Memory leaks in Android applications generally occur because objects are still strongly referenced by GC roots after a component has been destroyed, leading to continuous growth of heap memory, frequent garbage collection, UI freezing, and eventually memory overflow. Given the shortcomings of the previous method, such as lacking lifecycle semantics, limited dynamic evidence coverage, and no risk constraints on remediation strategies, this paper proposes a lifecycle-reference chain collaborative LC-Hybrid method: record component states as event sequences, calculate retention paths with a heap reference graph, identify asymmetric releases of listeners, tasks, contexts, and heavy resources using static rules, and finally form tiered remediation recommendations through evidence fusion scoring. Based on the previous three years' public research and benchmark data, a supporting secondary statistics table has been prepared. According to the results from DroidLeaks, InferROI is a large-scale model intent inference and static analysis tool with a defect detection rate of 59.3% and an F1 score of 0.667; after fixing the problems, the total heap memory in the LeakPair public project decreased from 830.3 MB to 734.0 MB, and the project-level pairing showed significant improvement in the Wilcoxon test. Based on the above, this paper proposes an all-encompassing optimisation process of discovery, localisation and remediation, followed by regression validation, and clarifies its scope.

Wenhao Song, Research on Memory Leakage Detection and Optimization Methods for Android Applications. International Journal of Social Sciences and Economic Management (2026), Vol. 7, Issue 2: 58-66. https://doi.doi.org/10.3807/IJSSEM.2026.070206.

[1] Liao D, Pan S, Yang S, Wang Y. Automatically Analyzing Performance Issues in Android Apps: How Far Are We? arXiv:2407.05090, 2024. DOI: 10.48550/arXiv.2407.05090.

[2] Rua R, Saraiva J. A Large-Scale Empirical Study on Mobile Performance: Energy, Run-Time and Memory. Empirical Software Engineering, 2024, 29: 31. DOI: 10.1007/s10664-023-10391-y.

[3] Mahmud T, Che M, Ngu AHH, Yang G. Why Android App Testing Falls Short: Empirical Insights from Open-Source Projects and a Practitioner Survey. Empirical Software Engineering, 2025, 30: 163. DOI: 10.1007/s10664-025-10726-x.

[4] Murali A, Alfadel M, Nagappan M, Xu M, Sun C. AddressWatcher: Sanitizer-Based Localization of Memory Leak Fixes. IEEE Transactions on Software Engineering, 2024, 50(9): 2398–2411. DOI: 10.1109/TSE.2024.3438119.

[5] Guo, X. (2025, April). Research on Financial Trading Algorithms Based on Deep Reinforcement Learning. In 2025 IEEE 3rd International Conference on Control, Electronics and Computer Technology (ICCECT) (pp. 1711-1715). IEEE.

[6] Guo, X. (2025, March). Research on Blockchain-Based Financial AI Algorithm Integration Methods and Systems. In 2025 IEEE International Conference on Electronics, Energy Systems and Power Engineering (EESPE) (pp. 816-821). IEEE.

[7] Han, W. (2025, November). An Empirical Study on Building a High-precision Forecasting System for Financial Market Trends Based on the PyTorch Deep Learning Framework. In International Conference on Computational Technologies for Research in Data Analytics (pp. 518-528). Cham: Springer Nature Switzerland.

[8] Jing, X. (2025, November). In-Depth Analysis of the Optimization Decision Model of Financial Investment Portfolio Based on Machine Learning Algorithms in Python Scikit-Learn Library. In International Conference on Computational Technologies for Research in Data Analytics (pp. 130-140). Cham: Springer Nature Switzerland.