Extracting Information from Negative Interactions in Multiplex Networks Using Mutual Information

Many interesting real-world systems are represented as complex networks with multiple types of interactions and complicated dependency structures between layers. These interactions can be encoded as having a valence with positive links marking interactions such as trust and friendship and negative links denoting distrust or hostility. Extracting information from these negative interactions is challenging since standard topological metrics are often poor predictors of negative link formation, particularly across network layers. In this paper, we introduce a method based on mutual information which enables us to predict both negative and positive relationships. Our experiments show that SMLP (Signed Multiplex Link Prediction) can leverage negative relationship layers in multiplex networks to improve link prediction performance.