Web Guardian: Harnessing Web Mining to Combat Online Terrorism
Prerna Sharma, Prakhranshu Singh, C. N. S. Vinoth Kumar · 2024
In recent years, terrorism has surged dramatically in specific regions, demanding immediate intervention to prevent its destructive impact on lives and property. Exploiting technological advancements, particularly the internet, terrorist groups utilize online platforms to spread propaganda, defame individuals, and recruit members for criminal activities. To effectively create a system, it is essential to use both web mining and data mining approaches to combat this danger. Web mining involves using several techniques for text mining to derive valuable insights from unorganized data. At the same time, data mining algorithms handle organized datasets, enhancing web mining's ability to navigate the intricacies of online material. However, the diverse data structures of websites pose challenges for unified algorithms. Web data mining assumes a critical role in detecting online terrorism spread, especially through the proliferation of phishing websites used for recruitment and coordination. By employing machine learning algorithms such as AdaBoost, SVM, Gradient Boosting and Random Forest, web data analysis can differentiate between legitimate and terrorist-associated sites by extracting features such as URL structure and domain age. Rigorous evaluation ensures the accuracy of these methods, enabling proactive measures like blocking identified terrorist activity sites. Through ongoing research and refinement, machine learning-driven web data mining emerges as a potent tool in combating the online dissemination of terrorist propaganda, reinforcing global security efforts.