Statistical Ranking: A Voting-Based Ensemble Approach to Feature Selection in Android Malware Detection

• Anna Luiza Gomes da Silva UNIPAMPA
• Lucas Ferreira Areias de Oliveira UNIPAMPA
• Angelo Diniz UNIPAMPA
• Diego Kreutz UNIPAMPA https://orcid.org/0000-0003-0830-0238
• Dionatan R. Schmidt UNIPAMPA
• Rodrigo Mansilha UNIPAMPA https://orcid.org/0000-0002-2083-653X
• Kayuã Oleques Paim UFRGS https://orcid.org/0000-0002-3394-795X

DOI: https://doi.org/10.5753/sbseg.2026.29203

Resumo

High-dimensional Android malware datasets inflate training time and damage model generalization. We propose Statistical Ranking, an ensemble feature selection framework that enforces multi-perspective consensus: a feature is retained only when ranked in the top-k by at least two of three independent criteria (χ2, Mutual Information, and Random Forest importance), under a percriterion budget that scales with the original dimensionality. Across 11 public datasets under stratified 5-fold cross-validation, the method removes 89.8%–99.6% of the feature space with a median recall drop of 0.04, retaining recall ≥ 0.85 in 9 of 11 environments. An ablation over ranking criteria and voting thresholds shows that the ≥ 2-of-3 consensus maximizes the reduction, recall trade-off, and that the contribution of a third ranker depends on the redundancy between the marginal criteria, which varies with feature modality. On MH100K, 24,833 features are reduced to 93 (99.6%) at a cost of 6.5% in F1-score, and the one-off cost of selection is recovered after roughly eight retraining cycles, yielding a 27% reduction in total cost over a ten-update horizon.

Referências

Abdo, A., Mostafa, R., and Abdel-Hamid, L. (2024). An optimized hybrid approach for feature selection based on chi-square and particle swarm optimization algorithms. Data, 9(2).

Alalhareth, M. and Hong, S.-C. (2023). An improved mutual information feature selection technique for intrusion detection systems in the internet of medical things. Sensors, 23(10).

Bragança, H., Kreutz, D., Rocha, V., Assolin, J., and Feitosa, E. (2026). MH-1M: A 1.34 million-sample multi-feature android malware dataset with rich metadata. Scientific Data, 13:153.

da Silva, A. R. and Silveira, C. G. (2022). Handcrafted feature selection techniques for pattern recognition: A survey.

Dey, S. K., Uddin, K. M. M., Babu, H. M. H., Rahman, M. M., Howlader, A., and Uddin, K. A. (2022). Chi2-mi: A hybrid feature selection based machine learning approach in diagnosis of chronic kidney disease. Intelligent Systems with Applications, 16:200144.

Iranzad, R. and Liu, X. (2025). A review of random forest-based feature selection methods for data science education and applications. International Journal of Data Science and Analytics, 20(2):197–211.

Jain, S., Goyal, H., Arora, A., and Kumar, D. (2026). Enfestdroid: Ensembled feature selection techniques based android malware detection. Computers and Electrical Engineering, 129:110763.

Paim, K., Nogueira, A., Kreutz, D., Cordeiro, W., and Mansilha, R. (2025). Maldatagen: A modular framework for synthetic tabular data generation in malware detection. In Anais Estendidos do XXV Simpósio Brasileiro de Cibersegurança, pages 38–47, Porto Alegre, RS, Brasil. SBC.

Pudjihartono, N., Fadason, T., Kempa-Liehr, A. W., and O’Sullivan, J. M. (2022). A review of feature selection methods for machine learning-based disease risk prediction. Frontiers in Bioinformatics, 2.

Rocha, V., Kreutz, D., Bragança, H., and Feitosa, E. (2026a). Limitless feature selection: Revolutionizing evaluation with mh-fsf. Journal of the Brazilian Computer Society, 32(1):73–84.

Rocha, V., Tschiedel, L., Kreutz, D., Bragança, H., Assolin, J., Mansilha, R. B., Quincozes, S. E., and Nogueira, A. G. D. (2026b). Generalizing feature selection in android malware detection: The sigapi autocraft approach. Journal of the Brazilian Computer Society, 32(1):250–263.

Sharma, Y. and Arora, A. (2025). Corrnetdroid: Android malware detector leveraging a correlation-based feature selection for network traffic features.

Theng, D. and Bhoyar, K. K. (2024). Feature selection techniques for machine learning: a survey of more than two decades of research. Knowledge and Information Systems, 66(3):1575–1637.

Tisoc, M. and Beltran, J. (2022). Mutual information and correlations. Revista de la Facultad de Ciencias, Universidad Nacional de Ingenieria. Received on February 14, 2022; Revised on May 25, 2022; Accepted on July 11, 2022.

Venkatesh, B. and Anuradha, J. (2019). A review of feature selection and its methods. Cybernetics and information technologies, 19(1):3–26.

Wu, Y., Li, M., Zeng, Q., Yang, T., Wang, J., Fang, Z., and Cheng, L. (2023). Droidrl: Feature selection for android malware detection with reinforcement learning. Computers and Security, 128:103126.