Analyses of Research Topics in the Field of Informetrics Based on the Method of Topic Modeling

Abstract

In this study, we used the approach of topic modeling to uncover the possible structure of research topics in the field of Informetrics, to explore the distribution of the topics over years, and to compare the core journals. In order to infer the structure of the topics in the field, the data of the papers published in the Journal of Informetricsand Scientometrics during 2007 to 2013 are retrieved from the database of the Web of Science as input of the approach of topic modeling. The results of this study show that when the number of topics was set to 10, the topic model has the smallest perplexity. Although data scopes and analysis methodsare different to previous studies, the generating topics of this study are consistent with those results produced by analyses of experts. Empirical case studies and measurements of bibliometric indicators were concerned important in every year during the whole analytic period, and the field was increasing stability. Both the two core journals broadly paid more attention to all of the topics in the field of Informetrics. The Journal of Informetricsput particular emphasis on construction and applications ofbibliometric indicators and Scientometrics focused on the evaluation and the factors of productivity of countries, institutions, domains, and journals.

Keywords

• Analyses of research topics
• Informetrics
• Topic modeling