Active learning model used for android malware detection
Smartphones have become one of the main products in today’s world. However, the security risks of smartphones are high compared with those of other devices. Smartphone users face threats to their privacy and property protection. Android malware identification is essential to prevent mobile applicati...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-09-01
|
| Series: | Machine Learning with Applications |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666827025000635 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Smartphones have become one of the main products in today’s world. However, the security risks of smartphones are high compared with those of other devices. Smartphone users face threats to their privacy and property protection. Android malware identification is essential to prevent mobile applications from being compromised because a number of new techniques for Android malware assaults have appeared and evolved. Researchers are constantly improving and inventing new approaches for identifying Android malware. Modern research has extensively used machine-learning techniques. This study introduces a technique known as active learning, which analyzes device behavior and traffic monitoring on devices to discover malware. To identify malware more accurately, this hybrid architecture includes active learning, supervised machine-learning models, and threat intelligence tools. The results of the experiments demonstrated the effectiveness of our proposed model, which outperformed both supervised machine learning and the traditional machine learning technique in terms of training and test accuracies of 92.36 % and 85.9 % and loss functions of 22.5 % and 33.2 %, respectively, and achieved the highest degree of accuracy. |
|---|---|
| ISSN: | 2666-8270 |