March 16, 2018


The Unique Value of MathSciNet®

MathSciNetLibrarians and researchers understand the value of a subject-specific database like MathSciNet®: they are focused and authoritative. Still, while most databases provide abstracting and indexing, MathSciNet® also provides users with 3rd party reviews of the published literature created by over 21,000 PhD reviewers. This approach dates back 75 years to the database’s origin as the publication Mathematical Reviews launched in 1940 by Otto Neugebauer.

“We spend a lot of time getting things right,” says Edward Dunne, Executive Editor, Mathematical Reviews at AMS. Dunne expects their reviewers to answer the following questions: What is the context? What is the content? What might this lead to?

Specificity in searches matters in large part due to the sheer volume of research being published. Over the last 30 years, the mathematics literature has grown at a rate of approximately 3.6% per year with the entire body of work doubling every 19 years. By 2036, we can expect to have 250,000 mathematics articles published per year.

MathSciNet® offers an alternative to pure machine learning approaches to indexing the scholarly literature. “Part of what we do is human-aided computing and part is computer-aided human endeavor,” says Dunne.

Dunne provides us a few examples of how MathSciNet® helps researchers and librarians:

  1. Andrea Bertozzi, a professor at UCLA, specializes in partial differential equations and their applications. In particular, some of her influential work is in the area of the mathematics of swarms. An example of one of her papers on swarms is Topaz, Chad M.; Bertozzi, Andrea L.; Lewis, Mark A., A nonlocal continuum model for biological aggregation. Bull. Math. Biol.68(2006), no. 7, 1601–1623, is findable in MathSciNet, but doesn’t have swarm in the titles. Here the reviewer connects “continuum model for biological aggregation” with “swarm”.
  2. Another example of the value of searching with reviews is a paper that discusses cryptocurrencies where the title does not mention blockchains, the fundamental idea underlying all cryptocurrencies. However, this key term can be found in the review text. A mathematician might be interested in blockchains, which is where the mathematics is, more than the currencies. Blockchains can also be used for contracts, as explained in the paper and the review, and other transactional settings where security and verification are issues, explains Dunne.
  3. An important active topic in mathematics is the Minimal Model Program, an area of algebraic geometry that has seen dramatic results in the last five years or so. A natural search is for the phrase “minimal model”. However, the constituent terms have a wide variety of meanings depending on context. For instance, a search in Web of Science for the phrase “minimal model” results in 4,541 matches. But 917 are in Endocrinology, 451 are in Physics: Particles and Fields, and 404 are in Mathematics. If you do the same search in MathSciNet, there are 1,685 matches. So, the subject-specific database not only covers more articles, but just as importantly, the researcher is confident that all of the matches have something to do with mathematics.

“As someone with a mathematics background and an appreciation of librarians, I am proud to be part of this effort to ensure the searchability and accuracy of the database of the mathematics literature,” says Dunne, who adds, “In many ways, mathematics literature is the wild, wild west, but if you want to efficiently and accurately tame the literature, MathSciNet® is the right tool for doing that.”

Edward Dunne is the Executive Editor for Mathematical Reviews. He has worked for the American Mathematical Society (AMS) since 1997. For the first seventeen years, he was an Editor for the AMS Book Program. Prior to coming to the AMS, he held academic positions at Oklahoma State University, Oxford University, and Rice University. For a time, he worked for Springer-Verlag in Heidelberg, Germany. He has PhD from Harvard and a BS from Santa Clara University.

MathSciNet® Data points:

  • 3,500,819 total publications
  • 882,474 authors indexed
  • Over 1,800 current journals
  • 21,946 active reviewers
  • 2,154,340 direct links from MathSciNet to the full-text version of the paper
  • 12,961,827 matched citations (from reference lists)
  • 376,265 authors cited
  • 3,441 journals cited
  • Approx. 125,000 items added to the database per year

*As of January 24, 2018

MathSciNet®, compiled, edited and delivered by the American Mathematical Society, is the authoritative gateway to the scholarly literature of mathematics. MathSciNet® contains information on over 3 million articles and books, and includes expert reviews, author profiles, and citation information on articles, books, journals, and authors. MathSciNet® is also available from EBSCO Information Services as MathSciNet via EBSCOhost®. For more information and a free trial, visit the EBSCO website.

 Sponsored by



Comment Policy:
  1. Be respectful, and do not attack the author, people mentioned in the article, or other commenters. Take on the idea, not the messenger.
  2. Don't use obscene, profane, or vulgar language.
  3. Stay on point. Comments that stray from the topic at hand may be deleted.
  4. Comments may be republished in print, online, or other forms of media, per our Terms of Use.

We are not able to monitor every comment that comes through (though some comments with links to multiple URLs are held for spam-check moderation by the system). If you see something objectionable, please let us know. Once a comment has been flagged, a staff member will investigate.

We accept clean XHTML in comments, but don't overdo it and please limit the number of links submitted in your comment. For more info, see the full Terms of Use.

Speak Your Mind