Improved Off-Line Intrusion Detection Using A Genetic Algorithm And RMI

This article proposes an optimization of using Genetic Algorithms for the Security Audit Trail Analysis Problem, which was proposed by L. Mé in 1995 and improved by Pedro A. Diaz-Gomez and Dean F. Hougen in 2005. This optimization consists in filtering the attacks. So, we classify attacks in “Certainly not existing attacks class”, “Certainly existing attacks class” and “Uncertainly existing attacks class”. The proposed idea is to divide the 3rd class to independent sub-problems easier to solve. We use also the remote method invocation (RMI) to reduce resolution time. The results are very significant: 0% false+, 0%false-, detection rate equal to 100%. We present also, a comparative study to confirm the given improvement.

@article{AHMIM2011,
  title     = {Improved Off-Line Intrusion Detection Using A Genetic Algorithm And RMI},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {2},
  number    = {1},
  year      = {2011},
  publisher = {The Science and Information Organization},
  author    = {Ahmed AHMIM and Nacira GHOUALMI and Noujoud KAHYA},
  doi       = {10.14569/IJACSA.2011.020116},
  url       = {https://doi.org/10.14569/IJACSA.2011.020116}
}