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

Strong feature sets from small samples.

Seungchan Kim¹, Edward R Dougherty, Junior Barrera

¹Department of Electrical Engineering, Texas A&M University, College Station, TX 77840, USA.

Journal of Computational Biology : a Journal of Computational Molecular Cell Biology

|March 26, 2002

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

Longitudinal Single-Cell RNA-Sequencing Reveals Evolution of Micro- and Macro-states in Chronic Myeloid Leukemia.

Cancer research·2026

Same author

A Model for Clinician and Staff Education in the Engagement of American Indians in Genomic Research.

Journal of cancer education : the official journal of the American Association for Cancer Education·2026

Same author

Genomic Landscape Analysis of Canine Pulmonary Adenocarcinoma Reveals Candidate Targetable Gene Fusions.

Veterinary and comparative oncology·2026

Same author

The rise of astrocytes: are they guardians or troublemakers of the brain disorder?

Experimental & molecular medicine·2026

Same author

Identifying Risk Factors for Dental Neglect in Children Who Failed to Complete Their Dental Surgery Appointments in Northeast Ohio: A Retrospective Study.

Children (Basel, Switzerland)·2025

Same author

A Concordance Study among 26 NGS Laboratories Participating in the NCI Molecular Analysis for Therapy Choice Clinical Trial.

Clinical cancer research : an official journal of the American Association for Cancer Research·2025

Same journal

Mosquito Species and Gender Identification System Based on Artificial Intelligence and Image Processing Methods.

Journal of computational biology : a journal of computational molecular cell biology·2026

Same journal

GMSA: A Graph Matching and Point Cloud Registration-Based Method for Spatial Transcriptomics Data Alignment.

Journal of computational biology : a journal of computational molecular cell biology·2026

Same journal

Investigations on Multiple Protein Scaffold Filling.

Journal of computational biology : a journal of computational molecular cell biology·2026

Same journal

Cell Type Prediction for Single-Cell RNA Sequencing Utilizing Unsupervised Domain Adaptation and Semi-Supervised Learning.

Journal of computational biology : a journal of computational molecular cell biology·2026

Same journal

PPIGAN: Prediction of Protein-Protein Interactions Using Generative Adversarial Networks.

Journal of computational biology : a journal of computational molecular cell biology·2026

Same journal

Deep Structure-Enhanced Cell Clustering Model for Single-Cell RNA Sequencing Data.

Journal of computational biology : a journal of computational molecular cell biology·2026

See all related articles

This study introduces a novel method to improve classifier accuracy with small datasets by spreading sample points. This technique enhances feature selection for robust cancer classification, even with limited data.

Area of Science:

Bioinformatics
Machine Learning
Computational Biology

Background:

Classifier design algorithms struggle with overfitting in small sample sizes.
Error estimators for small samples can be unbiased but have high variance, leading to optimistic estimates.
Existing methods lack robustness when dealing with limited data for classification tasks.

Purpose of the Study:

To mitigate the small-sample problem in classifier design.
To develop a method for robust feature selection in the presence of limited data.
To improve the accuracy of cancer classification using gene expression data.

Main Methods:

Proposing a novel algorithm that designs classifiers from a probability distribution by spreading sample points.
Parameterizing the algorithm by the variance of the spreading distribution to control classification difficulty.

Related Experiment Videos

Analytically deriving linear classifiers for computational efficiency.

Main Results:

The algorithm identifies gene sets with classification accuracy robust to increased data spreading.
The error measure quantifies feature set strength as a function of data spread.
The method successfully distinguishes between BRCA1/BRCA2 tumors using gene expression data from cDNA microarrays.

Conclusions:

The proposed method effectively addresses the small-sample problem in classifier design.
This approach yields reliable feature sets for classification beyond the original sample data.
The algorithm demonstrates practical application in cancer classification, specifically for BRCA1/BRCA2 tumors.