Using citation data to improve retrieval from MEDLINE

Elmer V. Bernstam, Jorge R. Herskovic, Yindalon Aphinyanaphongs, Constantin F. Aliferis, Madurai G. Sriram, William (Bill) Hersh

Research output: Contribution to journalArticle

36 Citations (Scopus)

Abstract

Objective: To determine whether algorithms developed for the World Wide Web can be applied to the biomedical literature in order to identify articles that are important as well as relevant. Design and Measurements: A direct comparison of eight algorithms: simple PubMed queries, clinical queries (sensitive and specific versions), vector cosine comparison, citation count, journal impact factor, PageRank, and machine learning based on polynomial support vector machines. The objective was to prioritize important articles, defined as being included in a pre-existing bibliography of important literature in surgical oncology. Results: Citation-based algorithms were more effective than noncitation-based algorithms at identifying important articles. The most effective strategies were simple citation count and PageRank, which on average identified over six important articles in the first 100 results compared to 0.85 for the best noncitation-based algorithm (p <0.001). The authors saw similar differences between citation-based and noncitation-based algorithms at 10, 20, 50, 200, 500, and 1,000 results (p <0.001). Citation lag affects performance of PageRank more than simple citation count. However, in spite of citation lag, citation-based algorithms remain more effective than noncitation-based algorithms. Conclusion: Algorithms that have proved successful on the World Wide Web can be applied to biomedical information retrieval. Citation-based algorithms can help identify important articles within large sets of relevant results. Further studies are needed to determine whether citation-based algorithms can effectively meet actual user information needs.

Original languageEnglish (US)
Pages (from-to)96-105
Number of pages10
JournalJournal of the American Medical Informatics Association
Volume13
Issue number1
DOIs
StatePublished - Jan 2006

Fingerprint

MEDLINE
Internet
Journal Impact Factor
Information Storage and Retrieval
Bibliography
PubMed

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Bernstam, E. V., Herskovic, J. R., Aphinyanaphongs, Y., Aliferis, C. F., Sriram, M. G., & Hersh, W. B. (2006). Using citation data to improve retrieval from MEDLINE. Journal of the American Medical Informatics Association, 13(1), 96-105. https://doi.org/10.1197/jamia.M1909

Using citation data to improve retrieval from MEDLINE. / Bernstam, Elmer V.; Herskovic, Jorge R.; Aphinyanaphongs, Yindalon; Aliferis, Constantin F.; Sriram, Madurai G.; Hersh, William (Bill).

In: Journal of the American Medical Informatics Association, Vol. 13, No. 1, 01.2006, p. 96-105.

Research output: Contribution to journalArticle

Bernstam, EV, Herskovic, JR, Aphinyanaphongs, Y, Aliferis, CF, Sriram, MG & Hersh, WB 2006, 'Using citation data to improve retrieval from MEDLINE', Journal of the American Medical Informatics Association, vol. 13, no. 1, pp. 96-105. https://doi.org/10.1197/jamia.M1909
Bernstam, Elmer V. ; Herskovic, Jorge R. ; Aphinyanaphongs, Yindalon ; Aliferis, Constantin F. ; Sriram, Madurai G. ; Hersh, William (Bill). / Using citation data to improve retrieval from MEDLINE. In: Journal of the American Medical Informatics Association. 2006 ; Vol. 13, No. 1. pp. 96-105.
@article{cc28581733be4a71b35e3c29b062410a,
title = "Using citation data to improve retrieval from MEDLINE",
abstract = "Objective: To determine whether algorithms developed for the World Wide Web can be applied to the biomedical literature in order to identify articles that are important as well as relevant. Design and Measurements: A direct comparison of eight algorithms: simple PubMed queries, clinical queries (sensitive and specific versions), vector cosine comparison, citation count, journal impact factor, PageRank, and machine learning based on polynomial support vector machines. The objective was to prioritize important articles, defined as being included in a pre-existing bibliography of important literature in surgical oncology. Results: Citation-based algorithms were more effective than noncitation-based algorithms at identifying important articles. The most effective strategies were simple citation count and PageRank, which on average identified over six important articles in the first 100 results compared to 0.85 for the best noncitation-based algorithm (p <0.001). The authors saw similar differences between citation-based and noncitation-based algorithms at 10, 20, 50, 200, 500, and 1,000 results (p <0.001). Citation lag affects performance of PageRank more than simple citation count. However, in spite of citation lag, citation-based algorithms remain more effective than noncitation-based algorithms. Conclusion: Algorithms that have proved successful on the World Wide Web can be applied to biomedical information retrieval. Citation-based algorithms can help identify important articles within large sets of relevant results. Further studies are needed to determine whether citation-based algorithms can effectively meet actual user information needs.",
author = "Bernstam, {Elmer V.} and Herskovic, {Jorge R.} and Yindalon Aphinyanaphongs and Aliferis, {Constantin F.} and Sriram, {Madurai G.} and Hersh, {William (Bill)}",
year = "2006",
month = "1",
doi = "10.1197/jamia.M1909",
language = "English (US)",
volume = "13",
pages = "96--105",
journal = "Journal of the American Medical Informatics Association",
issn = "1067-5027",
publisher = "Oxford University Press",
number = "1",

