EKRV: Ensemble of kNN and Random Committee Using Voting for Efficient Classification of Phishing
Any efficient anti-phishing tool must be able to classify phishing activity as ‘phishing’ with utmost accuracy. The key factor that influences the accuracy of an anti-phishing tool is the selection of a classification algorithm whose prediction accuracy is the maximum with nil or least false-positive rate. This paper proposes the implementation of a hybrid approach involving random committee that is a type of Ensemble classification technique and k-nearest neighbor (kNN) algorithm which is available as IBK (instance-based with k neighbors) on WEKA, resulting in most encouraging prediction accuracy values. The proposed scheme is followed after the preprocessing phase that involves feature extraction using Consistency Subset Eval algorithm with the Greedy Stepwise search technique.
KeywordsRandom committee kNN Phishing Voting Ensemble classifiers
- 1.Niranjan, A., Nitish, A., Deepa Shenoy, P., Venugopal, K.R.: Security in data mining—a comprehensive survey. Global J. Comput. Sci. Technol. 16(5), 52—73 (2017)Google Scholar
- 2.Lakshmi, V.S., Vijaya, M.S.: Efficient prediction of phishing websites using supervised learning algorithms. Procedia Eng. 30, 798—805 (2012)Google Scholar
- 4.Yan, Z.: A genetic algorithm based model for Chinese phishing E-commerce websites detection. In: International Conference on HCI in Business, Government and Organizations, pp. 270—279 (2016)Google Scholar
- 6.Shah, R.K., Hossin, M.A., Khan, A.: Intelligent phishing possibility detector. Int. J. Comput. Appl. 148(7), pp. 1–8 (2016)Google Scholar
- 8.Mohammad, R.M., Thabtah, F., McCluskey, L.: Phishing Websites Features (2015). http://eprints.hud.ac.uk/24330/6/RamiPhishing_Websites_Features.pdf