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

Handling missing values in support vector machine classifiers.

K Pelckmans1, J De Brabanter, J A K Suykens

  • 1Katholieke Universiteit Leuven, ESAT-SCD/SISTA, Kasteelpark Arenberg 10, B-3001 Leuven, Belgium. kristiaan.pelckmans@esat.kuleuven.ac.be

Neural Networks : the Official Journal of the International Neural Network Society
|August 23, 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

Monitoring the evolution of free and cysteinylated aldehydes from malt to fresh and forced aged beer.

Food research international (Ottawa, Ont.)·2021
Same author

First-trimester intrauterine hematoma and pregnancy complications.

Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology·2019
Same author

Validation of an ultra-high-performance liquid chromatography-mass spectrometry method for the quantification of cysteinylated aldehydes and application to malt and beer samples.

Journal of chromatography. A·2019
Same author

Early-pregnancy events and subsequent antenatal, delivery and neonatal outcomes: prospective cohort study.

Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology·2019
Same author

Ultrasound characteristics of endometrial cancer as defined by International Endometrial Tumor Analysis (IETA) consensus nomenclature: prospective multicenter study.

Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology·2017
Same author

International Endometrial Tumor Analysis (IETA) terminology in women with postmenopausal bleeding and sonographic endometrial thickness ≥ 4.5 mm: agreement and reliability study.

Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology·2017
Same journal

Anchor-based disentanglement framework for incremental multi-view clustering.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Complex-valued amplitude-phase interference modeling for adversarially robust classification.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

TraNce: Type-aware hypergraph neural network with biological mediators for drug repositioning.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Decentralized ADMM for factorization-based Low-rank matrix estimation.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Memristive neuromorphic circuit design inspired by the neural mechanisms of conditioned fear.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Q-learning based asynchronous Boolean control networks stabilization with data loss.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

This study introduces a new method for building classifiers with missing input data. The approach handles uncertainty from missing values, generalizing mean imputation and standard Support Vector Machines (SVMs).

Area of Science:

  • Machine Learning
  • Statistical Learning Theory

Background:

  • Handling missing data is crucial in machine learning.
  • Standard algorithms often struggle with incomplete datasets.
  • Existing methods like mean imputation can introduce bias.

Purpose of the Study:

  • To develop a robust non-parametric classifier for data with missing inputs.
  • To generalize existing imputation techniques and Support Vector Machines (SVMs).
  • To extend the method for multivariate additive models.

Main Methods:

  • A modified risk function is defined to account for output uncertainty due to missing values.
  • A non-parametric approach is adopted.
  • Componentwise kernel machines are used for additive models.

Related Experiment Videos

Main Results:

  • The proposed method generalizes mean imputation in the linear case.
  • The kernel machine reduces to a standard SVM when no data is missing.
  • An efficient implementation is derived using the Least Squares Support Vector Machine (LS-SVM) formulation.

Conclusions:

  • The developed method provides a principled way to handle missing data in classification.
  • It offers a flexible framework applicable to both linear and non-linear models.
  • The LS-SVM based implementation ensures computational efficiency.