MODELING AND MINING COMPLEX NETWORKS WITH FEATURE-RICH NODES

Real-world complex networks describe connections between objects; in reality, those objects are typically endowed with features. How does the presence or absence of such features interplay with the network link structure? The idea is to be able to represent a wide range of scenarios — not only homophily and heterophily. In this work, as a first thing we will present an ad-hoc statistical model, showing it displays the same global topological properties of a real-world social network. Then, we will use this model to design and analyze learning algorithms for graph mining problems – such as predicting links, anomaly detection, discovering missing features, and so on. Finally, we will present some results on real complex networks of different kinds (citation networks and semantic networks).