Image-based Malware Family Detection: An Assessment between Feature Extraction and Classification Techniques.

Authors

Giacomo Iadarola, Fabio Martinelli, Francesco Mercaldo, Antonella Santone

Publication

Proceedings of the 5th International Conference on Internet of Things, Big Data and Security - Volume 1: AI4EIoTs , 499-506, 2020
https://doi.org/10.5220/0009817804990506

Abstract

The increasing number of malware in mobile environment follows the continuous growth of the app stores, which required constant research in new malware detection approaches, considering also the weaknesses of signature-based anti-malware software. Fortunately, most of the malware are composed of well-known pieces of code, thus can be grouped into families sharing the same malicious behaviour. One interesting approach, which makes use of Image Classification techniques, proposes to convert the malware binaries to images, extract feature vectors and classifying them with supervised machine learning models. Realizing that researchers usually evaluate their solutions on private datasets, it is difficult to establish whether a model can be generalized on another dataset, making it difficult to compare the performance of the various models. This paper presents a comparison between different combination of feature vector extraction methods and machine learning models. The methodology aimed t o evaluate feature extractors and supervised machine learning algorithms, and it was tested on more than 20 thousand images of malware, grouped into 10 different malware families. The best classifier, a combination of GIST descriptors and Random Forest classifiers, achieved an accuracy of 0.97 on average.