Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

L2-norm multiple kernel learning and its application to biomedical data fusion.

Shi Yu¹, Tillmann Falck, Anneleen Daemen

¹Bioinformatics Group, Department of Electrical Engineering, Katholieke Universiteit Leuven, Kasteelpark Arenberg 10, Heverlee B-3001, Belgium. shee.yu@gmail.com

BMC Bioinformatics

|June 10, 2010

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

The effect of acute fatigue on somatosensory function differs between trained and untrained individuals.

Scientific reports·2026

Same author

Macromolecular crowding reshapes the conformational landscapes of intrinsically disordered proteins: mechanisms, cellular contexts, and functional consequences.

Current opinion in structural biology·2026

Same author

Force Field Evaluation for an Intrinsically Disordered Domain: MD-NMR-FCS Benchmarking of Protein 4.1G Headpiece Ensembles.

Journal of chemical information and modeling·2026

Same author

Hydrogen-bonded Organic Frameworks Decorated Polymeric High Internal Phase Emulsions as Adsorbents for Solid-Phase Extraction of Estrogens.

Journal of separation science·2026

Same author

Profiling of porcine B-cell receptor heavy-chain repertoires indicates the development of a wide public pseudorabies virus-specific immune response after vaccination and challenge.

Discovery immunology·2026

Same author

Concanavalin A targets phylogenetically conserved N-linked glycans on coronavirus spike proteins for broad-spectrum antiviral activity.

Journal of virology·2026

Same journal

Covariance decomposition for distance based species tree estimation.

BMC bioinformatics·2026

Same journal

SNPio: a Python interface for population genomic data processing.

BMC bioinformatics·2026

Same journal

SpaHNR: a spatial domain identification method via sparse attention-based hierarchical node representation and multi-view contrastive learning.

BMC bioinformatics·2026

Same journal

OpenIMC: an open-source platform for analyzing single-cell and spatial proteomics by imaging mass cytometry.

BMC bioinformatics·2026

Same journal

NAP: an open source pipeline for cross-domain microbiome profiling using Nanopore sequencing-derived amplicon data.

BMC bioinformatics·2026

Same journal

SurvGME: an R package for survival analysis with graphical and measurement error models.

BMC bioinformatics·2026

See all related articles

This study introduces L2 Multiple Kernel Learning (MKL), a novel method for combining heterogeneous data sources. L2 MKL offers advantages over sparse methods by allowing non-sparse kernel coefficients, improving performance in biomedical applications.

Area of Science:

Machine Learning
Computational Biology
Statistical Learning

Background:

Introduces optimizing different norms in the dual problem of support vector machines with multiple kernels.
Discusses extensions of multiple kernel learning (MKL) such as L(infinity), L1, and L2 MKL.
Highlights L2 MKL as a novel method yielding non-sparse optimal kernel coefficients, distinct from sparse coefficients in L(infinity) MKL.

Purpose of the Study:

To theoretically analyze the relationship between dual L2 kernel optimization and primal L2 coefficient regularization.
To extend the L2 MKL method to various machine learning problems.
To implement and evaluate L2 MKL for ranking and classification, comparing it with L1 and L(infinity) MKL methods.

Main Methods:

Theoretical analysis of dual L2 optimization and primal L2 regularization.

Related Experiment Videos

Implementation of L2 MKL for ranking and classification tasks.

Development of novel L2 MKL least squares support vector machine (LSSVM) algorithms for computational efficiency.

Main Results:

L2 MKL demonstrates superior performance on most benchmark datasets compared to L1 and L(infinity) MKL.
The proposed L2 MKL LSSVM algorithm shows efficiency and promise for large-scale dataset processing.
Theoretical analysis provides a unified view on MKL and enables broader application of the L2 method.

Conclusions:

L2 MKL extends genomic data fusion, allowing non-sparse weights for potentially more relevant data source integration.
The L2 kernel optimization framework is applicable to diverse algorithms including ranking, classification, regression, and clustering.
LSSVM-based MKL algorithms offer efficient solutions for large-scale MKL problems, with performance comparable to conventional SVM MKL.