Analysis of Users Behaviour from a? Movie Preferences Perspective

Abstract

Despite their tremendous popularity, Online Social Networks (OSNs) have several issues related to the privacy of social users. These issues have motivated researchers to develop OSN services that take advantage of the decentralized platforms (such as P2P systems or opportunistic networks). Decentralized Online Social Networks (DOSNs) need specific approaches to manage the decentralization of social data. In particular, data availability is one of the main issues and current proposals exploit properties of the social relationships to manage it. At the best of our knowledge, there are no proposals which exploit similarity between users, expressed with the term homophily. Homophily has been well studied in existing sociology literature, however, it is not easily extensible in Online Social Networks due to the limitations of real datasets. In this paper, we propose a preliminary analysis of similarity of social profiles in term of movie preferences. Results reveal that user’s friends are characterized by a different levels of similarity which can be exploited to propose solutions for the data availability problem.