Abstract : In the influence maximisation problem, we are given a social network, modeled as a digraph, and a diffusion model which describes how a piece of information will propagate in the network. Particular nodes in the network, called seed nodes, will start the propagation of a message. The goal of the influence maximization problem is usually to select a set of seed nodes under a budget constraint to maximise the expected number of reached nodes (the nodes who will receive the piece of information). Many variations of this problem have been addressed. These variations are, for example, related to, game theory (competing brands on a social networks), or voting theory (manipulation of elections). After presenting some of the main results in the literature on these problems, I will present some recent works [1,2,3] that have tackled several new aspects as diversity (maximizing the diversity of information that nodes receive), fairness (maximizing the minimum expected coverage of a node or group of nodes) and centrality (finding the most important nodes for the diffusion process).
 Ruben Becker, Federico Corò, Gianlorenzo D'Angelo, Hugo Gilbert: Balancing Spreads of Influence in a Social Network. AAAI 2020: 3-10
 Ruben Becker, Gianlorenzo D'Angelo, Sajjad Ghobadi, Hugo Gilbert: Fairness in Influence Maximization through Randomization. CoRR abs/2010.03438 (2020)
 Eugenio Angriman, Ruben Becker, Gianlorenzo D’Angelo, Hugo Gilbert, Alexander van der Grinten, Henning Meyerhenke: Group-Harmonic and Group-Closeness Maximization -- Approximation and Engineering. ALENEX 2021 (to appear)