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

LDA/SVM driven nearest neighbor classification.

Jing Peng1, D R Heisterkamp, H K Dai

  • 1Electr. Eng. and Comput. Sci. Dept., Tulane Univ., New Orleans, LA, USA.

IEEE Transactions on Neural Networks
|February 2, 2008
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

Simulating rainfall runoff and assessing low impact development (LID) facilities in sponge airport.

Water science and technology : a journal of the International Association on Water Pollution Research·2020
Same author

Resolution Improvement of Light Field Imaging via a Nematic Liquid Crystal Microlens with Added Multi-Walled Carbon Nanotubes.

Sensors (Basel, Switzerland)·2020
Same author

The Recommendations for the Management of Chinese Children With Epilepsy During the COVID-19 Outbreak.

Frontiers in pediatrics·2020
Same author

Effects of caprolactam content on curdlan-based food packaging film and detection by infrared spectroscopy.

Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2020
Same author

MicroRNA-Related Prognosis Biomarkers from High-Throughput Sequencing Data of Colorectal Cancer.

BioMed research international·2020
Same author

The effect of qigong for pulmonary function and quality of life in patients with covid-19: A protocol for systematic review and meta-analysis.

Medicine·2020
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

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

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
Same journal

Data-based identification and control of nonlinear systems via piecewise affine approximation.

IEEE transactions on neural networks·2011
Same journal

Data-core-based fuzzy min-max neural network for pattern classification.

IEEE transactions on neural networks·2011
See all related articles

Nearest neighbor classification struggles with high-dimensional data. Our adaptive neighborhood morphing method reduces bias by adjusting feature space, improving classification accuracy.

Area of Science:

  • Machine Learning
  • Pattern Recognition
  • Data Science

Background:

  • Nearest Neighbor (NN) classification assumes local constancy of class conditional probabilities.
  • The curse of dimensionality with finite samples violates this assumption in high dimensions.
  • NN classification introduces significant bias under these high-dimensional conditions.

Purpose of the Study:

  • To propose a novel locally adaptive neighborhood morphing classification method.
  • To minimize classification bias in high-dimensional spaces.
  • To enhance classification performance by adapting to local data structures.

Main Methods:

  • Utilizing local support vector machine (SVM) learning to estimate an effective metric.
  • Morphing neighborhoods to be elongated along less discriminant dimensions and constricted along more discriminant ones.

Related Experiment Videos

  • Modifying neighborhoods to approximate local constancy of class conditional probabilities.

    • Main Results:

    • Demonstrated reduction in classification bias.
    • Achieved improved classification performance compared to traditional methods.
    • Validated efficacy across multiple diverse datasets.

    Conclusions:

    • Locally adaptive neighborhood morphing offers a robust solution to NN classification limitations in high dimensions.
    • The method effectively addresses the curse of dimensionality by adapting feature space.
    • This approach enhances predictive accuracy by ensuring more stable probability estimates within modified neighborhoods.