Phishing emails are a primary mode of entry for attackers into an organization. A successful phishing attempt leads to unauthorized access to sensitive information and systems. However, automatically identifying phishing emails is often difficult since many phishing emails have composite features such as body text and metadata that are nearly indistinguishable from valid emails. This paper presents a novel machine learning-based framework, the DARTH framework, that characterizes and combines multiple models, with one model for each composite feature, that enables the accurate identification of phishing emails. The framework analyses each composite feature independently utilizing a multi-faceted approach using Natural Language Processing (NLP) and neural network-based techniques and combines the results of these analyses to classify the emails as malicious or legitimate. Utilizing the framework on more than 150,000 emails and training data from multiple sources, including the authors’ emails and phishtank.com, resulted in the precision (correct identification of malicious observations to the total prediction of malicious observations) of 99.97% with an f-score of 99.98% and accurately identifying phishing emails 99.98% of the time. Utilizing multiple machine learning techniques combined in an ensemble approach across a range of composite features yields highly accurate identification of phishing emails.

Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License