Phishing Detection System through Hybrid Machine Learning Based on URL

Aim:

         The aim of this research is to develop an advanced phishing detection system that leverages a hybrid machine learning approach to analyse URLs effectively and accurately identify potential phishing attempts.

Abstract:

       This research introduces an innovative phishing detection system centered on URL analysis through a hybrid ensemble of machine learning models. Aimed at enhancing accuracy and efficiency in identifying phishing attempts, the system extracts and thoroughly examines URL features. By employing a diverse ensemble comprising Logistic Regression, Decision Tree Classifier, Random Forest Classifier, Support Vector Classifier, XGBoost, and Gradient Boosting models, the system can effectively differentiate between legitimate and malicious URLs. This approach aims to reduce false positives, bolster accuracy, and confront the ever-evolving nature of phishing tactics. The system’s adaptability and comprehensive analysis of URL characteristics signify a significant stride in fortifying cybersecurity measures against phishing attacks.

Proposed Method:

        The proposed approach begins by extracting the destination URL and utilizes an ensemble of machine learning models such as Logistic Regression, Decision Tree Classifier, Decision Tree Regressor, Random Forest Classifier, Random Forest Regressor, Support Vector Classifier, XGB Classifier, XGB Regressor, XGB Model, and Gradient Boosting Classifier for enhanced phishing detection.

Advantages:

• The diverse ensemble of machine learning models enhances the system’s accuracy and robustness in identifying phishing URLs.
• The incorporation of multiple models allows for a more comprehensive analysis of URL features, increasing the system’s ability to adapt to evolving phishing tactics.