Prompt-Based Learning for Thread Structure Prediction In Cybersecurity Forums

Anonymous

Abstract: With recent trends indicating cyber crimes increasing in both frequency and cost, it is imperative to develop new methods that leverage data-rich hacker forums to assist in combating ever evolving cyber threats. Defining interactions within these forums is critical as it facilitates identifying highly skilled users, which can improve prediction of novel threats and future cyber attacks. We propose a method called Next Paragraph Prediction with Instructional Prompting (NPP-IP) to predict thread structures while grounded on the context around posts. The experimental evaluation shows that our proposed method can predict the thread structure significantly better than existing methods allowing for better social network prediction based on forum interactions.

Paper Type: short