Graphical models for studying museum networks: the Abbonamento Musei Torino Piemonte

Abstract: Probabilistic graphical models are a powerful tool to represent real-word phenomena and to learn network structures starting from data. This paper applies graphical models in a new framework to study association rules driven by consumer choices in a network of museums. The network consists of the museums participating in the program of Abbonamento Musei Torino Piemonte, which is a yearly subscription managed by the Associazione Torino Citt`a Capitale Europea. Consumers are card-holders, who are allowed to entry to all the museums in the network for one year. We employ graphical models to highlight associations among the museums driven by card-holder visiting behaviour. We use both undirected graphs to investigate the strength of the network and directed graphs to highlight asimmetry in the association rules.