Labeled Dataset of Cybersecurity Events from Dark Web and Surface Web Forums

This dataset contains a collection of cybersecurity posts identified and classified from Dark Web and Surface Web forums. The dataset aims to provide a labeled resource for training and validating machine learning models focused on Cyber Threat Intelligence (CTI). The datasets labeled as "Events" were generated using a custom Event Detector developed as part of this Master's research.
It is structured in four versions:

Raw Dark Web Posts: Contains the complete set of posts collected from darknet forums. This file represents the initial corpus before event filtering.

Raw Surface Web Posts: Contains the full collection of posts gathered from clear web cybersecurity forums.

Events Dark Web Labeled : This subset contains only the posts identified as cybersecurity events within the Dark Web corpus. Each entry is labeled with a relevance score (High, Medium, or Low) as validated in the study.

Events Surface Web Labeled : This subset contains the cybersecurity events detected within the Surface Web corpus, categorized by their respective relevance levels.

This dataset can support machine learning research in threat classification, text mining, and cybersecurity automation. The code associated with the data can be found in https://github.com/rodrigobernabesilveira/deteccao-eventos-dark-surface