We use cookies to improve your experience with our site.

基于安卓重打包应用程序的恶意代码定位研究

On Identifying and Explaining Similarities in Android Apps

  • 摘要: 应用更新和应用重打包的行为在Android生态中普遍存在。这个现象导致应用市场中存在很多相似的应用程序,这些相似性应用程序需要被识别出来。尽管目前已经有一些研究能够提升重打包应用/克隆应用检测的可扩展性,为了理解和验证应用之间的相似性,研究者和实践者最终还是会面临复杂地成对比较(或者多个应用同时比较)。本文设计并实现了一个原型系统SimiDroid,用于Android应用的多级别相似性比较。SimiDroid旨在对应用的不同版本以及重打包应用的相似/变化之处进行全面分析。特别的,通过探究在不同场景下重打包应用的构建情况,本文通过实验表明实现一个重打包应用的检测以及相似性解释工具的重要性和必要性。本文的实验结果进一步表明,SimiDroid中实现的相似性比较比现有的方法更加准确。

     

    Abstract: App updates and repackaging are recurrent in the Android ecosystem, filling markets with similar apps that must be identified. Despite the existence of several approaches to improving the scalability of detecting repackaged/cloned apps, researchers and practitioners are eventually faced with the need for a comprehensive pairwise comparison (or simultaneously multiple app comparisons) to understand and validate the similarities among apps. In this work, we present the design and implementation of our research-based prototype tool called SimiDroid for multi-level similarity comparison of Android apps. SimiDroid is built with the aim to support the comprehension of similarities/changes among app versions and among repackaged apps. In particular, we demonstrate the need and usefulness of such a framework based on different case studies implementing different dissection scenarios for revealing various insights on how repackaged apps are built. We further show that the similarity comparison plugins implemented in SimiDroid yield more accurate results than the state of the art.

     

/

返回文章
返回