A study reported in a recent article in the Journal of the Medical Library Association (JMLA) [1] examined how different information systems maneuver or map from users' natural language search terms to the Medical Subject Headings (MeSH®) and other access points in MEDLINE® data.
The study looked at the PubMed® MeSH Browser and the PubMed Index/Preview feature, but it did not include the most-used PubMed method for mapping the user's input, the PubMed query box itself. PubMed's logic for parsing unqualified terms entered in the query box involves automatic translation tables, one of which incorporates not only MeSH but also enhancements from the Unified Medical Language System (UMLS®). While some searchers may enter a PubMed search via the PubMed MeSH Browser, the majority start with the PubMed query box. NLM has made UMLS mapping available through the PubMed query box with no additional steps or clicks needed by the user. For example, a user's search for “heart attack” maps as: “myocardial infarction” [MeSH Terms] OR heart attack [Text Word]. This query box feature has the added benefit of including a text word search for the searcher's term that will retrieve in-process citations as well. Searchers can see how the PubMed query box has translated their input by clicking on the Details option under the query box, once a search has been run. More information on the automatic translation tables is available from the PubMed Help at http://www.ncbi.nlm.nih.gov:80/entrez/query/static/help/pmhelp.html#AutomaticTermMapping.
JMLA readers should also be aware that the Internet Grateful Med mapping feature examined in the study lives on as the Find Terms feature in the NLM Gateway. The Gateway searches a number of NLM databases simultaneously, including PubMed. The Find Terms feature is initiated by the first button under the Gateway query box and utilizes both MeSH and the UMLS, the same mapping file used by the PubMed query box. More information about this feature is available from the NLM Gateway Help at http://gateway.nlm.nih.gov/gw/Cmd?Help.x#Find_Terms.
Not all terms in the mapping file are suitable for inclusion in MeSH for one reason or another. Users of MeSH are encouraged to send vocabulary enrichment requests to NLM's MeSH Section using the MeSH Suggestion form available at http://www.nlm.nih.gov/mesh/meshsugg.html. Any problems encountered with the mapping file can be reported through the same mechanism.
Contributor Information
Lou Wave S. Knecht , Email: lou_knecht@nlm.nih.gov.
Stuart J. Nelson, Email: nelson@nlm.nih.gov.
Reference
- Gault LV, Shultz M, and Davies KJ. Variations in Medical Subject Headings (MeSH) mapping: from the natural language of patron terms to the controlled vocabulary of mapped lists. J Med Libr Assoc. 2002 Apr; 90(2):173–80. [PMC free article] [PubMed] [Google Scholar]