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

Multiclass posterior probability support vector machines.

Mehmet Gonen1, Ayşe Gönül Tanugur, Ethem Alpaydin

  • 1Department of Computer Engineering, Boğaziçi University, Instanbul, Turkey. gonen@boun.edu.tr

IEEE Transactions on Neural Networks
|February 14, 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

Crimean-Congo Hemorrhagic Fever in Türkiye, 2017-2024: Evolving Epidemiology and Implications for Surveillance.

International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases·2026
Same author

Nicotinamide N-methyltransferase as a therapeutic target in taxane-resistant castration-resistant prostate cancer.

Cell death discovery·2026
Same author

Spatial prediction of COVID-19 pandemic dynamics in the United States.

ISPRS international journal of geo-information·2025
Same author

The Analysis of Missed Antibiotic De-Escalation Opportunities in Gram-Negative Bloodstream Infections.

Antibiotics (Basel, Switzerland)·2025
Same author

Interpretable and integrative analysis of single-cell multiomics with scMKL.

Communications biology·2025
Same author

Transfer learning framework via Bayesian group factor analysis incorporating feature-wise dependencies.

bioRxiv : the preprint server for biology·2025
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

The posterior probability support vector machine (PPSVM) offers improved accuracy and robustness against noise by using soft labels. This enhanced model requires fewer support vectors compared to traditional methods.

Area of Science:

  • Machine Learning
  • Pattern Recognition
  • Data Mining

Background:

  • Traditional Support Vector Machines (SVMs) can be sensitive to noise and outliers.
  • Existing posterior probability SVM (PPSVM) models offer robustness but are limited to binary classification and specific density estimators.

Purpose of the Study:

  • To extend the posterior probability support vector machine (PPSVM) to multiclass classification.
  • To introduce a novel neighbor-based density estimator for improved posterior probability estimation.
  • To evaluate the performance and efficiency of the proposed multiclass PPSVM.

Main Methods:

  • Developed a multiclass extension of the posterior probability support vector machine (PPSVM).
  • Implemented a neighbor-based density estimator to derive soft labels from posterior probabilities.

Related Experiment Videos

  • Conducted bias-variance analysis to understand error reduction mechanisms.
  • Evaluated performance on 20 benchmark datasets against canonical SVM.

    • Main Results:

    • The proposed multiclass PPSVM achieves higher or comparable accuracy to canonical SVMs.
    • PPSVM demonstrates increased robustness to noise and outliers.
    • The method significantly reduces the number of support vectors required.
    • Bias-variance analysis indicates error reduction is primarily due to decreased bias.

    Conclusions:

    • The multiclass PPSVM with a neighbor-based density estimator is an effective extension for robust classification.
    • This approach offers a more efficient alternative to canonical SVMs by reducing support vector usage.
    • The findings suggest PPSVM is a promising direction for handling noisy and complex datasets.