Visualizing Research Fields based on Scholarly Communication on the Web

Abstract

In science, information overload is not a contemporary issue. At the beginning of a scientific study, it is therefore usually quite cumbersome to get an overview of a research field. In this thesis, I aim to address this problem of classic literature search with visualizations based on scholarly communication on the web. Knowledge domain visualizations are usually based on citations. Co-citation is an established measure of subject similarity and can thus be used to structure a field. Due to the publication lag, however, the appearance of citations is considerably delayed. For this thesis, I propose to employ readership statistics instead of citations to calculate subject similarities. Readership statistics have a distinct advantage over citations: they are available shortly after the paper has been published. In this thesis, I present work on an interactive visualization of research fields based on readership statistics from the online reference management system Mendeley. As a use case I have chosen educational technology, because it represents a field that is multidisciplinary and highly dynamic in nature. The visualization created from co-readership patterns contains 91 papers which are attributed to 13 areas. The visualization is fully automated with the exception of choosing the number of publications to include and correcting some of the names from the naming algorithm. In comparison to citation analyses, the proposed visualization is more diverse. Furthermore, the visualization is a very recent representation of the field: 80% of the publications included were published in the last 10 years. Being based on the readers, however, their characteristics may introduce biases to the visualization. Knowledge domain visualizations based on readership statistics therefore present a timely alternative to citation-based overviews, but it is important that the characteristics of the underlying sample are made transparent.