Measuring the completeness of scholarly communications databases

Abstract

As scholarly communication has been digitised and moved online, large streams of data are being generated by the millions of publications, citations, or viewership statis- tics. This data, gathered by a few specialised services, serves an important role in help- ing individual researchers conduct literature review, science policymakers to analyse the impact of research, and science as a whole to progress effectively.

This research is aiming to summarise the requirements regarding scope, quality, trans- parency and accessibility of scholarly communication databases, create a uniform method- ology of analysis of these datasets based on these requirements. The methodology is then used to analyse Google Scholar, Microsoft Academic and Scopus and the re- sults are compared to other studies of these datasets. High similarity of the results ob- tained using designed methodology to established publications show that the method- ology may be a promising method of partially automated, cross-disciplinary analysis of scholarly databases. Finally, a method of conducting an automated overlap analysis of datasets is presented as a methodological contribution, alongside relevant statistics of precision and recall.

More information

Identifiers

ORCID for Bartosz Paszcza: orcid.org/0000-0001-6394-3573

ORCID for Leslie Carr: orcid.org/0000-0002-2113-9680

ORCID for Stevan Harnad: orcid.org/0000-0001-6153-1129

ORCID for Jeremy Frey: orcid.org/0000-0003-0842-4302

Catalogue record

Export record