Generating Clustered Journal Maps: An Automated System for Hierarchical Classification

TL;DR

This paper introduces an automated system that generates hierarchical clustered maps of journals, enabling visualization and analysis of scientific communities and their fluctuations over time.

Contribution

It presents a novel routine using VOSviewer for hierarchical clustering and mapping of a large journal dataset, facilitating flexible and dynamic analysis.

Findings

Journal maps are sensitive to fluctuations between years.

The routine allows for customizable labeling of fields and subfields.

Clusters of related journals can be visualized to understand communities.

Abstract

Journal maps and classifications for 11,359 journals listed in the combined Journal Citation Reports 2015 of the Science and Social Sciences Citation Indexes are provided at http://www.leydesdorff.net/jcr15. A routine using VOSviewer for integrating the journal mapping and their hierarchical clustering is also made available. In this short communication, we provide background on the journal mapping/clustering and an explanation and instructions about the routine. We compare 2015 journal maps with those for 2014 and show the delineations among fields and subfields to be sensitive to fluctuations. Labels for fields and sub-fields are not provided by the routine, but can be added by an analyst for pragmatic or intellectual reasons. The routine provides a means for testing one's assumptions against a baseline without claiming authority, clusters of related journals can be visualized to understand communities. The routine is generic and can be used for any 1-mode network.