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

Obtaining interpretable fuzzy classification rules from medical data.

D Nauck1, R Kruse

  • 1University of Magdeburg, Faculty of Computer Science (FIN-IWS), Germany. detlef.nauck@cs.uni-magdeburg.de

Artificial Intelligence in Medicine
|June 23, 1999
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

Improved mesoscale segmentation of concrete from 3D X-ray images using contrast enhancers.

Cement & concrete composites·2024
Same author

Placental expression of microRNAs in infants born small for gestational age.

Placenta·2019
Same author

The oestrogen receptor 2 (ESR2) gene in female-pattern hair loss: replication of association with rs10137185 in German patients.

The British journal of dermatology·2013
Same author

Evidence for a polygenic contribution to androgenetic alopecia.

The British journal of dermatology·2013
Same author

Investigation of selected cytokine genes suggests that IL2RA and the TNF/LTA locus are risk factors for severe alopecia areata.

The British journal of dermatology·2012
Same author

Oral immunosuppressive medication for growing pigs in transplantation studies.

Laboratory animals·2012

This study enhances neuro-fuzzy classification (NEFCLASS) algorithms for interpretable fuzzy rule-based classifiers. Interactive pruning improves rule and variable readability in data analysis.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Classifiers support decision-making but 'black box' models are problematic in domains like medicine.
  • Fuzzy rule-based classifiers offer linguistic interpretability, overcoming limitations of symbolic or crisp methods.
  • Learning classifiers from data is crucial due to incomplete expert knowledge.

Purpose of the Study:

  • To discuss extensions to neuro-fuzzy classification (NEFCLASS) learning algorithms.
  • To present interactive strategies for pruning rules and variables to enhance classifier readability.
  • To demonstrate the enhanced approach on a practical example.

Main Methods:

  • Utilizing neuro-fuzzy approaches for learning fuzzy classifiers from data.

Related Experiment Videos

  • Implementing interactive strategies for rule and variable pruning.
  • Applying the NEFCLASS framework for data analysis.

    • Main Results:

    • Demonstrated extensions to NEFCLASS learning algorithms.
    • Introduced effective interactive pruning techniques for improved interpretability.
    • Successfully applied the enhanced approach to a small dataset.

    Conclusions:

    • Neuro-fuzzy approaches provide a convenient method for learning fuzzy classifiers.
    • Interactive pruning significantly enhances the readability and understandability of fuzzy classifiers.
    • The extended NEFCLASS approach offers a valuable tool for interpretable data analysis.