Dear Manoharan,

As today many optimization algorithms are available to find the best solution of the problem by the minimization or maximization of the   objective function. so these algorithm will definitely improve your  results for the website phishing.

Hi Monoharan,

The swarm based optimization techniques like ACO (Ant Colony Opt.), BFOA (Bacterial Foraging Opt.), PSO(Particle swarm  opt.) are good choice for website phishing kind of problems. You need to identify accurate parameters and objective function (if it is MOO, the problem will be complex) to run any of the aforementioned intelligent optimization algorithm. The time can definitely a constraint here and your fitness function should be chosen to address the same formally.

you can also think hybrid Genetic algorithm for complex problems. 

I have also read your another post about difference between SVM and ant optimization. Optimization algorithms can be applied to improve the performance of classifiers by tuning parameters/inputs.  In your question, how do you want to solve the website phishing by optimization algorithm?  Classify/ Identify phishing sites from the perspective of optimizaiton?

I wish to recommend you to use differential evolution and harmony search.

Let me put this in a simple way. When you a number of possible choices that depend on a number of factors (parameters), optimization algorithms could play a crucial role in choosing the "Best" solution based on a value obtained from an objective function. This function will provide a measure to indicate how good/bad the choice is. Probably the most important thing in all of these meta-heuristic algorithms (ACO, HS, BCO, etc.) is how to build this function. In your case, you must select the factors that represent how likely the website is a phishing website and create a "function" based on these parameters. In this case the function will either give higher or lower value to indicate the result. The optimization algorithm will use this function to select which one is more likely a threat or not a threat. All meta-heuristic algorithms depend on the same concept of objective function but they have different methods.

Kindly, go through the following publication.

Y. del Valle, G. Venayagamoorthy, S. Mohagheghi, J.-C. Hernandez, R. Harley, Particle swarm optimization: Basic concepts, variants and applications in power systems, IEEE Transactions on Evolutionary Computation, 12 (2) (2008) 171-195.

I hope it will be very helpful to you.

Regards,

M.N.Alam