Extremal Coalitions for Influence Games Through Swarm Intelligence-Based Methods

An influence game is a simple represented over an graph (i.e., labeled, weighted graph) on which the spread phenomenon exerted. Influence games allow applying different properties and parameters coming from cooperative theory to contexts of social network analysis, decision-systems, voting systems, collective behavior. The exact calculation several these computationally hard, even for small number players. Two examples are length width game. size its smaller winning coalition, while larger losing coalition. Both relevant know levels difficulty in reaching agreements decision-making systems. Despite above, new bio-inspired metaheuristic algorithms have recently been developed solve NP-hard maximization problem efficient approximate way, being able find coalitions that maximize within graph. In this article, we apply some variations solution extreme coalitions, thus solutions games. As case study, consider two real networks, one formed by 58 members European Union Council under nice rules, other 705 Parliament, connected political affinity. Results promising show it feasible generate games, reduced solving time.