App Risk Analyzer: Android-Based Application for Risk Scoring and Safe Uninstallation

Abstract

The exponential growth of the Android ecosystem has led to an unprecedented increase in the number of mobile applications available to users. While this expansion has enhanced functionality and accessibility, it has also introduced critical security and privacy challenges. Many Android applications request permissions that grant access to sensitive user data such as contacts, messages, location, camera, and storage. In most cases, users approve these permissions without fully understanding their implications, thereby exposing themselves to potential threats such as data leakage, unauthorized tracking, and malicious activities. This research presents the design and implementation of an App Risk Analyzer, an Android-based application that evaluates installed applications and assigns a quantitative risk score based on both static and dynamic parameters. The proposed system primarily focuses on permission-based analysis combined with lightweight behavioural insights to determine the level of risk associated with each application. Unlike traditional antivirus solutions that rely on signature-based detection, the proposed model emphasizes user awareness and interpretability, transforming complex security data into an intuitive scoring mechanism. The system introduces a structured methodology where permissions are categorized based on sensitivity and assigned weighted values. Additionally, behavioural attributes such as background execution, resource consumption, and network activity are incorporated to enhance accuracy. One of the key features of the application is the direct uninstall functionality, which allows users to immediately remove potentially harmful applications without navigating through complex system settings. Experimental evaluation demonstrates that the proposed system effectively identifies high-risk applications while maintaining low computational overhead. The results indicate improved user awareness, better decision-making, and enhanced mobile security. The proposed solution bridges the gap between technical risk assessment and user-friendly interaction, making it a practical tool for real-world deployment.

Introduction

The widespread adoption of Android smartphones has improved mobile computing but has also increased security risks due to the platform’s open-source nature and widespread use of third-party applications. Although Android employs a permission-based security model, users frequently grant permissions without careful review, allowing malicious applications to access sensitive data. Traditional antivirus solutions mainly rely on signature-based detection, making them ineffective against new or evolving threats while providing limited transparency and requiring users to manually remove harmful applications. To address these shortcomings, this paper proposes an App Risk Analyzer that evaluates installed applications using a hybrid approach combining permission analysis and lightweight behavioural monitoring. The system assigns each application a clear risk score, presents the results through an intuitive interface, and includes a direct uninstall feature to help users quickly mitigate security risks.

The implementation follows a three-layer modular architecture consisting of a Data Collection Layer, Risk Analysis Layer, and User Interface Layer. The data collection module uses the Android Package Manager API to gather information about installed applications, including permissions, installation details, and metadata, while ensuring all processing is performed locally to preserve user privacy. The risk analysis engine categorizes permissions into high, medium, and low-risk groups, assigning weighted values based on their sensitivity. To improve detection accuracy, a behavioural analysis module monitors lightweight runtime indicators such as CPU usage, battery consumption, background activity, and network traffic using Android system services. These behavioural insights are combined with permission weights to compute a normalized risk score ranging from 1 to 5, classifying applications as low, medium, or high risk.

The user interface is designed to enhance usability and transparency by displaying installed applications with color-coded risk indicators, detailed explanations of permission contributions and behavioural anomalies, and a direct uninstall option that simplifies the removal of potentially harmful applications. Performance optimization techniques, including asynchronous processing, caching, and efficient memory management, ensure real-time analysis with minimal impact on device resources. The system also prioritizes security and privacy by performing all analysis locally and requesting only essential permissions.

Experimental evaluation on multiple Android devices demonstrated that the App Risk Analyzer effectively distinguishes applications based on their risk levels while maintaining analysis times below two seconds per application and low resource consumption. Behavioural monitoring improved detection accuracy by identifying suspicious activities not evident through permission analysis alone. User feedback indicated that the intuitive interface, transparent risk explanations, and integrated uninstall feature significantly enhanced awareness and encouraged proactive security practices.

Future enhancements include integrating machine learning for adaptive risk prediction, cloud-based threat intelligence for real-time malware detection, advanced behavioural and network traffic analysis, real-time security alerts, community-driven risk evaluation, personalized recommendations, cross-platform support, and visual analytics dashboards. These improvements aim to create a more intelligent, adaptive, and comprehensive mobile security solution capable of addressing emerging Android security threats while maintaining high usability and user trust.

Conclusion

The increasing reliance on mobile applications in everyday life has made security and privacy critical concerns for users. The Android platform, while offering flexibility and accessibility, presents unique challenges due to its open nature and permission-based architecture. Many users unknowingly grant access to sensitive data, making them vulnerable to potential threats. Addressing this issue requires solutions that not only detect risks but also present them in a manner that is understandable and actionable for users. The App Risk Analyzer proposed in this research provides an effective solution to this problem by introducing a user-centric approach to mobile security. Unlike traditional antivirus systems that focus primarily on detecting known malware, this system emphasizes risk awareness and transparency. By analysing application permissions and behavioural characteristics, the system generates a quantitative risk score that simplifies complex security information into an intuitive format. One of the key strengths of the proposed system is its ability to balance accuracy, efficiency, and usability. The use of permission-based analysis ensures low computational overhead, making the system suitable for real-time mobile environments. At the same time, the incorporation of behavioural indicators enhances the depth of analysis, enabling the detection of potential risks that may not be evident through static methods alone. The inclusion of a direct uninstall feature further distinguishes the system from existing solutions. By enabling users to take immediate action against high-risk applications, the system reduces the gap between risk detection and risk mitigation. This feature enhances user control and promotes proactive security practices. The experimental evaluation demonstrates that the system effectively identifies high-risk applications and provides meaningful insights to users. The intuitive interface and clear risk categorization improve user understanding and encourage informed decision-making. These results highlight the importance of designing security solutions that prioritize both technical effectiveness and user experience. Despite its advantages, the system has certain limitations, such as reliance on predefined permission weights and limited behavioural analysis. However, these limitations also present opportunities for future enhancement, as discussed in the previous section. By integrating advanced technologies such as machine learning and cloud-based intelligence, the system can be further improved to address more complex security challenges.

