Abstract :

Keylogger Malware Poses A Serious Cybersecurity Threat By Secretly Recording User Keystrokes And Transmitting Sensitive Information Such As Passwords, Financial Data, And Personal Messages To Unauthorized Entities. Traditional Signature-based Detection Methods Are Often Ineffective Against Newly Developed Or Modified Keylogger Variants. To Address This Challenge, This Study Proposes A Machine Learning-based Framework For Detecting Keylogger Activities Through Behavioral Analysis Of System Processes. The Proposed Approach Analyzes System-level Features Such As Keyboard API Calls, Keystroke Frequency, CPU Usage Patterns, Memory Access Behavior, And Network Communication Characteristics. The Dataset Containing Both Benign Processes And Malicious Keylogger Samples Is Preprocessed Using Normalization And Feature Selection Techniques To Improve Model Performance. Several Machine Learning Classifiers Including Decision Tree, KNearest Neighbor, Support Vector Machine, And Random Forest Are Implemented Using Pythonbased Tools Such As Scikit-learn, Pandas, And NumPy. Experimental Results Demonstrate That The Random Forest Model Achieves The Highest Detection Accuracy Of 96.3%, Outperforming Other Classification Algorithms And Traditional Detection Methods. The Proposed System Provides An Efficient And Scalable Solution For Detecting Keylogger Malware Based On Behavioral Characteristics, Thereby Enhancing System Security And Protecting Sensitive User Information. Keywords: Keylogger Detection, Machine Learning, Behavioral Analysis, Malware Detection, Cybersecurity

Published:

09-3-2026

Issue:

Vol. 26 No. 3 (2026)

Page Nos:

69-76

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite