Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Using hit curves to compare search algorithm performance.

Jorge R Herskovic1, M Sriram Iyengar, Elmer V Bernstam

  • 1School of Health Information Sciences, The University of Texas Health Science Center at Houston, Houston, TX, USA.

Journal of Biomedical Informatics
|February 14, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Rural-Urban Differences in Hypertension Prevalence and Control in a Large Regional Health System Cohort.

Journal of clinical medicine·2026
Same author

Long COVID Persistence and Surveillance Gaps Across 58 US Hospitals.

JAMA network open·2026
Same author

Opportunities for informatics to improve patient experiences: observations and reflections of ACMI fellows.

Journal of the American Medical Informatics Association : JAMIA·2026
Same author

A conceptual model for provider scheduling: insights from an EHR implementation.

Journal of the American Medical Informatics Association : JAMIA·2026
Same author

Advancing delirium detection through the Open Health Natural Language Processing Consortium and the Evolve to Next-Gen Accrual to Clinical Trials Network.

The journals of gerontology. Series A, Biological sciences and medical sciences·2025
Same author

Review of tools to support Target Trial Emulation.

Journal of biomedical informatics·2025
Same journal

Evaluation of temporal preservation in synthetic longitudinal patient data.

Journal of biomedical informatics·2026
Same journal

ARKE: An ontology-driven framework for automated mapping of local radiology procedure terms to the LOINC-RadLex playbook using large language model.

Journal of biomedical informatics·2026
Same journal

A validation-driven training controller for cross-lingual biomedical NER via reinforcement learning-based adaptive loss weighting.

Journal of biomedical informatics·2026
Same journal

ASP-HR: An Adaptive Spatial Perception and Hierarchical Reasoning mechanism for document-level biomedical relation extraction.

Journal of biomedical informatics·2026
Same journal

Beyond Accuracy: Safety-Centered guidelines for the evaluation of LLM-based therapy recommendation systems for chronic multimorbidity patients.

Journal of biomedical informatics·2026
Same journal

DeepEN: A deep reinforcement learning framework for personalized enteral nutrition in critical care.

Journal of biomedical informatics·2026
See all related articles

New weighted hit curves offer a better way to evaluate information retrieval systems, especially for large databases like MEDLINE. These curves account for document relevance and ranking, improving performance measurement.

Area of Science:

  • Information Science
  • Computer Science
  • Data Science

Background:

  • Information retrieval system evaluation metrics have not evolved despite database growth.
  • Traditional metrics like precision and recall do not adequately assess ranking performance.
  • Large datasets (e.g., MEDLINE, World Wide Web) necessitate effective ranking for relevant document retrieval.

Purpose of the Study:

  • To introduce a novel method for measuring information retrieval system performance.
  • To develop evaluation metrics that incorporate document relevance, importance, and quality.
  • To address the limitations of existing metrics in evaluating ranked retrieval from large collections.

Main Methods:

  • Adapted weighted hit curves from statistical detection theory.

Related Experiment Videos

  • Applied hit curves to represent the position of relevant documents in large result sets.
  • Developed a formal model for information retrieval to integrate hit curve analysis.
  • Defined statistical methods for comparing the performance of multiple systems using hit curves.

    • Main Results:

    • Weighted hit curves effectively represent information retrieval system performance, including ranking.
    • Hit curves provide a more nuanced evaluation than traditional metrics like precision and recall for large datasets.
    • Demonstrated scenarios where hit curves are superior to existing measures for evaluating ranked retrieval.

    Conclusions:

    • Weighted hit curves offer a valuable new approach for evaluating information retrieval systems.
    • This method is particularly useful for large collections where ranking is critical.
    • Hit curves can better reflect desirable characteristics such as relevance and quality in retrieval performance.