TRUSIS · Trust in Social Internetworking Systems

Social media applications, such as blogs, multimedia and link sharing sites, question and answering systems, wikis and online forums, are growing at an unprecedented rate and are estimated to generate a significant amount of the contents currently available on the Web. In real cases a user can decide to affiliate to many social networks and to create an account for each of them and the resulting scenario is called social internetworking system (SIS). In the SIS scenario, mechanisms allowing users to locate “reliable” users with whom they can profitably interact and detect malicious users are mandatory. In this project we address the problem of defining mechanisms for finding reliable users in a SIS. The proposed methods significantly extends the concepts of trust and reputation which are present in virtual communities. The goals of this project are: (i) To define a simple mathematical model of social internetworking and analyze factors influencing the computation of trust and reputation with a special emphasis on some typical Web 2.0 features. (ii) To extend the basic model with “context-awareness” functionalities in order to specify “trust/reputation” of users in concrete domains. (iii) To build an ontology capable of representing trust and reputation data in multiple social networks. (iv) To carry out long-term iterative testing and validation activities on real users.