Very few fields in science are completely isolated; most disciplines have a fair amount of interdisciplinarity, even more so when the field is so settled that new results come only through interaction among one or several disciplines within a large field. This means that authors from initially separate fields end up converging, and creating new links in the coauthorship network. Besides, this implies a transfer of knowledge, so that authors initially focused in a single field (say, evolutionary computation) end up authoring papers in several other fields (such as, say, tabu search or scatter search) and thus extending their coauthorship ego-network and the coauthorship network in the field where both disciplines interact. This interaction also comes through horizontal applications (such as, for instance, scheduling), or broad problem areas, such as multiobjective optimization. Authors specialized on applications or on problem areas eventually coauthor papers in many different method areas, and thus build bridges that convey problem-solving (or, at least, paper-writing) expertise from a narrow field (such as evolutionary computation) to a broader one (such as metaheuristics).
[1]
Carlos Cotta,et al.
The complex network of EC authors
,
2006,
SEVO.
[2]
Carlos Cotta,et al.
The Complex Network of Evolutionary Computation Authors: an Initial Study
,
2005,
physics/0507196.
[3]
Á. M. Hernáez.
Borgatti, Stephen; Martin Everett i Lin Freeman. UCINET IV. Network Analysis Software. Version 1.0. Columbia: Analytic Technologies, 1992
,
1995
.
[4]
David H. Wolpert,et al.
No free lunch theorems for optimization
,
1997,
IEEE Trans. Evol. Comput..
[5]
M E Newman,et al.
Scientific collaboration networks. I. Network construction and fundamental results.
,
2001,
Physical review. E, Statistical, nonlinear, and soft matter physics.