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

The nearest subclass classifier: a compromise between the nearest mean and nearest neighbor classifier.

Cor J Veenman1, Marcel J T Reinders

  • 1Department of Mediamatics, Delft University of Technology, P.O. Box 5031, 2600 GA Delft, The Netherlands. C.J.Veenman@ewi.tudelf.nl

IEEE Transactions on Pattern Analysis and Machine Intelligence
|September 22, 2005
PubMed
Summary

The Nearest Subclass Classifier (NSC) offers a flexible yet robust approach to classification. This prototype-based method balances performance and efficiency, unifying nearest neighbor and nearest mean classifier properties.

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

Precision-guided immunomodulatory therapy in sepsis.

The Lancet. Respiratory medicine·2026
Same author

On the state of protein function prediction: a report on the fourth CAFA challenge.

bioRxiv : the preprint server for biology·2026
Same author

Effects of a Combined Dietary and Physical Activity Intervention on Bone Density, Lean Mass and Fat Mass in Adults: The GOTO Trial.

Journal of cachexia, sarcopenia and muscle·2026
Same author

Decoding exon inclusion in the human brain reveals more divergent splicing mechanisms in neurons than glia.

Genome biology·2026
Same author

Domain mapping of disease mutations reveals pathogenic SORL1 variants in Alzheimer's disease.

Molecular neurodegeneration·2025
Same author

All-atom protein sequence design using discrete diffusion models.

Journal of cheminformatics·2025

Area of Science:

  • Machine Learning
  • Computer Science
  • Data Mining

Background:

  • Nearest neighbor and nearest mean classifiers are widely used but have limitations.
  • Prototype-based classifiers offer an alternative but require careful parameter tuning.

Purpose of the Study:

  • To introduce the Nearest Subclass Classifier (NSC), a novel classification algorithm.
  • To unify the strengths of nearest neighbor and nearest mean classifiers.
  • To provide a regularized, prototype-based classification method.

Main Methods:

  • The NSC algorithm is based on the Maximum Variance Cluster algorithm.
  • It utilizes a variance constraint parameter for regularization, preventing overfitting.
  • The number of prototypes can range from the entire training set to one per class.

Related Experiment Videos

Main Results:

  • The NSC demonstrated comparable performance to the k-nearest neighbor classifier on several datasets.
  • It achieved favorable data set compression ratios.
  • The NSC offers good classification speed and low storage requirements.

Conclusions:

  • The NSC presents a valuable compromise between classification performance and computational efficiency.
  • It serves as a versatile, regularized, prototype-based classifier.
  • The NSC is a promising alternative for various classification tasks.