquisition have been a very active area in
both scientific research and real word applications. Recently, we have
built an interval valued fuzzy rule matrix model for decision making
systems; and applied it for network intrusion detection systems (IDS)
with mining network state database. In that model, a decision is
selected according to a predefined threshold and an index that obtained
from historical data. In real time IDS, one need to consider the
evolution of the rules. We believe that Taylor methods can play an
important role for this purpose, and will discuss a few basic ideas.

References:
1.    A. de Korvin, C. Hu, and P. Chen, Generating and Applying Rule
for Interval Valued Fuzzy Observations, Lecture Notes in Computer
Science, Vol. 3177, pp. 279-284, Springer-Verlag, 2004
2.    Q. Duan, C. Hu, and H. Wei, Enhancing Network Intrusion
Detection Systems with Interval Methods, the 20th ACM Symposium on
Applied Computing, to appear, 2005