Information security technology for computer networks through classification of cyber-attacks using soft computing algorithms

The Internet is the global platform which revolutionized the computer and communications domain. Although it becomes one of the most useful tools in people's lives, the presence of cyber-attacks that can cause damage, modification, and theft of vital data and information over this platform has increased. Utilization of soft-computing based on the behavior of the network may detect new or modified old attacks. An information security system is developed for the recognition the network infrastructure's behavior. This is limited to Normal, DoS, Probe, U2R, and R2L. The packets on the network are processed in MATLAB and analyze using Fuzzy Logic, Artificial Neural Network, and Fuzzy-Neural Network. Different tests are done with different datasets of varied parameters. The best model for each algorithm, which is rendered from the tests, is used for the information security system. The cyber-attacks were identified within a short period: 51.64us for Fuzzy Logic, 1.34us for Artificial Neural Network, and 14.23us for the Fuzzy Neural Network. The detection rate and accuracy of the three algorithms are 94.84%, 98.51%, 98.60% and 89.74%, 96.09%, 96.19% respectively. The Fuzzy Neural Network has the best performance which used the advantage of Fuzzy Logic and Artificial Neural Network.

[1]  Mohamed Ben Ahmed,et al.  Intrusion detection based on “Hybrid” propagation in Bayesian Networks , 2009, 2009 IEEE International Conference on Intelligence and Security Informatics.

[2]  Yasser Morgan,et al.  Real-time Support Vector Machine based Network Intrusion Detection system using Apache Storm , 2016, 2016 IEEE 7th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON).

[3]  Nouria Harbi,et al.  Real detection intrusion using supervised and unsupervised learning , 2013, 2013 International Conference on Soft Computing and Pattern Recognition (SoCPaR).

[4]  Irfan-Ullah Awan,et al.  Cyber-Attack Modeling Analysis Techniques: An Overview , 2016, 2016 IEEE 4th International Conference on Future Internet of Things and Cloud Workshops (FiCloudW).

[5]  Ali A. Ghorbani,et al.  Toward developing a systematic approach to generate benchmark datasets for intrusion detection , 2012, Comput. Secur..

[6]  Atilla Özgür,et al.  A review of KDD99 dataset usage in intrusion detection and machine learning between 2010 and 2015 , 2016, PeerJ Prepr..

[7]  Yogita Danane,et al.  Intrusion detection system using fuzzy genetic algorithm , 2015, 2015 International Conference on Pervasive Computing (ICPC).

[8]  Ali A. Ghorbani,et al.  A detailed analysis of the KDD CUP 99 data set , 2009, 2009 IEEE Symposium on Computational Intelligence for Security and Defense Applications.

[9]  Pll Siinksen,et al.  Control , 1999, Diabetic medicine : a journal of the British Diabetic Association.

[10]  Sireesha Rodda,et al.  Class imbalance problem in the Network Intrusion Detection Systems , 2016, 2016 International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT).