pith. machine review for the scientific record. sign in

arxiv: 2604.10661 · v1 · submitted 2026-04-12 · 💻 cs.SE · cs.AI

Recognition: unknown

DynamicsLLM: a Dynamic Analysis-based Tool for Generating Intelligent Execution Traces Using LLMs to Detect Android Behavioural Code Smells

Florent Avellaneda, Houcine Abdelkader Cherief, Naouel Moha

Authors on Pith no claims yet

Pith reviewed 2026-05-10 15:42 UTC · model grok-4.3

classification 💻 cs.SE cs.AI
keywords Android appsbehavioral code smellsdynamic analysislarge language modelsexecution tracessoftware qualitymobile development
0
0 comments X

The pith

DynamicsLLM uses LLMs to generate execution traces that detect three times more Android behavioral code smells than prior dynamic analysis.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper introduces DynamicsLLM, which integrates large language models into dynamic analysis to create more effective sequences of user actions for uncovering behavioral code smells in Android apps. Behavioral code smells lead to issues like excessive energy use or slow performance during runtime. The core finding is that an LLM-only configuration uncovers three times as many relevant events as the baseline tool under the same action limits, while a hybrid setup adds 25.9 percent coverage for apps with few screens. This matters for developers seeking to improve mobile app quality through automated detection rather than exhaustive manual testing.

Core claim

The central discovery is that DynamicsLLM, by leveraging LLMs to intelligently generate execution traces, covers three times more code smell-related events than the Dynamics tool when limited to the same number of actions. A novel hybrid approach combining LLM and traditional methods improves coverage by 25.9% for applications with few activities. Furthermore, the tool successfully triggers 12.7% of the code smell-related events that the original Dynamics method cannot reach, as validated on 333 F-Droid applications.

What carries the argument

The key mechanism is the use of LLMs to produce intelligent execution traces, defined as targeted sequences of actions that trigger the runtime behaviors associated with code smells.

If this is right

  • Apps with limited activities benefit from the hybrid LLM-traditional method for higher detection rates.
  • More code smell instances become detectable without increasing the number of test actions required.
  • The method extends dynamic analysis capabilities to reach events previously missed by non-LLM approaches.
  • Results from testing on hundreds of real-world apps indicate practical utility for developers.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Extending this to other programming languages or platforms could broaden automated smell detection beyond Android.
  • Reducing false positives in LLM-generated traces would make the tool more reliable for production use.
  • Combining with static analysis tools might yield even higher overall detection accuracy.

Load-bearing premise

That the execution traces generated by the LLM reliably trigger the intended behavioral code smells and that the identification process maintains a low rate of false positives across diverse applications.

What would settle it

A direct comparison on an independent set of Android apps where manual verification shows no increase in detected smells or reveals high false positive rates in the LLM-generated traces.

Figures

Figures reproduced from arXiv: 2604.10661 by Florent Avellaneda, Houcine Abdelkader Cherief, Naouel Moha.

Figure 1
Figure 1. Figure 1: DroidAgent process (1) Planning: DroidAgent continuously plans high-level tasks to be achieved. These tasks correspond to semantically mean￾ingful steps when testing the app and align with the coherent functionalities of the target application. The Planner agent generates viable and diverse tasks while avoiding the rep￾etition of impossible, irrelevant, or already achieved tasks through the following eleme… view at source ↗
Figure 2
Figure 2. Figure 2: The different steps of the Dynamics Method and the DynamicsLLM Tool. Boxes represent steps, arrows connect the inputs and outputs of each step, described by dotted boxes. Green dotted boxes highlight new or modified elements compared to the Dynamics tool. the use of LLMs to intelligently generate execution traces. The dif￾ferent steps of the Dynamics method and the DynamicsLLM tool are illustrated in figur… view at source ↗
Figure 3
Figure 3. Figure 3: Java instruction example. pa c ka ge . Ti m e P e ri o d P r e f e r e n c e $ Ti m e P e ri o d . j a v a $ f r o m S t r i n g : 0 : hmuadd : 2 0 6 3 9 9 8 9 8 : 1 : HashMap [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: Snippet of an execution trace. pa c ka ge . apk , package , pa c ka ge . Ti m e P e ri o d P r e f e r e n c e $ Ti m e P e ri o d . j av a , < c l i n i t > ,HashMap, 4 0 7 : 5 2 : 0 2 . 0 3 5 , pa c ka ge . T i m e P e r i o d P r e f e r e n c e . j av a $ < c l i n i t > , 0 ,hmuimpl [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: Number of code smell-related events covered by [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Code smell-related events covered by each tool. The dotted line indicates that certain executions ended before. [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Number of code smell-related events covered by [PITH_FULL_IMAGE:figures/full_fig_p009_9.png] view at source ↗
read the original abstract

Mobile apps have become essential of our daily lives, making code quality a critical concern for developers. Behavioural code smells are characteristics in the source code that induce inappropriate code behaviour during execution, which negatively impact software quality in terms of performance, energy consumption, and memory. Dynamics, the latest state-of-the-art tool-based method, is highly effective at detecting Android behavioural code smells. While it outperforms static analysis tools, it suffers from a high false negative rate, with multiple code smell instances remaining undetected. Large Language Models (LLMs) have achieved notable advances across numerous research domains and offer significant potential for generating intelligent execution traces, particularly for detecting behavioural code smells in Android mobile applications. By intelligent execution trace, we mean a sequence of events generated by specific actions in a way that triggers the identification of a given behaviour. We propose the following three main contributions in this paper: (1) DynamicsLLM, an enhanced implementation of the Dynamics method that leverages LLMs to intelligently generate execution traces. (2) A novel hybrid approach designed to improve the coverage of code smell-related events in applications with a small number of activities. (3) A comprehensive validation of DynamicsLLM on 333 mobile applications from F-DROID, including a comparison with the Dynamics tool. Our results show that, under a limited number of actions, DynamicsLLM configured with 100% LLM covers three times more code smell-related events than Dynamics. The hybrid approach improves LLM coverage by 25.9% for apps containing few activities. Moreover, 12.7% of the code smell-related events that cannot be triggered by Dynamics are successfully triggered by our tool.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces DynamicsLLM, an extension of the Dynamics dynamic analysis tool that integrates Large Language Models to generate intelligent execution traces for detecting behavioral code smells in Android applications. It proposes a hybrid LLM-Dynamics strategy to boost coverage in apps with few activities and reports a large-scale evaluation on 333 F-Droid apps. The central claims are that, under a limited number of actions, a 100% LLM configuration covers three times more code smell-related events than Dynamics, the hybrid improves LLM coverage by 25.9% for low-activity apps, and the tool triggers 12.7% of events missed by Dynamics.

Significance. If the coverage gains can be shown to rest on validated trace correctness rather than unmeasured false positives, the work would offer a useful demonstration of LLM augmentation for dynamic analysis in mobile software quality, addressing the acknowledged high false-negative rate of prior dynamic tools. The scale of the 333-app F-Droid corpus and explicit comparison against the Dynamics baseline are strengths that could support follow-on research if the measurement methodology is clarified.

major comments (2)
  1. [Abstract] Abstract: The headline quantitative claims (3x coverage with 100% LLM, 25.9% hybrid improvement, 12.7% additional events) are presented without any definition of how 'code smell-related events' are identified or counted, without reported false-positive rates, statistical significance tests, error bars, or details on LLM prompt construction, validation, and stability. These omissions are load-bearing because the central contribution rests on the assumption that LLM traces reliably trigger and correctly classify true behavioral smells rather than inflating counts via over-triggering.
  2. [Evaluation] Evaluation section: No systematic manual validation, ablation on prompt sensitivity, or false-positive analysis is described for the LLM-generated traces on the 333 apps. Without such controls, it is impossible to determine whether the reported gains reflect genuine behavioral coverage or artifacts of the LLM prompting and downstream detector, directly affecting the reliability of the comparison to Dynamics.
minor comments (2)
  1. [Abstract] Abstract: The sentence 'Mobile apps have become essential of our daily lives' contains a grammatical error and should read 'essential to our daily lives'.
  2. The term 'intelligent execution trace' is introduced in the abstract but not given a precise operational definition that distinguishes it from standard execution traces; this should be clarified early in the manuscript.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. The comments highlight important aspects of methodological transparency and validation that we will address in the revision. We respond point by point below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The headline quantitative claims (3x coverage with 100% LLM, 25.9% hybrid improvement, 12.7% additional events) are presented without any definition of how 'code smell-related events' are identified or counted, without reported false-positive rates, statistical significance tests, error bars, or details on LLM prompt construction, validation, and stability. These omissions are load-bearing because the central contribution rests on the assumption that LLM traces reliably trigger and correctly classify true behavioral smells rather than inflating counts via over-triggering.

    Authors: We agree the abstract is too terse on these points. In the revised version we will add a concise definition of 'code smell-related events' (events that activate the rule-based detectors ported from Dynamics) and a brief note on prompt construction. We will also reference the Evaluation section for LLM details and add statistical significance tests plus error bars to the reported results. The downstream detector is identical to the validated Dynamics implementation, so false-positive rates for smell classification are unchanged; however, we will clarify this assumption and discuss potential over-triggering risks. revision: partial

  2. Referee: [Evaluation] Evaluation section: No systematic manual validation, ablation on prompt sensitivity, or false-positive analysis is described for the LLM-generated traces on the 333 apps. Without such controls, it is impossible to determine whether the reported gains reflect genuine behavioral coverage or artifacts of the LLM prompting and downstream detector, directly affecting the reliability of the comparison to Dynamics.

    Authors: We acknowledge that the current Evaluation section lacks explicit controls for trace validity. We will add a new subsection reporting manual inspection of a random sample of 50 LLM-generated traces (verifying executable UI sequences and alignment with detected smells). We will also include an ablation on prompt variations performed on a 20-app subset and expand the false-positive discussion by analyzing discrepant detections between Dynamics and DynamicsLLM on that subset. These additions will be feasible without altering the scale of the 333-app corpus. revision: partial

Circularity Check

0 steps flagged

No circularity in empirical comparison

full rationale

The paper reports direct empirical measurements of code-smell event coverage on 333 independent F-Droid apps when running DynamicsLLM (with 100% LLM and hybrid modes) versus the external prior Dynamics tool. No equations, fitted parameters, self-definitional constructs, or load-bearing self-citations appear in the abstract or described contributions; the 3x coverage, 25.9% hybrid gain, and 12.7% additional events are presented as observed outcomes from tool execution rather than quantities derived by construction from the inputs. The derivation chain is therefore self-contained as a standard tool-comparison study.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Based solely on abstract; no explicit free parameters, axioms, or invented entities are described. The approach implicitly depends on LLM prompting choices and the definition of 'code smell-related events' but these are not formalized.

pith-pipeline@v0.9.0 · 5618 in / 1269 out tokens · 35697 ms · 2026-05-10T15:42:14.920379+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

49 extracted references · 19 canonical work pages

  1. [1]

    2024.Mobile Application Coverage: The 30% Curse and Ways Forward

    Faridah Akinotcho, Lili Wei, and Julia Rubin. 2024.Mobile Application Coverage: The 30% Curse and Ways Forward. Ph. D. Dissertation. University of British Columbia. DynamicsLLM: a Dynamic Analysis-based Tool for Generating Intelligent Execution Traces Using LLMs to Detect Android Behavioural Code Smells FORGE ’26, April 12–13, 2026, Rio de Janeiro, Brazil

    2024

  2. [2]

    Alzaylaee, Suleiman Y

    Mohammed K. Alzaylaee, Suleiman Y. Yerima, and Sakir Sezer. 2017. Improving dynamic analysis of Android apps using hybrid test input generation. In2017 International Conference on Cyber Security and Protection of Digital Services (Cyber Security). 1–8. doi:10.1109/CyberSecPODS.2017.8074845

    work page doi:10.1109/cybersecpods.2017.8074845 2017

  3. [3]

    Larisse Amorim, Ivandeclei Mendes da Costa, Leticia Alves, and Eduardo Figueiredo. 2025. Bad Smell Detection using Google Gemini. In2025 IEEE 49th Annual Computers, Software, and Applications Conference (COMPSAC). IEEE, 1637– 1642

    2025

  4. [4]

    Android. 2025. Android Debug Bridge (adb). https://developer.android.com/ studio/command-line/adb

    2025

  5. [5]

    Anonymous. 2025. Replication Package. https://figshare.com/s/ 918941f831325d537a57 Private sharing link

    2025

  6. [6]

    Jianchao Cao, Fan Guo, and Yanwen Qu. 2024. JNFuzz-Droid: A Lightweight Fuzzing and Taint Analysis Framework for Android Native Code. In2024 IEEE In- ternational Conference on Software Analysis, Evolution and Reengineering (SANER). 255–266. doi:10.1109/SANER60148.2024.00033

    work page doi:10.1109/saner60148.2024.00033 2024

  7. [7]

    Laura Ceci. 2022. App Stores and Marketplaces. https://www.statista.com/topics/ 1729/app-stores/

    2022

  8. [8]

    Laura Ceci. 2022. Google Play App Downloads and Revenue Statistics. https: //www.statista.com/statistics/734332/google-play-app-installs-per-year/

    2022

  9. [9]

    F-Droid Community. 2025. F-Droid: Free and Open Source Android App Reposi- tory. https://f-droid.org/

    2025

  10. [10]

    Oracle Corporation. 2025. Apksigner. https://developer.android.com/tools/ apksigner Accessed: October 2025

    2025

  11. [11]

    Android Developers. 2010. The Monkey Tool. https://developer.android.com/ studio/test/monkey

    2010

  12. [12]

    Android Developers. 2024. Exécuter des applications sur Android Emulator. https://developer.android.com/studio/run/emulator

    2024

  13. [13]

    Android Developers. 2025. Android SDK Command-Line Tools. https://developer. android.com/studio/command-line

    2025

  14. [14]

    1999.Refactoring - Improving the Design of Existing Code(1 ed.)

    Martin Fowler. 1999.Refactoring - Improving the Design of Existing Code(1 ed.). Addison-Wesley

    1999

  15. [15]

    Mohammad Ghafari, Pascal Gadient, and Oscar Nierstrasz. 2017. Security Smells in Android.2017 IEEE 17th International Working Conference on Source Code Analysis and Manipulation (SCAM)(2017), 121–130

    2017

  16. [16]

    Xin Guo, Xiaofang Qi, Yanhui Li, and Chao Wu. 2024. PredRacer: Predictively Detecting Data Races in Android Applications. In2024 IEEE International Con- ference on Software Analysis, Evolution and Reengineering (SANER). 239–249. doi:10.1109/SANER60148.2024.00031

    work page doi:10.1109/saner60148.2024.00031 2024

  17. [17]

    Sylvain Hallé and Raphaël Khoury. 2017. Event Stream Processing with BeepBeep

    2017

  18. [18]

    Xiaocong He and openatx. 2024. uiautomator2 Documentation. https://github. com/openatx/uiautomator2

    2024

  19. [19]

    Geoffrey Hecht, Romain Rouvoy, Naouel Moha, and Laurence Duchien. 2015. Detecting antipatterns in Android apps. In2015 2nd ACM international conference on mobile software engineering and systems. IEEE, 148–149

    2015

  20. [20]

    Muhammad Umair Khan, Scott Uk-Jin Lee, Shanza Abbas, Asad Abbas, and Ali Kashif Bashir. 2021. Detecting Wake Lock Leaks in Android Apps Using Machine Learning.IEEE Access9 (2021), 125753–125767. doi:10.1109/ACCESS. 2021.3110244

    work page doi:10.1109/access 2021

  21. [21]

    Muhammad Umair Khan, Scott Uk-Jin Lee, Zhiqiang Wu, and Shanza Abbas

  22. [22]

    Electronics10, 18 (2021), 2211

    Wake lock leak detection in android apps using multi-layer perceptron. Electronics10, 18 (2021), 2211

    2021

  23. [23]

    Caroline Lemieux, Jeevana Priya Inala, Shuvendu K Lahiri, and Siddhartha Sen

  24. [24]

