University of California, Riverside (UC Riverside) computer scientists have developed a tool that pinpoints online malware source code repositories with 89% accuracy.
The UC Riverside researchers employed a supervised learning strategy to scan 97,000 malware-related software repositories, locating more than 75,000 malware source code repositories.
The team utilized malware-related keywords to find and download 1,000 repositories on GitHub, rigorously probing each and labeling those designated as malicious.
These were divided into subsets, with components of one subset's repositories used to train a supervised machine learning algorithm.
UC Riverside's Michalis Faloutsos said, "The implications are huge: having such a large database of malware can really help security researchers develop better defenses."
From UC Riverside News
View Full Article
Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA