MeSH on demand: an easy way to identify relevant MeSH terms and related articles from text [Poster].
Cho D, Mork JG, Aronson A, Demner-Fushman D, Schmidt S, Ozga D, Pash J, Kilbourne J
National Cancer Institute (NCI) Symposium "RNA Biology 2015".
Abstract:
The National Library of Medicine (NLM) has developed MeSH on Demand, a Web‐based tool that recommends Medical Subject Heading (MeSH®) terms from text such as an abstract or grant summary using the NLM Medical Text Indexer (MTI) software. MeSH on Demand has been developed in close collaboration between MeSH, NLM Index Section, and the Lister Hill National Center for Biomedical Communications to address the authors’ needs for assigning MeSH‐based key terms to their manuscripts. The NLM Medical Text Indexer (MTI) is the main product of the Indexing Initiative project and has been providing indexing recommendations based on the MeSH vocabulary since 2002. MTI combines human NLM Index Section expertise and Natural Language Processing technology to curate the biomedical literature more efficiently and consistently. In 2011, NLM expanded MTI's role by designating it as the first‐line indexer (MTIFL) for a select number of journals where it performs particularly well; today the MTIFL workflow includes over 200 journals for which MTIFL provides initial indexing that is then curated by human indexers. The MeSH on Demand tool is now freely available at www.nlm.nih.gov/mesh /MeSHonDemand.html and requires no software downloads or prior knowledge of the MeSH vocabulary. Users simply input text in the box labelled “Text to be processed.” The input text can contain up to 10,000 characters and may include a variety of text types including title, abstract, grant summary, or keywords. Well‐defined sentences provide the best results. The user then selects the “Find Mesh Terms” button, and MeSH on Demand provides as a result a list of MeSH terms that MTI identifies as being relevant to the text; the terms are displayed in alphabetical order, and a link to the MeSH Browser is provided for each term. For example, with input text that contains the phrase “treatment‐resistant depression,” MeSH on Demand automatically identifies the MeSH heading Depressive Disorder, Treatment‐Resistant. MeSH on Demand also includes links to the top ten related citations in PubMed (in text similarity‐ranked order) that were used in computing the MeSH on Demand results. Although the MeSH terms provided by MeSH on Demand are machine generated by the MTI software and do not reflect any human review, they do provide a good summary of the text.
Cho D, Mork JG, Aronson A, Demner-Fushman D, Schmidt S, Ozga D, Pash J, Kilbourne J. MeSH on demand: an easy way to identify relevant MeSH terms and related articles from text [Poster].
National Cancer Institute (NCI) Symposium "RNA Biology 2015".