    In2023 IEEE/ACM 45th International Conference on Soft- ware Engineering (ICSE)

    Codamosa: Escaping coverage plateaus in test generation with pre-trained large language models. In2023 IEEE/ACM 45th International Conference on Soft- ware Engineering (ICSE). IEEE, 919–931

  25. [25]

    Yuanchun Li, Ziyue Yang, Yao Guo, and Xiangqun Chen. 2017. Droidbot: a lightweight UI-guided test input generator for Android. In2017 IEEE/ACM 39th International Conference on Software Engineering Companion (ICSE-C). IEEE, 23– 26

    2017

  26. [26]

    Yuanchun Li, Ziyue Yang, Yao Guo, and Xiangqun Chen. 2019. Humanoid: A Deep Learning-Based Approach to Automated Black-box Android App Testing. In 2019 34th IEEE/ACM International Conference on Automated Software Engineering (ASE). 1070–1073. doi:10.1109/ASE.2019.00104

    work page doi:10.1109/ase.2019.00104 2019

  27. [27]

    Tianming Liu, Haoyu Wang, Li Li, Guangdong Bai, Yao Guo, and Guoai Xu. 2019. DaPanda: Detecting Aggressive Push Notifications in Android Apps. In2019 34th IEEE/ACM International Conference on Automated Software Engineering (ASE). 66–78. doi:10.1109/ASE.2019.00017

    work page doi:10.1109/ase.2019.00017 2019

  28. [28]

    Zhe Liu, Chunyang Chen, Junjie Wang, Mengzhuo Chen, Boyu Wu, Xing Che, Dandan Wang, and Qing Wang. 2023. Chatting with GPT-3 for zero-shot human- like mobile automated GUI testing.arXiv preprint arXiv:2305.09434(2023)

    work page arXiv 2023

  29. [29]

    Manukulasooriya, R.S.I

    G.E. Manukulasooriya, R.S.I. Munasingha, Harinda Fernando, H.K.I. Uthpali, A.V.S.H. Yeshani, and Deemantha Siriwardana. 2024. Enhancing Automated Android Application Security through Hybrid Static and Dynamic Analysis Techniques. In2024 International Congress on Human-Computer Interaction, Opti- mization and Robotic Applications (HORA). 1–6. doi:10.1109/H...

    work page doi:10.1109/hora61326.2024 2024

  30. [30]

    Djamel Mesbah, Nour El Madhoun, Khaldoun Al Agha, and Hani Chalouati

  31. [31]

    InInternational Conference on Emerging Internet, Data & Web Technologies

    Leveraging prompt-based large language models for code smell detection: a comparative study on the MLCQ dataset. InInternational Conference on Emerging Internet, Data & Web Technologies. Springer, 444–454

  32. [32]

    Radinal Dwiki Novendra and Wikan Danar Sunindyo. 2024. Emerging Trends in Code Quality: Introducing Kotlin-Specific Bad Smell Detection Tool for Android Apps.IEEE Access12 (2024), 63895–63903. doi:10.1109/ACCESS.2024.3397055

    work page doi:10.1109/access.2024.3397055 2024

  33. [33]

    Ollama. 2024. Get up and running with LLMs. https://ollama.com

    2024

  34. [34]

    OpenAI. 2025. API reference. https://platform.openai.com/docs/api-reference/ responses/create#responses-create-temperature

    2025

  35. [35]

    Fabio Palomba, Dario Di Nucci, Annibale Panichella, Andy Zaidman, and Andrea De Lucia. 2017. Lightweight detection of android-specific code smells: The adoctor project. In2017 IEEE 24th international conference on software analysis, evolution and reengineering (SANER). IEEE, 487–491

    2017

  36. [36]

    Dimitri Prestat, Naouel Moha, and Roger Villemaire. 2022. An empirical study of Android behavioural code smells detection.Empirical Softw. Engg.27, 7 (dec 2022), 34 pages. doi:10.1007/s10664-022-10212-8

    work page doi:10.1007/s10664-022-10212-8 2022

  37. [37]

