Detection Of phishing URLs using a Term Frequency Inverse Document Frequency (TF-IDF).
M - Sibhathallah, Sathya Srinivas D - · International Journal For Multidisciplinary Research · 2024
Phishing attacks continue to pose a threat to online security as hackers employ more sophisticated tactics to trick users into revealing sensitive information. One common method is creating fake login URLs that mimic legitimate websites, making it challenging for users to distinguish between safe and dangerous links. This study focuses on detecting real-world phishing URLs, particularly analyzing login URLs. Our aim is to develop effective techniques and tools for early detection and prevention of phishing attempts by examining the common traits and patterns associated with them. The paper seeks to enhance people's and organizations' resilience to phishing attacks by exploring advanced technology and machine learning algorithms, ultimately contributing to a safer online environment. The research evaluates the effectiveness of deep learning and machine learning techniques in classifying phishing URLs, using statistical features such as Term Frequency-Inverse Document Frequency (TF-IDF) in combination with character N-gram, as well as proposed handcrafted features. CNN models are utilized for the deep learning approaches. Following model training, it is used to classify phishing URLs. The primary aim of this experiment is to assess the effectiveness of our proposed method. The results demonstrate that our phishing URL detection method achieves an accuracy rate of 96.6%