Research Guides: Scite.ai Guide: Things to Know Before You Get Started

Strengths & Weaknesses

Scite is an innovative platform for AI-driven citation analysis that offers features similar to those of traditional citation index databases like Web of Science or Scopus. These features include, but not limited to, searching for and finding citations, viewing visual representations of citation networks, and exporting references into citation management programs.

As part of the first wave of AI-powered citation tools, it has limitations compared to more established platforms. For example, Scite utilizes AI to categorize citations as mentioning, supporting, or contrasting; however, misclassification issues have been reported. Scite's subject coverage and size of content collection are limited, in certain ways, compared to Web of Science and Scopus databases. Before using Scite.ai for research, check if using Scite or other GenAI tools is allowed in your academic work, and if so, how to use and cite them responsibly. Refer to the "How to Cite & Disclose" section of this guide for more information.

Strengths	Weaknesses
AI-driven citation analysis: Scite uses machine learning to categorize citations as mentioning, supporting, or contrasting, to help understand contextual information on how a publication is being cited.	Precision and recall issues: Precision is high for some categories of citations, but recall for contrasting and supporting citations was reportedly lower.
Simple user interface: Scite enables users to pose questions conversationally and receive answers supported by scientific citations. It also gives users direct access to the text snippets where the citations appear.	Lack of subject headings such as MeSH: Lack of support for controlled vocabulary may contribute to a large number of results with irrelevant information.
Reference Check tool: Automatically identifies retracted articles, saving time for researchers.	Smaller content collection: Scite has a more limited scope than larger competitors like Web of Science and Scopus.
Helps track research evolution: Assists in identifying experts and tracking the development of research topics.	Focus on scientific research: Primarily designed for scientific literature. More useful for STEM disciplines, with less applicability to broader fields.
Visualization tools: Users can track and visualize citation trends, create dashboards, and set alerts.	Limited integration with non-scientific sources: Its functionality is primarily tailored to scholarly articles and journals with DOI, and not to books and non-journal literature.
Browser extension: Users can highlight text on websites and use the Scite extension to find related scientific articles.	Dashboard bugs: The dashboard feature often failed to update when new articles were added.

Articles about Scite

How it works:

Nicholson, J. M. (2021). Smart(er) Citations. Matter, 4(3), 756–758. https://doi.org/10.1016/j.matt.2021.02.007
Nicholson, J. M., Mordaunt, M., Lopez, P., Uppala, A., Rosati, D., Rodrigues, N. P., Grabitz, P., & Rife, S. C. (2021). scite: A smart citation index that displays the context of citations and classifies their intent using deep learning. Quantitative Science Studies, 2(3), 882–898. https://doi.org/10.1162/qss_a_00146
Uppala, A., Rosati, D., Nicholson, J. M., Mordaunt, M., Grabitz, P., & Rife, S. C. (2022). Title detection: a novel approach to automatically finding retractions and other editorial notices in the scholarly literature. arXiv preprint https://doi.org/10.48550/arxiv.2210.09553

Reviews:

Bakker, C., Theis-Mahon, N., & Brown, S. J. (2023). Evaluating the Accuracy of scite, a Smart Citation Index. Hypothesis: Research Journal for Health Information Professionals, 35(2). https://doi.org/10.18060/26528
Brody S. (2021). Scite. Journal of the Medical Library Association : JMLA, 109(4), 707–710. https://doi.org/10.5195/jmla.2021.1331
deLaubell, L. (2023). Scite. The Charleston Advisor, 24(4), 60-63.