Advertisement
Published in:
2020 | OriginalPaper | Chapter
Classification of Grayscale Malware Images Using the K-Nearest Neighbor Algorithm
Authors : Ikram Ben Abdel Ouahab, Mohammed Bouhorma, Anouar Abdelhakim Boudhir, Lotfi El Aachak
Published in: Innovations in Smart Cities Applications Edition 3
Publisher: Springer International Publishing
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Abstract
The biggest problem with recovering from cyberattacks is that security professionals rarely get the chance to deal with them immediately. So, using advanced intelligent techniques, we can defense systems against malware the moment it begins to download. For that reason, a new type of feature has been recently introduced for malware classification task, borrowing techniques from computer vision community called malware visualization technique. Malware classification goal is to know how they work, and then we can rapidly defend them especially in the case of zero-days attacks. In this paper, we adopt KNN algorithm to classify malwares based on their image visualization. So, a malware binary is converted to grayscale image. Then to extract similarities and dis-similarities from these images a GIST descriptor is computed. We used a database of 9339 samples of malwares belonging to 25 families. Our malware classifier reached a high score of 97%, which is very close to the results found in literature.