Info: Zenodo’s user support line is staffed on regular business days between Dec 23 and Jan 5. Response times may be slightly longer than normal.

Published March 1, 2020 | Version v1
Journal article Open

Phishing detection system using machine learning classifiers

  • 1. University Malaysia Pahang
  • 2. Universitas Sriwijaya
  • 3. Universiti Tun Hussein Onn Malaysia
  • 4. Universitas Ahmad Dahlan

Description

The increasing development of the Internet, more and more applications are put into websites can be directly accessed through the network. This development has attracted an attacker with phishing websites to compromise computer systems. Several solutions have been proposed to detect a phishing attack. However, there still room for improvement to tackle this phishing threat. This paper aims to investigate and evaluate the effectiveness of machine learning approach in the classification of phishing attack. This paper applied a heuristic approach with machine learning classifier to identify phishing attacks noted in the web site applications. The study compares with five classifiers to find the best machine learning classifiers in detecting phishing attacks. In identifying the phishing attacks, it demonstrates that random forest is able to achieve high detection accuracy with true positive rate value of 94.79% using website features. The results indicate that random forest is effective classifiers for detecting phishing attacks.

Files

05 21155.pdf

Files (347.8 kB)

Name Size Download all
md5:138214473251e52a7c60db9cf8a5885f
347.8 kB Preview Download