References

[1] Anuradha Dahiya, Sukhdip Singh, Gulshan Shrivastava. ‘Risk-Oriented Taxonomy of Android App Assessment Models’Revolutionary Advance in Computing and Electronics(RACE)Volume 1, Dec 2025. [2] Singh, A., & Singh, K. ‘Permission-Based Android Malware Detection’Indian Journal of Computer Science, 10(5), 33–52. 2025 [3] Patel, R., & Gupta, ‘A Comprehensive Framework for Android Application Risk Analysis Using Permission and Behavior Features’ S. Journal of Mobile Security Research, 12(2), 101–118.2025 [4] Chen, L., & Zhang, Y. ‘Advanced Permission Misuse Detection Strategies in Android Platforms’International Journal of Cybersecurity and Mobile Computing, 8(1), 45–63. [5] ‘Efficient Risk Analysis for Android Applications’Manoj T. Joy2015 IEEE Recent Advances in Intelligent Computational Systems (RAICS) | 10-12 December 2015. [6] S. Maganur, Y. Jiang, J. Huang, and F. Zhong, ‘Feature-Centric Approaches to Android Malware Analysis: A Survey’, Computers, vol. 14, no. 11, p. 482, Nov. 2025, doi: 10.3390/computers14110482. [7] A. Ruggia, D. Nisi, S. Dambra, A. Merlo, D. Balzarotti, and S. Aonzo, ‘Unmasking the Veiled: A Comprehensive Analysis of Android Evasive Malware’, in Proceedings of the 19th ACM Asia Conference on Computer and Communications Security, Singapore: ACM, Jul. 2024, pp. 383–398, doi: 10.1145/3634737.3637658. [8] A. Kar, N. Stakhanova, and E. Branca, ‘Detecting Overlay Attacks in Android’, Procedia Computer Science, vol. 231, pp. 137–144, 2024, doi: 10.1016/j.procs.2023.12.185. [9] A. Dahiya, S. Singh, and G. Shrivastava, ‘Android malware analysis and detection: A systematic review’, Expert Systems, p. e13488, Oct. 2023, doi: 10.1111/exsy.13488. [10] Q. Wu, X. Zhu, and B. Liu, ‘A Survey of Android Malware Static Detection Technology Based on Machine Learning’, Mobile Information Systems, vol. 2021, pp. 1–18, Mar. 2021, doi: 10.1155/2021/8896013. [11] Z. Qu, S. Alam, Y. Chen, X. Zhou, W. Hong, and R. Riley, ‘DyDroid: Measuring Dynamic Code Loading and Its Security Implications in Android Applications’, in 2017 47th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN), Jun. 2017, pp. 415–426, doi: 10.1109/DSN.2017.14. [12] A. Dahiya, S. Singh, and G. Shrivastava, ‘Lightweight and Efficient Android Malware Detection Using Manifest File Analysis’, in 2025 International Conference on Networks and Cryptology (NETCRYPT), May 2025, pp. 1246–1251, doi: 10.1109/NETCRYPT65877.2025.11102561. [13] T. Sutter, T. Kehrer, M. Rennhard, B. Tellenbach, and J. Klein, ‘Dynamic Security Analysis on Android: A Systematic Literature Review’, IEEE Access, vol. 12, pp. 57261–57287, 2024, doi: 10.1109/ACCESS.2024.3390612. [14] B. Kondracki, B. A. Azad, N. Miramirkhani, and N. Nikiforakis, ‘The Droid is in the Details: Environment-aware Evasion of Android Sandboxes’, in Proceedings 2022 Network and Distributed System Security Symposium, 2022, doi: 10.14722/ndss.2022.23056. [15] A. Dahiya, S. Singh, and G. Shrivastava, ‘Malware Detection Insights, Mechanisms and Future Perspectives for Android Applications’, in Innovative Computing and Communications, vol. 1021, Singapore: Springer Nature Singapore, 2024, pp. 381–403, doi: 10.1007/978-981-97-3591-4_31. [16] M. Choudhary and B. Kishore, ‘HAAMD: Hybrid Analysis for Android Malware Detection’, in 2018 International Conference on Computer Communication and Informatics (ICCCI), Jan. 2018, pp. 1–4, doi: 10.1109/ICCCI.2018.8441295.

Copyright

Copyright © 2026 Mohit Vaidya, Rudresh Sarode, Moksh Chandekar, Khushi Hajare, Kshama Bobade, Nikhil Shende, Mr. Amol Dhankar. This 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.