Analysis of Machine Learning Techniques on URL Phishing Dataset

  • Robiah Yusof Universiti Teknikal Malaysia Melaka
  • Muhammad Aidil Hakim Ahmad Universiti Teknikal Malaysia Melaka
  • Nurul Azma Zakaria Universiti Teknikal Malaysia Melaka
  • Najiahtul Syafiqah Ismail Universiti Teknologi Mara Kuala Terengganu


The Internet has become a vital part of daily life, as do almost all online social and financial practices. However, rising phishing sites today faced significant threats because of their appallingly imperceptible danger. Phishing is an online fraudulent act that uses social engineering and technical subterfuge to trick internet users and capture their sensitive data or critical information online. There is a lack of knowledge on implementing a suitable classification technique on machine-learning tools for analyzing phishing URLs or Websites. This research aims to identify the best classification technique using the orange tool on these three datasets and implement the phishing URL analysis methodology comprising six phases. Based on the result, Decision Tree is the best classification technique for identifying the URL phishing attack. It has obtained the highest accuracy result of 88.30% and 70.70% in Dataset 2 and Dataset 3, respectively. In the future, more classification techniques or machine learning tools with different performances are explored to analyze Phishing URLs / Websites for better results.

How to Cite
Yusof, R., Ahmad, M. A., Zakaria, N. A., & Ismail, N. S. (2023). Analysis of Machine Learning Techniques on URL Phishing Dataset. Journal of Advanced Computing Technology and Application (JACTA), 5(1), 57-70. Retrieved from