A Framework of New Hybrid Features for Intelligent Detection of Zero Hour Phishing Websites

Nagunwa, Thomas and Naqvi, S. and Fouad, Shereen and Shah, H. (2019) A Framework of New Hybrid Features for Intelligent Detection of Zero Hour Phishing Websites. In: International Conference on Computational Intelligence in Security for Information Systems (CISIS 2019), 13-15 May 2019, Seville, Spain.

Framework for detection of zero hour phishing webpages - Camera Ready.pdf - Accepted Version

Download (162kB)


Existing machine learning based approaches for detecting zero hour phishing websites have moderate accuracy and false alarm rates and rely heavily on limited types of features. Phishers are constantly learning their features and use sophisticated tools to adopt the features in phishing websites to evade detections. Therefore, there is a need for continuous discovery of new, robust and more diverse types of prediction features to improve resilience against detection evasions. This paper proposes a framework for predicting zero hour phishing websites by introducing new hybrid features with high prediction performances. Prediction performance of the features was investigated using eight machine learning algorithms in which Random Forest algorithm performed the best with accuracy and false negative rates of 98.45% and 0.73% respectively. It was found that domain registration information and webpage reputation types of features were strong predictors when compared to other feature types. On individual features, webpage reputation features were highly ranked in terms of feature importance weights. The prediction runtime per webpage measured at 7.63s suggest that our approach has a potential for real time applications. Our framework is able to detect phishing websites hosted in either compromised or dedicated phishing domains.

Item Type: Conference or Workshop Item (Paper)
13 March 2019Accepted
28 April 2019Published Online
Uncontrolled Keywords: Phishing, Phishing webpage detection, Zero hour phishing website, Webpage features, Machine learning
Subjects: CAH11 - computing > CAH11-01 - computing > CAH11-01-01 - computer science
CAH11 - computing > CAH11-01 - computing > CAH11-01-03 - information systems
CAH11 - computing > CAH11-01 - computing > CAH11-01-05 - artificial intelligence
Divisions: Faculty of Computing, Engineering and the Built Environment > School of Computing and Digital Technology
Depositing User: Syed Naqvi
Date Deposited: 16 Mar 2019 12:17
Last Modified: 22 Mar 2023 12:01
URI: https://www.open-access.bcu.ac.uk/id/eprint/7254

Actions (login required)

View Item View Item


In this section...