Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{177415,
        author = {Prasanna Kumari.Posina and Mr. Yerrabathana Guravaiah},
        title = {ANDROID AUTOMATED MALWARE DETECTION BASED ON INFROMATION SECURITY},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {12},
        pages = {555-560},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=177415},
        abstract = {With the exponential rise in Android-based smartphone usage, securing mobile devices against malware has become a paramount concern. Traditional signature-based detection mechanisms are often inefficient against rapidly evolving and zero-day threats. This project proposes an automated malware detection system for Android applications, rooted in core principles of information security. The system uses static and dynamic analysis techniques, enhanced with machine learning, to identify anomalous behavior and malicious code patterns in Android apps. By analyzing permissions, API calls, and network behaviors, this approach offers a proactive and adaptive security mechanism. The automated system ensures real-time threat detection and aims to minimize false positives, offering a robust and scalable solution to Android malware threats. Because of its open-source nature and the support it receives from Google, the Android platform currently holds the biggest market share worldwide. Since it is the most widely used operating system in the world, it has attracted the attention of cybercriminals, notably through the widespread distribution of malicious programs. The purpose of this research is to offer an effective machine-learning and deep-learning-based solution for Android Malware Detection. The approach makes use of an evolutionary chi-square algorithm for discriminatory feature selection. For the purpose of training machine learning and deep learning classifiers, selected features from the chi-square algorithm are utilized. The competence of these classifiers to identify malware depending on feature selection is then compared. It has been demonstrated through the results of the experiments that the chi-square method provides the best-optimized feature subset, which assists in the reduction of the feature dimension to less than half of the initial feature set. Machine learning and deep learning-based classifiers, such as RF, ETC, ANN, and CNN, are able to maintain a classification accuracy that is higher than the previous percentage after feature selection. This is accomplished while working on a significantly reduced feature dimension, which positively impacts the computational complexity of learning classifiers. When compared to machine-learning models, deep-learning models have demonstrated the highest level of accuracy in detecting malware. This conclusion is based on the evaluations of our experimental models.},
        keywords = {RF, ETC, ANN, CNN, Malware},
        month = {May},
        }