}

TY - JOUR

T1 - Using citation data to improve retrieval from MEDLINE

AU - Bernstam, Elmer V.

AU - Herskovic, Jorge R.

AU - Aphinyanaphongs, Yindalon

AU - Aliferis, Constantin F.

AU - Sriram, Madurai G.

AU - Hersh, William (Bill)

PY - 2006/1

Y1 - 2006/1

N2 - Objective: To determine whether algorithms developed for the World Wide Web can be applied to the biomedical literature in order to identify articles that are important as well as relevant. Design and Measurements: A direct comparison of eight algorithms: simple PubMed queries, clinical queries (sensitive and specific versions), vector cosine comparison, citation count, journal impact factor, PageRank, and machine learning based on polynomial support vector machines. The objective was to prioritize important articles, defined as being included in a pre-existing bibliography of important literature in surgical oncology. Results: Citation-based algorithms were more effective than noncitation-based algorithms at identifying important articles. The most effective strategies were simple citation count and PageRank, which on average identified over six important articles in the first 100 results compared to 0.85 for the best noncitation-based algorithm (p <0.001). The authors saw similar differences between citation-based and noncitation-based algorithms at 10, 20, 50, 200, 500, and 1,000 results (p <0.001). Citation lag affects performance of PageRank more than simple citation count. However, in spite of citation lag, citation-based algorithms remain more effective than noncitation-based algorithms. Conclusion: Algorithms that have proved successful on the World Wide Web can be applied to biomedical information retrieval. Citation-based algorithms can help identify important articles within large sets of relevant results. Further studies are needed to determine whether citation-based algorithms can effectively meet actual user information needs.

AB - Objective: To determine whether algorithms developed for the World Wide Web can be applied to the biomedical literature in order to identify articles that are important as well as relevant. Design and Measurements: A direct comparison of eight algorithms: simple PubMed queries, clinical queries (sensitive and specific versions), vector cosine comparison, citation count, journal impact factor, PageRank, and machine learning based on polynomial support vector machines. The objective was to prioritize important articles, defined as being included in a pre-existing bibliography of important literature in surgical oncology. Results: Citation-based algorithms were more effective than noncitation-based algorithms at identifying important articles. The most effective strategies were simple citation count and PageRank, which on average identified over six important articles in the first 100 results compared to 0.85 for the best noncitation-based algorithm (p <0.001). The authors saw similar differences between citation-based and noncitation-based algorithms at 10, 20, 50, 200, 500, and 1,000 results (p <0.001). Citation lag affects performance of PageRank more than simple citation count. However, in spite of citation lag, citation-based algorithms remain more effective than noncitation-based algorithms. Conclusion: Algorithms that have proved successful on the World Wide Web can be applied to biomedical information retrieval. Citation-based algorithms can help identify important articles within large sets of relevant results. Further studies are needed to determine whether citation-based algorithms can effectively meet actual user information needs.

UR - http://www.scopus.com/inward/record.url?scp=29244472467&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=29244472467&partnerID=8YFLogxK

U2 - 10.1197/jamia.M1909

DO - 10.1197/jamia.M1909

M3 - Article

VL - 13

SP - 96

EP - 105

JO - Journal of the American Medical Informatics Association

JF - Journal of the American Medical Informatics Association

SN - 1067-5027

IS - 1

ER -