Analyzing XSS Attack Information Content on Social Media

Social media has become more than just a communication tool; it's now a rapidly evolving platform for sharing information, including cybersecurity threats like XSS attacks. This study investigates if analyzing collections of tweets can identify those containing XSS attack information. By accurately extracting this information, we can effectively classify tweets as containing or not containing XSS attacks. Our research focuses on evaluating how well feature extraction methods perform in achieving this classification accuracy. As the result, the 2-gram model with TF-IDF Vectorizer might be a more suitable choice with accuracy exceeding 97% and demonstrates better tolerance for overfitting due to the inherent properties of TF-IDF.