DFRMIdroid: A Comprehensive Fusion Approach Utilizing Permissions and Intents Analysis with the DFR-MI Algorithm for Enhanced Malware Detection on Android Devices
Bibliographic record
Abstract
Smartphones based on the Android operating system are increasingly popular due to their multifunctional capabilities in various fields.However, these functions have also encouraged intruders to develop applications that perform malicious actions, including stealing sensitive data, encrypting files for ransom, and sending unauthorized SMS messages without user consent.In this study, we proposed a new and comprehensive dataset containing 32,170 samples distributed equally between malicious and benign applications developed on different API levels.The dataset includes two categories of features: permissions and intents.We also proposed a new feature selection method called Discriminative Feature Ranking-Mutual Information (DFR-MI) which selects optimal features from the more frequent features in the dataset and this helped the predictive model to achieve high performance.Nine machine-learning algorithms were tested, and the results show that our dataset outperforms the Drebin dataset by at least 2.22% in combining permissions with intents for detecting malicious apps.Additionally, the DFR-MI algorithm obtained better results in selecting features and took less time than the mutual information algorithm.Among all tested machine-learning algorithms, the random forest algorithm achieved high scores in terms of accuracy, precision, recall, and F1 score, which were 98.52%, 98.62%, 98.41%, and 98.52%, respectively.Our proposed method enhances mobile security by scrutinizing an app's declared permissions and communication patterns between its components.This approach allows for a more comprehensive understanding of an app's behavior, enabling early detection of potential threats generated from Android applications.
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.
How this classification was reachedexpand
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 | 0.002 |
| Science and technology studies | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".