Predicting Privacy Behavior on Online Social Networks

Abstract

Online Social Networks (OSNs) have come to play an increasingly important role in our social lives, and their inherent privacy problems have become a major concern for users. Can we assist consumers in their privacy decision-making practices, for example by predicting their preferences and giving them personalized advice? In order to accomplish this, we would need to study the factors that affect users’ privacy decision-making practices. In this paper, we intend to comprehensively investigate these factors in light of two common OSN scenarios: the case where other users request access to the user’s information, and the case where the user shares this information voluntarily. Using a real-life dataset from Google+ and three location-sharing datasets, we identify behavioral analogs to psychological variables that are known to affect users’ disclosure behavior: the trustworthiness of the requester/information audience, the sharing tendency of the receiver/information holder, the sensitivity of the requested/shared information, the appropriateness of the request/sharing activity, as well as some contextual information. We also explore how these factors work to affect the privacy decision making. Based on these factors we build a privacy decisionmaking prediction model that can be used to give users personalized advice regarding their privacy decisionmaking practices.