The Citation Cloud of a Biomedical Article: Enabling Citation Analysis

Background: Citation analysis is crucial for tracing the diffusion of knowledge. It has wide applicability to medical libraries, for instance to study research impact, to analyze the continued citation of retracted publications, and to find related articles via reference searching and citation tracking. Current tools are limited because they require expensive subscriptions (e.g., Web of Science, Scopus), lack a user-friendly public interface (e.g., Open Citations), or do not support backwards reference searching (e.g., Google Scholar). The objective of the Citation Cloud is to provide a fully open, user-friendly public interface for anyone to access or analyze citation data in biomedicine.

Description: We have built an extension to PubMed that allows any user to visualize and analyze important citation relationships around given article, using open citations provided by iCite, Microsoft Academic Graph, Semantic Scholar, Open Citations, PubMed Central, and ArnetMiner. The citation cloud allows tracking both backwards (“Articles Citing this Article”) and forwards via citation tracking. It also displays two important “sideways” citation-relationships: bibliographically coupled articles and co-cited articles. Two articles are bibliographically coupled articles when their reference lists overlap, and this remains constant over time. By contrast, two articles are co-cited if they are both cited by the same article(s). Co-citation is a measure of similarity that changes over time as newer articles may potentially co-cite new pairs of articles.

Conclusion: The Citation Cloud can be accessed by running any query on the Anne O’Tate value-added PubMed search interface https://arrowsmith.psych.uic.edu/cgi-bin/arrowsmith_uic/AnneOTate.cgi and clicking on the Citations button next to any retrieved article. One can view and export citations in each of the four different categories – citing articles, bibliography articles, co-cited articles, and bibliographically coupled articles – individually. Citation data is more easily available via the availability of open, user-friendly public interfaces. The Citation Cloud tool should greatly enable the study of biomedical citations by the scientific community. A preprint describing the Citation Cloud in detail has been posted in medRxiv.