You are here:, UncategorizedIdentifying Malware By Turning It Into An Image?

Sounds crazy? Well that’s exactly what the researchers at Microsoft and Intel have done. The two companies are collaborating in a research projectwhere they tried a new approach to classifying Malware.

They call this new process Static Malware as image Network Analysis, STAMINA in short. This new algorithm converts Malware samples into a grey scale image, then uses image analysis (AI) to detect patterns that are specific to Malware.

To help in the research Microsoft reportedly submitted 2.2million infected executable for both training and testing.

The research team reported that STAMINA was able to identifyMalware samples with an  99.07%

“The results certainly encourage the use of deeptransfer learning for the purpose of malware classification,” said Jugal Parikh and Marc Marino, two of the researchers involved in the project.

There was  a caveat with this new approach however, Microsoft reported though STAMINA was fast and accurate with smaller files, it “faultered” with larger ones.

Microsoft has been making concentrated efforts andinvestments in using deep learning in efforts to improving malware detections.

It’s a pretty ingenious way to detect Malware, check out the research paper if you are interested in reading more

About the author:

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Top