NORMA eResearch @NCI Library

Novel Text and Image Based Approach to Android Malware Detection

Pal, Agnideep (2023) Novel Text and Image Based Approach to Android Malware Detection. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


This research examined two unique, image and text-based malware detection strategies before proposing novel classification techniques. It is clear from previous research in the field that feature engineering is crucial for producing high-quality training datasets. This study mainly focused on feature engineering techniques for proper classification of android malwares with the data obtained from either Dex or manifest files, or both. While reshaping the malware detection challenge to a text processing problem, a novel approach was identified, proposed and tested. Along with it, another innovative technique was developed in the image-based approach. Both the approaches demonstrated encouraging results in terms of classifier accuracy, precision, recall, and F1 score. Although, the dataset was highly unbalanced, the text-based models attained approx. 95% accuracy. On the other hand, using the same dataset, the image-based models had an accuracy rate of about 84%.

Item Type: Thesis (Masters)
Simiscuka, Anderson Augusto
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing
Q Science > QA Mathematics > Computer software > Computer Security
T Technology > T Technology (General) > Information Technology > Computer software > Computer Security
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 23 May 2023 16:02
Last Modified: 23 May 2023 16:02

Actions (login required)

View Item View Item