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

Extracting regression rules from neural networks.

Kazumi Saito1, Ryohei Nakano

  • 1NTT Communication Science Laboratories, NTT Corporation, Soraku-gun, Kyoto, Japan. saito@cslab.kecl.ntt.co.jp

Neural Networks : the Official Journal of the International Neural Network Society
|November 12, 2002
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

Non-destructive monitoring of cyenopyrafen residues on strawberry via 280 nm UV-excited fluorescence of co-formulated organic solvents.

Food chemistry: X·2026
Same author

Non-climacteric but genotype-dependent ethylene response characteristics in highbush blueberry (Vaccinium corymbosum) fruit ripening.

Journal of experimental botany·2026
Same author

Agricultural activities maintain vigor and plant-based protein intake in older adults with type 2 diabetes: A feasibility study.

JAR life·2026
Same author

Phenological variation associates with the stability of fruit quality traits in cultivated tetraploid blueberry.

G3 (Bethesda, Md.)·2025
Same author

The haplotype-phased genome assembly facilitated the deciphering of the bud dormancy-related QTLs in Prunus mume.

DNA research : an international journal for rapid publication of reports on genes and genomes·2024
Same author

A low-cost dpMIG-seq method for elucidating complex inheritance in polysomic crops: a case study in tetraploid blueberry.

Horticulture research·2024
Same journal

Dynamic analysis and reliable mechanical optimization application of ring HNN effected with a memristive neuron.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

DAFF-Net: A detection and search method for small-scale low surface brightness galaxies.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Quasi-synchronization for complex networks with hybrid pinning intermittent control.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Physics-encoded convolutional neural operators for parametric PDEs: A convergence-guaranteed framework via pre-computed kernel fields.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Exploiting audio-visual modalities in videos: Object detection via multi-stage bilateral coupling network.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Reliability-aware modality completion with cross-modal distillation for federated learning with missing modalities.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

This study introduces a novel method for extracting interpretable regression rules from neural networks. The approach effectively handles mixed data types, yielding accurate and generalizable rules.

Area of Science:

  • Machine Learning
  • Data Mining
  • Artificial Intelligence

Background:

  • Neural networks are powerful predictive models but often lack interpretability.
  • Extracting understandable rules from complex models is crucial for trust and validation.
  • Multivariate data with mixed variable types (nominal and numeric) presents unique challenges for rule extraction.

Purpose of the Study:

  • To propose a new framework and method for extracting regression rules from neural networks.
  • To handle multivariate data containing both nominal and numeric variables.
  • To generate accurate and interesting regression rules that are interpretable.

Main Methods:

  • The method generates an initial regression rule per training sample.
  • It employs the kappa-means algorithm for rule generalization and condition simplification.

Related Experiment Videos

  • Decision-tree induction is used to form logical conditions for nominal variables.

    • Main Results:

    • The proposed method successfully extracts regression rules from neural networks.
    • The extracted rules are accurate and exhibit interesting patterns.
    • Experiments on four datasets demonstrate the method's effectiveness.

    Conclusions:

    • The framework provides an effective way to interpret neural networks trained on mixed data.
    • The extracted rules offer a balance between accuracy and comprehensibility.
    • This work contributes to the field of explainable artificial intelligence (XAI).