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

Large margin nearest neighbor classifiers.

Carlotta Domeniconi1, Dimitrios Gunopulos, Jing Peng

  • 1Computer Science Department, University of California, Riverside, CA 92521, USA. carlotta@ise.gmu.edu

IEEE Transactions on Neural Networks
|August 27, 2005
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

Gradient-Based Local Causal Structure Learning.

IEEE transactions on cybernetics·2023
Same author

Isoform function prediction by Gene Ontology embedding.

Bioinformatics (Oxford, England)·2022
Same author

Personalized Federated Few-Shot Learning.

IEEE transactions on neural networks and learning systems·2022
Same author

Artificial Intelligence for Hospital Health Care: Application Cases and Answers to Challenges in European Hospitals.

Healthcare (Basel, Switzerland)·2021
Same author

DMIL-IsoFun: predicting isoform function using deep multi-instance learning.

Bioinformatics (Oxford, England)·2021
Same author

Multiview Multi-Instance Multilabel Active Learning.

IEEE transactions on neural networks and learning systems·2021
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

This study introduces a novel nearest neighbor classification method using support vector machines (SVMs) to create adaptive metrics. This approach effectively reduces bias caused by the curse of dimensionality in machine learning.

Area of Science:

  • Machine Learning
  • Pattern Recognition
  • Data Science

Background:

  • The nearest neighbor technique is a common classification method but suffers from bias in high-dimensional data due to the curse of dimensionality.
  • Locally adaptive metrics are essential to mitigate bias by maintaining uniform class conditional probabilities.
  • Existing methods for locally adaptive nearest neighbor classification can be computationally intensive.

Purpose of the Study:

  • To propose a novel technique for nearest neighbor classification that employs a locally adaptive metric.
  • To leverage Support Vector Machines (SVMs) for constructing a flexible, locally adaptive metric.
  • To enhance classification accuracy and computational efficiency in high-dimensional spaces.

Main Methods:

  • A locally flexible metric is computed using Support Vector Machines (SVMs).

Related Experiment Videos

  • The SVM decision function identifies discriminant directions within local neighborhoods.
  • This results in a local feature weighting scheme, effectively adapting the metric to the data distribution.
  • The method formally increases the margin in the weighted classification space.

    • Main Results:

    • The proposed method effectively minimizes bias in nearest neighbor classification, particularly in high-dimensional settings.
    • Demonstrated online computational efficiency compared to other locally adaptive techniques.
    • Validation of the method's efficacy using both simulated and real-world datasets.

    Conclusions:

    • The SVM-based locally adaptive metric offers a significant improvement for nearest neighbor classification.
    • This technique effectively addresses the curse of dimensionality and reduces estimation bias.
    • The method provides an efficient and accurate solution for classification tasks in complex data environments.