Vigorous Malware Detection in IoT Devices Using Machine Learning

EasyChair Preprint 9838

Abstract

The use of Internet of Things (IoT) devices has grown significantly due to the expansion of the internet.
However, these devices now contain large amounts of data, making them vulnerable to malware attacks. As a result, detecting malware in IoT devices has become a critical issue. Although many researchers have proposed various methods, accurately identifying advanced malware still poses a challenge.
To tackle this problem, we suggest a deep learning-based ensemble classification method for identifying malware in IoT devices. Our method comprises three steps:
(1) preprocessing the data using scaling, normalization, and de-noising,
(2) selecting features and applying one-hot encoding, and
(3) using an ensemble classifier that combines convolutional neural network (CNN) and long short-term memory (LSTM) outputs for malware detection.

We have evaluated our proposed method using standard datasets and compared it to state-of-the-art techniques. Our approach outperforms existing techniques and achieved an average accuracy of 99.5%.

Keyphrases: Malware, N-Gram sequence algorithm, Support Vector Machine, Vigorous malware detection, deep learning, machine learning

@booklet{EasyChair:9838,
  author    = {T Renny and Sk Khadar Baba and P Abhilash Reddy and P Deepanth},
  title     = {Vigorous Malware Detection in IoT Devices Using Machine Learning},
  howpublished = {EasyChair Preprint 9838},
  year      = {EasyChair, 2023}}