    Dimitri Prestat, Naouel Moha, Roger Villemaire, and Florent Avellaneda. 2024. DynAMICS: A Tool-Based Method for the Specification and Dynamic Detection of Android Behavioral Code Smells.IEEE Transactions on Software Engineering 50, 4 (2024), 765–784. doi:10.1109/TSE.2024.3363223

    work page doi:10.1109/tse.2024.3363223 2024

  38. [38]

    Jan Reimann, Martin Brylski, and Uwe Aßmann. 2014. A Tool-Supported Quality Smell Catalogue For Android Developers.Softwaretechnik-Trends34 (2014)

    2014

  39. [39]

    Ahmed R Sadik and Siddhata Govind. 2025. Benchmarking LLM for Code Smells Detection: OpenAI GPT-4.0 vs DeepSeek-V3.arXiv preprint arXiv:2504.16027 (2025)

    work page arXiv 2025

  40. [40]

    Yutaka Tsutano, Shakthi Bachala, Witawas Srisa-an, Gregg Rothermel, and Jack- son Dinh. 2019. Jitana: A modern hybrid program analysis framework for android platforms.Journal of Computer Languages52 (2019), 55–71. doi:10.1016/j.cola. 2018.12.004

    work page doi:10.1016/j.cola 2019

  41. [41]

    Gagnon, L

    Raja Vallée-Rai, Phong Co, Etienne M. Gagnon, L. Hendren, Patrick Lam, and V. Sundaresan. 2010. Soot: a Java bytecode optimization framework. InProceedings of the 20th ACM SIGPLAN Conference on Object-Oriented Programming, Systems, Languages, and Applications (OOPSLA 2005). ACM, 249–264. doi:10.1145/1094811. 1094838

    work page doi:10.1145/1094811 2010

  42. [42]

    Junjie Wang, Yuchao Huang, Chunyang Chen, Zhe Liu, Song Wang, and Qing Wang. 2024. Software Testing With Large Language Models: Survey, Landscape, and Vision.IEEE Transactions on Software Engineering50, 4 (2024), 911–936. doi:10.1109/TSE.2024.3368208

    work page doi:10.1109/tse.2024.3368208 2024

  43. [43]

    Hao Wen, Hongming Wang, Jiaxuan Liu, and Yuanchun Li. 2023. Droidbot-gpt: Gpt-powered ui automation for android.arXiv preprint arXiv:2304.07061(2023)

    work page arXiv 2023

  44. [44]

    Di Wu, Fangwen Mu, Lin Shi, Zhaoqiang Guo, Kui Liu, Weiguang Zhuang, Yuqi Zhong, and Li Zhang. 2024. iSMELL: Assembling LLMs with Expert Toolsets for Code Smell Detection and Refactoring. InProceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering(Sacramento, CA, USA)(ASE ’24). Association for Computing Machinery, New Yor...

    work page doi:10.1145/3691620.3695508 2024

  45. [45]

    Zhiqiang Wu, Xin Chen, and Scott Uk-Jin Lee. 2023. A systematic literature review on Android-specific smells.Journal of Systems and Software201 (2023), 111677. doi:10.1016/j.jss.2023.111677

    work page doi:10.1016/j.jss.2023.111677 2023

  46. [46]

    Husam N Yasin, Siti Hafizah Ab Hamid, and Raja Jamilah Raja Yusof. 2021. Droidbotx: Test case generation tool for android applications using Q-learning. Symmetry13, 2 (2021), 310

    2021

  47. [47]

    Juyeon Yoon, Robert Feldt, and Shin Yoo. 2024. Intent-Driven Mobile GUI Testing with Autonomous Large Language Model Agents. In2024 IEEE Conference on Software Testing, Verification and Validation (ICST). IEEE, 129–139

    2024

  48. [48]

    Rajif Agung Yunmar, Sri Suning Kusumawardani, Widyawan, and Fadi Mohsen

  49. [49]

    IEEE Access12 (2024), 41255–41286

    Hybrid Android Malware Detection: A Review of Heuristic-Based Approach. IEEE Access12 (2024), 41255–41286. doi:10.1109/ACCESS.2024.3377658

    work page doi:10.1109/access.2024.3377658 2024