Page 24 -

Figure 7 shows network dynamics and representations of collaborative networks for all of
the energy technologies. For each technology area, we show the complete inventive
network (excluding non-collaborating inventors), the largest component of each network,
and the network of the most highly cited inventor (node and line coloring is described in
Figure 5). As noted above, accurate quantitative descriptions of inventive social
networks demand a level of technology-specific boundary robustness that is beyond the
scope of this preliminary examination. Additionally, inventor-level social influence can
only be accurately characterized by analyzing the social network signature across the full
USPTO dataset. But even though our datasets exhibit artificial social network
boundaries, preliminary results suggest significant collaborative network differences
between the technology fields at both the macro and micro levels. At the macro level,
biofuels patenting activity is more islanded with few inter-organizational linkages
connecting network components. Wind patenting activity sees moderate levels of inter-
organizational linkage, and solar sees by far the highest level of the three technology
areas.

Figure 7. Summary network visualizations for each technology area

As a preliminary test of whether these macro levels of isolation are a genuine network
feature or an artifact of the bounded dataset, we isolated the immediate inventive

17