Web Guardian: Harnessing Web Mining to Combat Online Terrorism

The rapid growth of the internet has increased the misuse of online platforms for phishing attacks and the spread of terrorism-related content.Manual detection of such threats is inefficient due to the large and dynamic nature of web data.This paper presents Web Guardian, an intelligent web mining-based system designed to detect phishing URLs and terrorism-related textual content using machine learning and deep learning techniques.The system integrates URL feature extraction with a Random Forest classifier for phishing detection and employs TF-IDF-based text vectorization combined with an Artificial Neural Network for terrorism content classification.Web scraping and keyword-based analysis are also used to compute heuristic risk scores for improved interpretability.The proposed framework is implemented as a Flask-based web application that enables efficient analysis of URLs and text, demonstrating reliable performance for early threat detection and web security support.