Comparative Analysis of Machine Learning Classifiers for Phishing Detection

In recent years, communication over the Internet has become the most effective media for leveraging social interactions during the COVID-19 pandemic. Nevertheless, the rapid increase use of digital platforms has led to a significant growth of Phishing Attacks. Phishing attacks are one of the most common security issues in digital worlds that can affects both individual and organization in keeping their confidential information secure. Various modern approaches can be used to target an individual and trick them into leaking their sensitive information, which can later, purposely be used to harm the targeted victim or entire organization depending on the cybercriminal's intent and type of data leaked. This paper evaluates phishing detection by using Naïve Bayes, Simple Logistic, Random Forest, Ada Boost and MLP classifications. This study discusses the comparative analysis on the effectiveness of classification for detecting phishing attacks. The results indicated that the detection system trained with the Random Forest produce higher accuracy of 97.98% than another classifier method.