Comparison of Supervised Machine Learning Algorithms for Malware Detection - Focus on Detection Accuracy Results

Aug 3, 2026

Uploaded by Patrick

Cybersecurity is essential amid the prevalence of security issues and ongoing cyberattacks, as malware has evolved rapidly in recent years. Machine learning (ML) is increasingly used to detect malware by comparing supervised algorithms with detection accuracy as the main evaluation criterion. The study develops supervised techniques including Decision Tree, K-Nearest Neighbors, Naive Bayes, Random Forest, and Neural Networks using a Windows malware dataset. Performance is compared via confusion-matrix-derived metrics: False Positive Rate, True Positive Rate, and False Negative Rate, showing Decision Tree and Random Forest at 96% accuracy and K-NN at 95%.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Highlights

Abstract

Introduction

Methodology

Supervised ML Algorithms

Dataset and Feature Extraction

Evaluation Metrics (Confusion Matrix)

Results and Discussion

Conclusion and Future Work

Journal of Computing Research and Innovation (JCRINN) Vol. 8 No.2 (2023) [https://jcrinn.com](https://jcrinn.com : eISSN: 2600-8793 /)[ : eISSN: 2600-8793 /](https://jcrinn.com : eISSN: 2600-8793 /)[https://dx.doi.org/10.24191/jcrinn.v8i2.329](https://dx.doi.org/10.24191/jcrinn.v8i2.329) Comparison of Supervised Machine Learning Algorithms for Malware Detection Mohd Faris Mohd Fuzi1, Syamir Mohd Shahirudin2, Iman Hazwam Abd Halim3, Muhammad Nabil Fikri Jamaluddin4 1,2,3,4 College of Computing, Informatics, and Mathematics, Universiti Teknologi MARA Perlis Branch, Arau Campus, 02600 Arau, Perlis, Malaysia. Corresponding author: * [farisfuzi@uitm.edu.my](farisfuzi@uitm.edu.my) Received Date: 31 August 2022 Accepted Date: 27 April 2023 Revised Date: 5 May 2023 Published Date: 1 September 2023 HIGHLIGHTS ● Malware is becoming more sophisticated, making it difficult to detect using malware detection software. ● Machine learning algorithm techniques have grown in popularity among researchers for analysing malware detection. ● Focus on the supervised machine learning algorithm for malware detection. ● The best algorithm for malware detection will have the highest percentage of detection accuracy. ABSTRACT Due to the prevalence of security issues and cyberattacks, cybersecurity is crucial in today's environment. Malware has also evolved significantly over the past few years. With the advancement ofmalware analysis, Machine Learning (ML) is increasingly being used to detect malware. This study's major objective is to compare the best-supervised ML algorithms for malware detection based on detection accuracy. This study includes the scripting and development ofsupervised ML techniques such as Decision Tree (DT), K-Nearest Neighbors (KNN), Naive Bayes, Random Forest, and Neural Networks. This study was solely concerned with the Windows malware dataset. The malware classification was determined by testing and training the supervised ML algorithms using the extracted features from the malware dataset. Then, the percentage of detection accuracy was used to compare the detection performance of all five algorithms. The detection accuracy is calculated using the confusion matrix, which includes the False Positive Rate (FPR), the True Positive Rate (TPR), and the False Negative Rate (FNR). The results indicated that the Decision Tree and Random Forest algorithms provided the best detection accuracy at 96%, followed by the K-NN algorithm at 95%. To improve the detection accuracy for future research, it is suggested that the malware dataset be enhanced using several architectures, such as Linux and Android, and use additional supervised and unsupervised machine learning algorithms. Keywords: supervised machine learning; malware detection; detection accuracy; machine learning algorithms INTRODUCTION People nowadays use the internet for a variety of purposes, including shopping, watching videos, listening to music, and even filing taxes. It is no secret that the World Wide Web's quick expansion over the last two decades has brought us fantastic things and given us the ability to do our tasks from the comfort of our own Copyright© 2023 UiTM Press. This is an open access article licensed under CC BY-SA [https://creativecommons.org/licenses/by-sa/4.0/](https://creativecommons.org/licenses/by-sa/4.0/) Journal of Computing Research and Innovation (JCRINN) Vol. 8 No.2 (2023) [https://jcrinn.com](https://jcrinn.com : eISSN: 2600-8793 /)[ : eISSN: 2600-8793 /](https://jcrinn.com : eISSN: 2600-8793 /)[https://dx.doi.org/10.24191/jcrinn.v8i2.329](https://dx.doi.org/10.24191/jcrinn.v8i2.329) homes or offices. But, as with many excellent things, there is always the other side, which is not always sonice. Cybersecurity is a term that most people are not familiar with or are not interested in because it can be complicated. The basic purpose of cybersecurity is to secure the user, their data, and any other sen

FAQs

What is the main objective of the study?

To compare the best supervised machine learning algorithms for malware detection based on detection accuracy.

Which supervised machine learning algorithms are included?

Decision Tree, K-Nearest Neighbors (KNN), Naive Bayes, Random Forest, and Neural Networks.

How is detection performance evaluated in the study?

Using a confusion matrix to compute False Positive Rate (FPR), True Positive Rate (TPR), and False Negative Rate (FNR), and then comparing detection accuracy.

Aug 3, 2026

Uploaded by Patrick