Exploring Nuances of User Privacy Preferences on a Platform for Political Participation

A problematic gap between existing online privacy controls and actual user disclosure behavior motivates researchers to focus on a design and development of intelligent privacy controls. These intelligent controls intend to decrease the burden of privacy decision-making and generate user-tailored privacy suggestions. To do so, at first it is necessary to analyze user privacy preferences. Previous studies have shown that user privacy profiles tend to have a multidimensional structure, which in turn might bring issues of an inexact user classification. This paper proposes to apply a fuzzy clustering approach, where fuzzy membership degree values can be used for the calculation of more precise personalized privacy suggestions. Based on the real-world dataset collected from a political platform , the fuzzy c-means algorithm was applied to demonstrate the multidimensionality and the existence of imprecise user privacy profiles, where a user simultaneously possesses features inherent in several clusters.