Researchers at the University of Maryland Baltimore County and Booz Allen Hamilton, a U.S.-based consulting firm, have created a dataset to improve malware classification. It includes over four million malicious files labeled with 75 common malware categories and behaviors and nearly one million malicious files labeled with 43 file formats, operating systems, and programming languages. Researchers can use the dataset to train AI systems to better detect and guard against malware.
Image credit: Flickr user Daniel Foster