Permission-Based Classification of Android Malware Applications Using Random Forest
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Android is arguably the most widely used mobile operating system in the world. Due to its widespread use, it has attracted a lot of attention of cybercriminals who attempt to exploit its architecture and outsmart innocent users to install malware applications. The number of such applications is growing every day either by alternating a basic exploitation mechanism or by creating novel mechanisms to exfiltrate users’ data. As a result, there is an increasing need for detection mechanisms that can classify these applications to families based on their characteristics. A significant amount of research has already been devoted to analysing and mitigating this growing problem; however, this situation demands more efficient methods with higher precision. The paper proposes such a framework for analysing and classifying a malicious application to certain families relying on the permissions used. The proposed method involves the pre-processing of the applications to extract their permissions, the tokenization of permissions, the data cleansing and finally the application of the Random Forest Classifier to classify the applications in families. The proposed method is trained and tested with a dataset of 11,159 malicious applications categorized in 33 unique families. The precision, recall and f1-score achieved is 98%. The results of the proposed methodology are promising, since it even works in an unbalanced dataset and in many cases outperform other state-of-the-art approaches.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it