Characterizing Interdisciplinarity of Researchers and Research Topics Using Web Search Engines

Researchers\u27 networks have been subject to active modeling and analysis. Earlier literature mostly focused on citation or co-authorship networks reconstructed from annotated scientific publication databases, which have several limitations. Recently, general-purpose web search engines have also been utilized to collect information about social networks. Here we reconstructed, using web search engines, a network representing the relatedness of researchers to their peers as well as to various research topics. Relatedness between researchers and research topics was characterized by visibility boost-increase of a researcher\u27s visibility by focusing on a particular topic. It was observed that researchers who had high visibility boosts by the same research topic tended to be close to each other in their network. We calculated correlations between visibility boosts by research topics and researchers\u27 interdisciplinarity at the individual level (diversity of topics related to the researcher) and at the social level (his/her centrality in the researchers\u27 network). We found that visibility boosts by certain research topics were positively correlated with researchers\u27 individual-level interdisciplinarity despite their negative correlations with the general popularity of researchers. It was also found that visibility boosts by network-related topics had positive correlations with researchers\u27 social-level interdisciplinarity. Research topics\u27 correlations with researchers\u27 individual- and social-level interdisciplinarities were found to be nearly independent from each other. These findings suggest that the notion of interdisciplinarity of a researcher should be understood as a multi-dimensional concept that should be evaluated using multiple assessment means