The visualization and the understanding of large conceptual schemas require the use of specific methods. These methods generate clustered, summarized or focused schemas that are easier to visualize and to understand. All of these methods require computing the importance of the elements in the schema but, up to now, only the importance of entity types has been taken into account. In this paper, we present three methods for computing the importance of associations by taking into account the knowledge defined in the structural and behavioral parts of the schema. We experimentally evaluate these methods with large real-world schemas and present the main conclusions we have drawn from the experiments.

Peer Reviewed