<b>Enhancing Android Malware Detection: The Influence of ChatGPT on Decision-centric Task</b>
收藏资源简介：
We explore the transformative impact of non-decision models, specifically ChatGPT, on the traditional decision-centric task of Android malware detection. Through a series of carefully designed experiments using publicly available datasets, this study reveals a paradigm shift. It reveals a serious lack of interpretability in decision-driven solutions, raising concerns about their reliability. In contrast, ChatGPT, as a non-decision-making model, is good at providing comprehensive analysis reports and significantly enhances interpretability. We give developers more insights through a non-decision-making perspective.<b>ChatGPT</b>You can find ChatGPT’s analysis report on [APK_Analysis].<b>Project Structure</b>APK List: It contains the SHA256 of malicious and benign samples.<b>Dataset</b>All samples we used in our experiments you can find at [kronodroid].<b>Survey Results</b>We collect responses from <b>101 participants</b> and process their data by removing <b>personal or sensitive information</b>.Data preparation includes:Converting speech to textTranslating Chinese responses to EnglishRemoving redundant modal particlesThese processes ensure that the data is <b>clean and structured</b>, allowing for <b>accurate and efficient analysis</b>.<br>
本研究探讨了非决策模型（non-decision models）——尤以ChatGPT为例——对传统以决策为核心的Android恶意软件检测任务所产生的变革性影响。本研究通过一系列基于公开数据集的精心设计实验，揭示了该领域的范式转变。研究发现，传统决策驱动型检测方案存在严重的可解释性缺失问题，引发了学界对其可靠性的担忧。与之形成鲜明对比的是，作为非决策模型的ChatGPT能够输出全面的分析报告，显著提升了检测方案的可解释性。本研究从非决策视角为开发者提供了全新的研究洞见。 **ChatGPT** 您可在[APK_Analysis]中查阅ChatGPT的分析报告。 **项目结构** APK列表：收录了恶意样本与良性样本的SHA256哈希值。 **数据集** 本研究实验所用的全部样本均可在[kronodroid]中获取。 **调研结果** 本次调研共收集到**101名受访者**的反馈，并在数据处理阶段移除了所有**个人敏感信息**。 数据预处理流程包括： 1. 语音转文字 2. 将中文反馈翻译为英文 3. 去除冗余语气助词 上述处理流程确保了数据的**清洁性与结构化程度**，可为**精准高效的分析**提供保障。
