Automating search term identification with MeSH entry terms via API

URN zum Zitieren dieses Dokuments:: urn:nbn:de:bvb:355-epub-746353
DOI zum Zitieren dieses Dokuments:: 10.5283/epub.74635

Knüttel, Helge

; Hausner, Elke ; Kapp, Claudia

Lizenz: Creative Commons Namensnennung 4.0 International
PDF - Präsentation
(1MB)

Veröffentlichungsdatum dieses Volltextes: 14 Jan 2025 06:33

Details

Vorschau

Bibliographische Daten exportieren

Zusammenfassung

Zusammenfassung

Context: Information retrieval for evidence synthesis in bibliographic databases includes free text terms in addition to controlled vocabulary (e.g. MeSH). Free text search terms are usually identified from different sources (e.g. seed articles, text analytical approaches or controlled vocabulary). Text mining of seed articles provides empirically derived search terms, but the approach requires high-quality references that are relevant to the research question [1]. Entry terms are selected synonyms of a keyword in thesauri such as the Medical Subject Headings from NLM (MeSH) . They are valuable candidates for text-word searches and may be used routinely [2]. They are especially useful for multi-word search terms as these are difficult to extract in text analytical approaches. Entry terms are also a useful source in cases, where little or no relevant references for text mining are known in advance. Manually extracting these terms from a thesaurus can be tedious and especially so when descriptors are exploded.

Objectives: To automate the retrieval of potential terms for text-word searches from entry terms of known relevant MeSH descriptors.

Design: Two tools using the freely available NLM Entrez APIs were developed that serve different use cases when developing systematic searches as an information specialist. Firstly, we will present an R-based tool, which provides a graphical user interface. And secondly, we will show a command line script employing NLM’s Entrez Direct utilities [3] enabling additional use cases such as direct integration into workflows or tools such as text editors.

Evaluation: We will present feedback from expert searchers on barriers to uptake, general feasibility and perceived value of the tools.

Outcomes and next steps: Results of the evaluation will be used for further development of approaches for identifying multi-word search terms.

References

1. Simon M, Hausner E, Klaus SF, Dunton N. Identifying nurse staffing research in Medline: development and testing of empirically derived search strategies with the PubMed interface. BMC Med Res Methodol. 2010;10:76. PubMed PMID: 20731858.
2. Bramer WM, Jonge GBd, Rethlefsen ML, Mast F, Kleijnen J. A systematic approach to searching: An efficient and complete method to develop literature searches. J Med Libr Assoc 2018;106(4):531–41. PubMed PMID: 30271302.
3. Kans J. Entrez Direct: E-utilities on the UNIX Command Line [Internet]. Bethesda (MD): National Center for Biotechnology Information (US); 2010- [cited 2023 Oct 26]. Available from: https://www.ncbi.nlm.nih.gov/books/NBK179288/.

Nur für Besitzer und Autoren: Kontrollseite des Eintrags

Downloadstatistik

Weitere Literatur (mittels CORE)

Details

Vorschau

Bibliographische Daten exportieren

Zusammenfassung

Zusammenfassung

Downloadstatistik

Downloads

Weitere Literatur (mittels CORE)

Universitätsbibliothek