Optimizing Neural Network Parameters For Effective Classification of Benign and Malicious Websites

Bio-inspired algorithms are the most powerful way to solve optimization problems. The objective of this paper is to use optimized network parameters for website classification and the efficiency of Neural Network is improved by optimized network parameters. The network parameters are optimized using two bio-inspired algorithms: Particle Swarm Optimization (PSO) and Cuckoo Search (CS) and the optimized parameter values are used with two neural network models, a standard Multi-Layer Feed Forward Network with Backpropagation (BPN) and Radial Basis Function (RBF) Network. Security is one of the major concerns in this digital era. There are numerous websites, which are potentially risky in spreading malicious files. It is difficult to detect such websites. In this work, Neural Network is used to classify the websites as benign and malicious. The proposed neural network models are tested with URL dataset. The experimental results are assessed in terms of Error reduction, training time and classification accuracy. The experimental result shows that the optimized network parameters have given good improvement in classification with faster convergence.