Odoo • Bild und Text

Frederic Raber

Transferring Recommendations through Privacy User Models across Domains

Abstract: Although privacy settings are important not only for data privacy but also to prevent hacking attacks like social engineering that depend on leaked private data in order to be  successful , most users do not care about them. Research has tried to help users in doing their privacy settings by using some of the privacy settings that have already been adapted by the user or individual factors like personality as an input to predict the remaining settings that have not been adapted so far. But in some cases, none of both is available. However, the user might have already done privacy settings in another domain where sensitive personal data is present, for example, she already adapted the privacy settings on the smartphone, but not on her social network account. In this article, we investigate at the example of four domains (social network posts, location sharing, smartphone app permission settings and data of an intelligent retail store), whether and how precise the privacy settings of a domain can be predicted across domains. We performed an exploratory study to examine which of the privacy settings of the aforementioned domains could be useful and validated our findings in a validation study.  Our results indicate that such an approach is working with a prediction precision significantly better than random. We identified clusters of domains that allow a model transfer between their members, and discuss which kind of privacy settings (general or context-based) leads to better prediction accuracy. Based on the results in this article, we would like to conduct user studies to find out whether the prediction precision is perceived by the users as a significant improvement over a “one size fits all” solution, where every user is given the same privacy settings.
 
Bio: Since the early days of my childhood, I was always fascinated by electronics and software, so it became clear that I have to pursue a career as a computer scientist. After finishing my Bachelor's and Master's Degree, I started working at the German Research  Center  for Artificial Intelligence, where I did research on AI in Retail Environments, Privacy Recommenders and Privacy User Interfaces, that allowed me to received my PhD about "Supporting Lay Users in Privacy Decisions When Sharing Sensitive Data". Currently, I'm continuing my work as a post-doc at DFKI in the domain of Usable Privacy in addition to my new research topic "Enhancing e-learning for motor skills using sensors and human-computer interfaces"


Below you can download the slides from this presentation:

l