Link prediction algorithms can help to understand the structure and dynamics of complex systems, to reconstruct networks from incomplete data sets, and to forecast future interactions in evolving networks. Available algorithms based on similarity between nodes are bounded by the limited amount of links present in these networks. In this talk, we show how to reduce this latter intrinsic limitation and show that different kinds of relational data can be exploited to improve the prediction of new links. To this aim, we propose a link prediction algorithm by generalizing the Adamic-Adar method to multiplex networks composed by an arbitrary number of layers, that encode diverse forms of interactions. We show that this metric outperforms the classical single-layered Adamic-Adar score and other state-of-the-art methods, across several social, biological, and technological systems. As a by-product, the coefficients that maximize the multiplex Adamic-Adar metric indicate how the information structured in a multiplex network can be optimized for the link prediction task, revealing which layers are redundant. Interestingly, this effect can be asymmetric with respect to predictions in different layers. As an application, we specifically consider a multiplex representation of scientific networks, including collaboration, citations, common keywords between scientists, showing that the information included in publication records can be leveraged to improve the prediction of new scientific collaborations.

‍

Registration required: https://iu.zoom.us/webinar/register/WN_18kilHNBSdqRvZP4tH_5jw