Classification of Website Phishing Data through Machine Learning Algorithms

Muhammad Taseer Suleman

  • Muhammad Taseer Suleman Lahore Garrison University
Keywords: Phishing, Spamming, Features, Machine learning algorithms, Genetic algorithms.


Phishing is the dissemination of malicious web sites used to acquire passwords, credit card details or any sensitive personal information. Clients of web advancements deal with different security dangers and phishing is a standout amongst the most imperative dangers that should be addressed. Phishing sites have certain attributes and designs, in order to, distinguish those components that can help us to recognize phishing. In order to, recognize such elements information mining methods have been utilized. In this work, we depicted examination in arrangement of phishing sites utilizing diverse classification algorithms with genetic algorithms for enhancement, for example, as feature selection and generation. Keeping in mind the end goal to figure out which technique gives the prime outcomes in phishing sites characterization. Websites are characterized as “1” for "Legitimate”, “0” for "Suspicious" and “-1” for "Illegitimate". We have found that machine-learning algorithms along with feature selection algorithms were the best choice for detecting web phishing attacks.