A Machine Learning Approach To Improve The Efficiency Of Fake Website Detection Techniques

Abstract

Phishing is a kind of cyber-attack in which perpetrators use spoofed emails and fallacious web sites to lure unsuspecting online users into giving up personal data. This paper takes a gander at the phishing issue totally by looking at different research works and their countermeasures, and how to increase detection. It makes out of two studies. In the main study, focus was on dataset assembling, feature extraction and preprocessing for the classification process. In the second study, focus was on metric evaluation of a set of classifiers (SVM, C4.5, LR and KNN) utilizing the precision, accuracy, f-measure and recall metrics. The output of the classifier study is utilized to pick the best performed classifier. The subsequent result of the study demonstrates that the classifier technique performed better with an accuracy of 99.37%. This outcome can be attributed to the small size of dataset utilized as it was appeared as a part of past scrutinizes that K-NN performs better with a decreasing size of dataset while classifiers like C4.5 and SVM performs better with increasing size of dataset.