Year: 2025
Type: article
Abstract: Phishing attacks, which trick users into divulging private information like passwords, credit card numbers, and personal information, have grown to be a serious cybersecurity risk. Blacklists and rule-based systems are examples of traditional security solutions that frequently fail to stop increasingly popular phishing websites. This study investigates how machine learning approaches can be used to identify phishing websites based on a variety of variables, including domain-based attributes, HTML content, and URL characteristics. Decision Trees, Random Forest, Support Vector Machines (SVM), and deep learning techniques are among the classification models that are trained using a dataset that includes both authentic and fraudulent websites. The accuracy, precision, recall, and F1-score measures are used to assess these models' performance. Results from experiments show that machine learning algorithms are capable of accurately and reliably classifying phishing websites, offering a scalable and reliable web security solution.
DOI: 10.46632/cset/3/1/2
Language: English
Topic: Spam and Phishing Detection
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4408823019