Malytics: A Malware Detection Scheme

by   Mahmood Yousefi-Azar, et al.
University of Newcastle
Macquarie University

An important problem of cyber-security is malware analysis. Besides good precision and recognition rate, a malware detection scheme needs to be able to generalize well for novel malware families (a.k.a zero-day attacks). It is important that the system does not require excessive computation and is not itself vulnerable. In this paper, we propose a novel scheme to detect malware which we call Malytics. It is not dependent on any particular tool or operating system. It extracts static features of any given binary file to distinguish malware from benign. Malytics consists of three stages: feature extraction, similarity measurement and classification. The three phases are implemented by a neural network with two hidden layers and an output layer. We show feature extraction, which is performed by tf-simhashing, is equivalent to the first layer of a particular neural network. We evaluate Malytics performance on both Android and Windows platforms. Malytics outperforms a wide range of learning-based techniques and also individual state-of-the-art models on both platforms. The byte-level feature analysis of Malytics makes it more difficult to craft adversarial samples to mislead Malytics. We introduce a possible attack technique to Malytics and demonstrate both its resilience and robust performance in addressing zero-day malware samples and adversarial samples. The F1-score of Malytics is 97.36 files respectively, in the applied datasets. The speed and efficiency of Malytics are also evaluated.


page 1

page 2

page 3

page 4


Review of Deep Learning-based Malware Detection for Android and Windows System

Differentiating malware is important to determine their behaviors and le...

Feature Engineering Using File Layout for Malware Detection

Malware detection on binary executables provides a high availability to ...

Explaining Vulnerabilities of Deep Learning to Adversarial Malware Binaries

Recent work has shown that deep-learning algorithms for malware detectio...

SeqNet: An Efficient Neural Network for Automatic Malware Detection

Malware continues to evolve rapidly, and more than 450,000 new samples a...

Function Identification in Android Binaries with Deep Learning

Application security support has become a preference for the enterprise ...

"Influence Sketching": Finding Influential Samples In Large-Scale Regressions

There is an especially strong need in modern large-scale data analysis t...

Please sign up or login with your details

Forgot password? Click here to reset