Generating large-scale network analyses of scientific landscapes in seconds using Dimensions on Google BigQuery

TL;DR

This paper introduces an open-source Python library that simplifies accessing and visualizing large-scale bibliometric data from Dimensions on Google BigQuery, enabling rapid analysis of scientific landscapes.

Contribution

The authors present a new Python tool that streamlines data retrieval and visualization from Dimensions, making complex bibliometric analyses more accessible and efficient.

Findings

Enables analysis of COVID-19 research patterns

Reduces technical barriers for data visualization

Facilitates large-scale scholarly landscape studies

Abstract

The growth of large, programatically accessible bibliometrics databases presents new opportunities for complex analyses of publication metadata. In addition to providing a wealth of information about authors and institutions, databases such as those provided by Dimensions also provide conceptual information and links to entities such as grants, funders and patents. However, data is not the only challenge in evaluating patterns in scholarly work: These large datasets can be challenging to integrate, particularly for those unfamiliar with the complex schemas necessary for accommodating such heterogeneous information, and those most comfortable with data mining may not be as experienced in data visualisation. Here, we present an open-source Python library that streamlines the process accessing and diagramming subsets of the Dimensions on Google BigQuery database and demonstrate its use on the freely available Dimensions COVID-19 dataset. We are optimistic that this tool will expand access to this valuable information by streamlining what would otherwise be multiple complex technical tasks, enabling more researchers to examine patterns in research focus and collaboration over time.