当前位置: X-MOL 学术Journal of Cybersecurity › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Predicting smartphone location-sharing decisions through self-reflection on past privacy behavior
Journal of Cybersecurity Pub Date : 2020-09-22 , DOI: 10.1093/cybsec/tyaa014
Pamela Wisniewski 1 , Muhammad Irtaza Safi 1 , Sameer Patil 2 , Xinru Page 3
Affiliation  

Abstract
Smartphone location sharing is a particularly sensitive type of information disclosure that has implications for users’ digital privacy and security as well as their physical safety. To understand and predict location disclosure behavior, we developed an Android app that scraped metadata from users’ phones, asked them to grant the location-sharing permission to the app, and administered a survey. We compared the effectiveness of using self-report measures commonly used in the social sciences, behavioral data collected from users’ mobile phones, and a new type of measure that we developed, representing a hybrid of self-report and behavioral data to contextualize users’ attitudes toward their past location-sharing behaviors. This new type of measure is based on a reflective learning paradigm where individuals reflect on past behavior to inform future behavior. Based on data from 380 Android smartphone users, we found that the best predictors of whether participants granted the location-sharing permission to our app were: behavioral intention to share information with apps, the “FYI” communication style, and one of our new hybrid measures asking users whether they were comfortable sharing location with apps currently installed on their smartphones. Our novel, hybrid construct of self-reflection on past behavior significantly improves predictive power and shows the importance of combining social science and computational science approaches for improving the prediction of users’ privacy behaviors. Further, when assessing the construct validity of the Behavioral Intention construct drawn from previous location-sharing research, our data showed a clear distinction between two different types of Behavioral Intention: self-reported intention to use mobile apps versus the intention to share information with these apps. This finding suggests that users desire the ability to use mobile apps without being required to share sensitive information, such as their location. These results have important implications for cybersecurity research and system design to meet users’ location-sharing privacy needs.


中文翻译:

通过对过去的隐私行为的自我反思来预测智能手机的位置共享决策

摘要
智能手机位置共享是一种特别敏感的信息披露类型,会对用户的数字隐私和安全以及他们的人身安全产生影响。为了了解和预测位置公开行为,我们开发了一个Android应用,该应用从用户的手机中抓取元数据,要求他们向该应用授予位置共享权限,并进行了调查。我们比较了使用社会科学中常用的自我报告测度,从用户手机中收集的行为数据以及我们开发的一种新型测度的有效性,这种测度代表了自我报告和行为数据的混合体,以使用户对他们过去的位置共享行为的态度。这种新的衡量方法基于反思型学习范式,其中个人反思过去的行为以告知未来的行为。根据来自380个Android智能手机用户的数据,我们发现参与者是否授予我们的应用程序位置共享许可的最佳预测指标是:与应用程序共享信息的行为意图,“ FYI”通信方式以及我们的新混合动力之一会询问用户是否愿意与智能手机上当前安装的应用共享位置信息。我们对过去行为进行自我反思的新颖的,混合的结构显着提高了预测能力,并显示了将社会科学与计算科学方法相结合以改善对用户隐私行为的预测的重要性。进一步,在评估从以前的位置共享研究得出的行为意图构念的构想有效性时,我们的数据显示了两种不同类型的行为意图之间的明显区别:自我报告的使用移动应用程序的意图与与这些应用程序共享信息的意图。这一发现表明,用户希望能够使用移动应用程序而无需共享敏感信息(例如其位置)。这些结果对于满足用户的位置共享隐私需求的网络安全研究和系统设计具有重要意义。这一发现表明,用户希望能够使用移动应用程序而无需共享敏感信息(例如其位置)。这些结果对于满足用户的位置共享隐私需求的网络安全研究和系统设计具有重要意义。这一发现表明,用户希望能够使用移动应用程序而无需共享敏感信息(例如其位置)。这些结果对于满足用户的位置共享隐私需求的网络安全研究和系统设计具有重要意义。
更新日期:2020-09-22
down
wechat
bug