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

A kernel-based two-class classifier for imbalanced data sets.

Xia Hong¹, Sheng Chen, Chris J Harris

¹Cybernetic Intelligence Research Group, School of Systems Engineering, University of Reading, Reading RG6 6AY, UK. x.hong@reading.ac.uk

IEEE Transactions on Neural Networks

|February 7, 2007

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

Adaptive Multioutput Gradient RBF Tracker for Nonlinear and Nonstationary Regression.

IEEE transactions on cybernetics·2023

Same author

Deep Cascade Gradient RBF Networks With Output-Relevant Feature Extraction and Adaptation for Nonlinear and Nonstationary Processes.

IEEE transactions on cybernetics·2022

Same author

Multi-Output Selective Ensemble Identification of Nonlinear and Nonstationary Industrial Processes.

IEEE transactions on neural networks and learning systems·2020

Same author

Characterization of extended-spectrum β-lactamases (ESBLs)-producing Salmonella in retail raw chicken carcasses.

International journal of food microbiology·2017

Same author

Sclareolide enhances gemcitabine‑induced cell death through mediating the NICD and Gli1 pathways in gemcitabine‑resistant human pancreatic cancer.

Molecular medicine reports·2017

Same author

Improving the reversibility of thermal denaturation and catalytic efficiency of Bacillus licheniformis α-amylase through stabilizing a long loop in domain B.

PloS one·2017

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

This study introduces a new kernel classifier algorithm for imbalanced datasets. It optimizes model generalization using orthogonal forward selection and a novel weighted least squares estimator, improving performance metrics.

Area of Science:

Machine Learning
Computational Statistics

Background:

Traditional kernel classifiers often use classification accuracy, which is problematic for imbalanced datasets.
Equal weighting of data samples in parameter estimation can lead to suboptimal models with imbalanced data.

Purpose of the Study:

To present a kernel classifier construction algorithm optimized for imbalanced two-class datasets.
To enhance model generalization by addressing the limitations of standard evaluation metrics and weighting schemes.

Main Methods:

Utilizing orthogonal forward selection (OFS) for kernel classifier construction.
Employing a novel regularized orthogonal weighted least squares (ROWLS) estimator.
Adopting maximal leave-one-out area under the curve (LOO-AUC) of receiver operating characteristics (ROCs) as the model selection criterion.

Related Experiment Videos

Main Results:

The proposed algorithm calculates LOO-AUC analytically using the ROWLS estimator, eliminating the need for data splitting.
Orthogonalization procedure enables efficient LOO-AUC calculation.
Forward recursive updating formulas minimize computational expense during model term searching.

Conclusions:

The presented kernel classifier algorithm effectively optimizes model generalization for imbalanced datasets.
The ROWLS estimator and OFS provide an efficient and accurate method for constructing robust classifiers.
Numerical examples validate the algorithm's efficacy in improving performance on imbalanced data.