publications). While the sampling size of the data solicitations is small, Figure B-1 reveals several interesting features that may be useful for subsequent analysis.

For example, all hubs are located in the United States, and all but one of these hubs are U.S. research universities. The coauthorship network maps show that some areas of research (and some researchers specifically) tend to collaborate on a nation- or region-specific basis or both. For example, chip architecture and parallel programming networks are primarily U.S. based with limited participation by Europe and Asia. In contrast, semiconductor device-scaling networks show a significant number of collaborations with Asia. In particular, Taiwan holds the vast majority share of Asia’s representation, followed by Singapore and Japan (data not shown).

While the individual hubs do not generally show a significant degree of connectivity with one another (with exception to two hubs in the semiconductor scaling networks), the semiconductor scaling and chip architecture networks appear to be highly interconnected. In fact, both of these networks share a common hub. In contrast, researchers within the parallel programming networks display much less connectivity.

Increasing circle size, increasing connectedness between researchers, and widening lines between researchers may all be useful indicators for identifying emerging hubs of research activity. For instance, a small circle with many connections might suggest an individual who publishes less but collaborates frequently and is thereby more intimately connected to the global knowledge network. In addition, a wide line between a small circle and an established research hub might suggest a promising early-career researcher who hails from a strong research lineage. This analysis could also be extended by observing how coauthored publication networks change over time.

In summary, this methodology presents a unique approach for identifying emerging, as well as established, hubs of research activity in three areas of science and technology. However, given the small sampling size of the data solicitations, this experiment is not intended to provide any assessment or interpretation of the hubs themselves (or of trends apparent in the network maps). Rather, the goal of this experiment is to demonstrate an approach that could be extended and/or modified (e.g., to include statistically valid data-gathering methodologies) for subsequent in-depth exploration in any number of research areas.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement