Friendship Recommendations in Online Social Networks

Recommendation systems are popular both commercially and in the research community. For example, Online Social Networks (OSNs), like Twitter, are of special attention since a lot of connection are established between users without any previous knowledge. This highlights one of the key features of a lot of OSNs: the creation of relationships between users. Therefore, it is important to find new ways to provide interesting friendships suggestions.This work is the first step of an in-depth study whose goal is to find the right trade-offs between the number of factors explored in current state-of-the-art research. In particular, our contribution is an approach based on both Hubs And Authorities algorithm and similarity measures. The first one let us leverage triadic closures while the second one takes into account homophily. Even if the interplay between similarity and social ties is still an open issue in the analysis of OSNs, we lean towards the idea that it really counts.In order to support this hypothesis, a preliminary evaluation is performed on an implementation of the presented algorithm on datasets from Twitter. The deriving results show promising perspectives in terms of both effectiveness and scalability. These encouraging results are driving our future research efforts.