Abstract:Malware is one of the most significant threats in today’s computing world since the number of websites distributing malware is increasing at a rapid rate. Malware analysis and prevention methods are increasingly becoming necessary for computer systems connected to the Internet. This software exploits the system’s vulnerabilities to steal valuable information without the user’s knowledge, and stealthily send it to remote servers controlled by attackers. Traditionally, anti-malware products use signatures for detecting known malware. However, the signature-based method does not scale in detecting obfuscated and packed malware. Considering that the cause of a problem is often best understood by studying the structural aspects of a program like the mnemonics, instruction opcode, API Call, etc. In this paper, we investigate the relevance of the features of unpacked malicious and benign executables like mnemonics, instruction opcodes, and API to identify a feature that classifies the executable. Prominent features are extracted using Minimum Redundancy and Maximum Relevance (mRMR) and Analysis of Variance (ANOVA). Experiments were conducted on four datasets using machine learning and deep learning approaches such as Support Vector Machine (SVM), Naïve Bayes, J48, Random Forest (RF), and XGBoost. In addition, we also evaluate the performance of the collection of deep neural networks like Deep Dense network, One-Dimensional Convolutional Neural Network (1D-CNN), and CNN-LSTM in classifying unknown samples, and we observed promising results using APIs and system calls. On combining APIs/system calls with static features, a marginal performance improvement was attained comparing models trained only on dynamic features. Moreover, to improve accuracy, we implemented our solution using distinct deep learning methods and demonstrated a fine-tuned deep neural network that resulted in an F1-score of 99.1% and 98.48% on Dataset-2 and Dataset-3, respectively.

Multi-context Features for Detecting Malicious Programs

A Malware Detection Method Based on Sandbox, Binary Instrumentation and Multidimensional Feature Extraction

Malware Analysis Using Machine Learning and Deep Learning Techniques

Detection of Malicious Software by Analyzing Distinct Artifacts Using Machine Learning and Deep Learning Algorithms

A malware detection framework based on kolmogorov complexity

An android malware detection approach using multi-feature fusion and TF-IDF algorithm

MCES: Multi-classifier Ensemble System for Malware Detection and Identification

TS-Mal: Malware Detection Model Using Temporal and Structural Features Learnin

Deep Multi-Task Learning for Malware Image Classification

Malware Detection Based on Multi-level and Dynamic Multi-feature Using Ensemble Learning at Hypervisor

Deep hybrid approach with sequential feature extraction and classification for robust malware detection

A Novel Approach to Malicious Code Detection Using CNN-BiLSTM and Feature Fusion

Detection of Advanced Malware by Machine Learning Techniques

Learning from Context: Exploiting and Interpreting File Path Information for Better Malware Detection

Fast and Efficient Malware Detection with Joint Static and Dynamic Features Through Transfer Learning

A malware detection scheme based on mining format information

Malware Detection Using Deep Learning and Correlation-Based Feature Selection

A Comprehensive Review of Machine Learning Approaches for Detecting Malicious Software

Deep Neural Network Based Malware Detection Using Two Dimensional Binary Program Features

Decoding Android Malware with a Fraction of Features: An Attention-Enhanced MLP-SVM Approach

Interpretable Detection of Malicious Behavior in Windows Portable Executables Using Multi-Head 2D Transformers