Share

Methodology

Methodology for computing HepaTop ranking from BioPredictive Labs

BioPredictive R&D Team, BioPredictive Labs, Paris, France - contact@biopredictive.com.

Objectives

Hepatology is a competitive medical specialty with an exponential increase of publications. Key idea of this study is to highlight key authors of the specialty, trying to identify raising stars too.

Methods

Population

Inclusion criteria : authors in hepatology with at least 2 abstracts accepted in AASLD meeting in 2009.

h-index score

The publications database of each authors from the above population is built thanks to PubMed/Medline. Then Google Scholar[3] is used to collect citations of these publications, in order to compute h-index from Hirsch [1]. The index "h" (a reference for the bibliographists) represents the number of publications "h" cited at least h times in the literature. It identifies the productive authors whose work becomes a reference[2]. The higher h-index, the more the author is a reference. An audit of false positives related to homonyms has been practiced for the top 100 selected authors, and an algorithm has been developed to remove the obviously false results.

Ranking

To avoid a bias due to authors' age, the population have been divided into 4 : "Dinosaurs" publishing since more than 20 years, "Seniors" publishing since 10 to 20 years, "Juniors" publishing since 5 to 10 years and "Babies" publishing since less than 5 years. This segmentation relies on data from PubMed/Medline.

Limits, false positives, false negatives.

Requests are based on author name, according to PubMed naming convention [4] (i.e : for "Friedman, Scott L.", search is done using "Friedman SL"). A vulnerability of disambiguation is remaining, especially for very common names [5]. A security algorithm has been implemented to detect and exclude authors with high risk of "false positive", having more than 80 publications in average over the last 5 years or 50 years. Conversely, false negatives may be observed, with authors who have not published at AASLD in 2009.

Conclusion

Hepatop ranking is presented in real time on hepatop.com [6]. The results on the site are directly computed from the algorithms presented here and updated every month. The classification by age (dinosaurs, seniors, juniors and babies) is relevant, highlighting both eminences known in the art (dinosaurs), confirmed authors (seniors) but also no-doubt future leaders (juniors, babies).
To be noted as well : a prominent number of American authors appears in this ranking regardless of age.

Bibliography

  1. Hirsch JE. An index to quantify an individual's scientific research output. Proc Natl Acad Sci U S A. 2005;102:16569-72.
  2. Ball P. Achievement index climbs the ranks. Nature 2007;448:737.
  3. Jacso ́ P. Testing the calculation of a realistic h-index in Google Scholar, Scopus and Web of Science for F.W. Lancaster, Library Trends 2008.
  4. http://www.nlm.nih.gov/bsd/mms/medlineelements.html#fau
  5. Torvik V, Smalheiser N. Author name disambiguation in MEDLINE. ACM Trans Knowl Discov Data 2009;3:1-29.
  6. http://hepatop.biopredictive.com
Back to homepage

Due mainly to homonyms, errors are possible and if you detect "false positives" or "false negatives" don't be too frustrated and please contact the webmaster.