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

Learning weighted metrics to minimize nearest-neighbor classification error.

Roberto Paredes1, Enrique Vidal

  • 1Departamento de Sistemas Informáticos y Computación, Instituto Tecnológico de Informática, Universidad Politiécnica de Valencia, Spain. rparedes@iti.upv.es

IEEE Transactions on Pattern Analysis and Machine Intelligence
|June 24, 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

The Relentless Progression of Metabolic Dysfunction-associated Steatotic Liver Disease in Liver Transplantation in Spain.

Transplantation·2025
Same author

EWS::FLI1 expression in human embryonic mesenchymal stem cells leads to transcriptional reprograming, defective DNA damage repair and Ewing sarcoma.

Nature communications·2025
Same author

<i>BRCA2</i> Pre-mRNA Differential 5' Splicing: A Rescue of Functional Protein Properties from Pathogenic Gene Variants and a Lifeline for Fanconi Anemia D1 Patients.

International journal of molecular sciences·2025
Same author

Assessment of sustainable compositions to reduce emissions and sound pressure level of flash powder.

Environmental pollution (Barking, Essex : 1987)·2025
Same author

External validation of models to predict hepatocellular carcinoma in Hepatitis C Virus cured F3-F4 patients.

United European gastroenterology journal·2024
Same author

Improving Radiology Report Generation Quality and Diversity through Reinforcement Learning and Text Augmentation.

Bioengineering (Basel, Switzerland)·2024
Same journal

Benchmarking the Robustness of Autonomous Driving to Environmental Illusions: A Lane Perception Perspective.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Learning Topology-Aware Representations via Test-Time Adaptation for Anomaly Segmentation.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces a weighted distance approach to enhance Nearest-Neighbor classification accuracy. The method learns optimal weights, improving classification performance on diverse datasets, including sparse text data.

Area of Science:

  • Machine Learning
  • Pattern Recognition
  • Data Mining

Background:

  • Nearest-Neighbor (NN) classification is a fundamental algorithm.
  • Standard NN methods can be sensitive to feature weighting and data distribution.
  • Optimizing NN accuracy requires effective weight learning strategies.

Purpose of the Study:

  • To propose a novel weighted distance metric for Nearest-Neighbor classification.
  • To develop algorithms for automatic learning of class- and feature-specific weights.
  • To evaluate the proposed method's performance against existing techniques.

Main Methods:

  • A weighted distance metric is introduced for NN classification.
  • Learning algorithms are derived by minimizing Leaving-One-Out (LOO) classification error.

Related Experiment Videos

  • Weights can be customized per class, feature, or prototype.

    • Main Results:

    • The proposed weighted distance approach demonstrated uniformly good performance across experiments.
    • Results on UCI/STATLOG corpora and text classification tasks were competitive or superior to state-of-the-art.
    • The method effectively handles sparse data and high dimensionality in text classification.

    Conclusions:

    • The proposed weighted distance method significantly optimizes Nearest-Neighbor classification accuracy.
    • Automatic weight learning provides a robust and adaptable solution for various datasets.
    • This approach offers a valuable improvement for classification tasks, especially those with complex data characteristics.