16.4 Network Structure

We first consider the bipartite networks for each of the FP networks. Call the size of an organization the number of projects in which it takes part, and similarly call the size of a project the number of constituent organizations taking part in the project. These sizes correspond directly to the degrees of the relevant vertices in the bipartite networks. Both parts – organizations (Figure 16.1) and projects (Figure 16.2) – of each of the networks feature strongly skewed, heavy-tailed size distributions. The sizes of vertices can differ by orders of magnitude, pointing toward the existence of high degree hubs in the networks; hubs of this sort can play an important role in determining the network structure.

The organization size distributions are similar for each of the FPs. The underlying research activities thus have not altered the mix of organizations participating in a particular number of projects in each FP, despite changes in the nature of those research activities over time. In contrast, the rule changes in FP6 that favor larger project consortia are clearly seen in the project size distributions.

Turning to the projection networks, we see that both the organization projection (Table 16.1) and the project projection (Table 16.2) show small-world properties [37]. First, note that the great majority of the (N) vertices and (M) edges are in the largest connected component of the networks. In light of this, we focus on paths in only the largest component. ...

Get Analysis of Complex Networks now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.