Covid-19 Phishing Detection Based on Hyperlink Using K-Nearest Neighbor (KNN) Algorithm

Abstract

Phishing scam grow bigger during COVID-19 pandemic as the victim easily being deceived to click on the hyperlink that include latest information related to COVID-19. This link is sent by unknown user through email claimed to be from trusted organization. Although various way has been proposed to overcome this issue, number of phishing attack keep increasing. Our work focused on detecting phishing email related to COVID-19 using KNN Algorithm based on hyperlink approach. We consider using Uniform Resource Locator (URL) features such as Generic_TLD, URL_Length, Having_Sub_Domain, Prefix_Suffix and Having_Slash where the dataset is taken from Phishtank, SpyCloud, DomainTool and Kaggle. The phishing URL detection model will be tested on KNN Algorithm in terms of accuracy rate. This research produces promising results using 5 features with 97.80% accuracy for Dataset 1 and 99.60% accuracy for Dataset 2.