Description: The CybAttT dataset contains 36,071 manually labeled English cyberattack tweets These tweets are categorized into three classes: high-risk news, normal news, and not news.
Purpose: It's designed to enhance threat intelligence by providing a labeled dataset for training and testing cyber threat detection models
Access: CybAttT: A Dataset of Cyberattack News Tweets for Enhanced Threat Intelligence (mdpi.com)
GitHub Repository: twitter-cyberthreat-detection
Description: This repository holds the dataset used for experiments on cyberthreat detection from Twitter using deep neural networks4.
Contents: The repository includes data, models, and scripts for building and evaluating the neural network models
Access: twitter-cyberthreat-detection/README.md at master · ndionysus/twitter-cyberthreat-detection · GitHub
CyATT contains such tweets : "the keyword encompasses three fundamental attack categories: data breach, cyber breach, and cyberattack:
Cyberattack: It is broader than a data breach. As well it is deliberate and considered more disruptive to business. The most common types of cyberattacks are malware, Denial of Service (DoS), Distributed DoS (DDoS), Phishing, Ransomware, password attacks, poor security, spam, and SQL injection.
Hence, the CyAttT dataset was collected by searching for any posted tweet on X platform using 45 keywords (Figure 1) related cyberattacks, data breach, and cyber breach."