TrustHop: Building a Social Trust Network
In recent online social networks, each user can often assign a value to their immediate friends level of trustworthiness. Understanding a social trust value between any two nodes in an online social network is beneficial in a range of applications, like online marketing and recommendation systems. However, assessing social trust between two members in an online social network is difficult and time-consuming. This is because existing work either created handcrafted rules based on specialized domain knowledge or required a large number of computational resources, limiting its scalability. Graph-based techniques have recently been proved to be effective at learning from graph data. Even though social trust may be represented as graph data, its advantages have a lot of potential for trust evaluation. Therefore, we begin by reviewing the characteristics of online social networks and the properties of trust. After which the two types of graph-simplification and graph-analogy methodologies would be compared and contrasted as well as their respective problems and obstacles. We then conduct a quick examination of its pre- and post-processes to present an integrated view of trust evaluation. Finally, we discuss some unresolved issues that all trust models face.
Undergraduate thesis submitted to the Department of Computer Science and Information Systems, Ashesi University, in partial fulfillment of Bachelor of Science degree in Management Information Systems, May 2022
Papa Kwame Twumasi-Ntiamoah. (2022). TrustHop: Building a Social Trust Network. Ashesi University.