An Approach to Target Automated Malware Detection in Internet through IOT Devices Association Models

Abstract

Attackers are shifting aggressively to the
Internet of Things devices in the last few
years. In this paper, we propose the
association IoT malware detection scheme.
First, it extracts static features of the onechannel gray-scale images converted from
binaries and extracts the system called graph
dynamically. The extracted features are
trained through machine learning and deep
learning models to detect IoT malware. The
performance of the proposed method was
evaluated, and its detection accuracy was
99.14% than in the static analysis and
dynamic analysis, which had 99.06% and
98.08% detection accuracy, respectively, with
8339 samples collected from 25 different
malware families, this method had a 94.5%
accuracy rate. Deep convolutional neural
networks were developed by Farhan et al. to
visualize color images and identify malware
threats on the Internet which we decided to
compare. Identify malware threats technique
through our presented technique.
Improvements in categorization performance
were seen in our experimental findings, which
monitor dangers to computer security. This
more authenticates the Nasir et al. developed
approach based on Random Coefficient
Selection and Mean Modification
(RCSMMA). Therefore our approach covers
many common cyberattacks, through the
purposed method by comparing analysis of
other models and producing positive
performance.