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

Differential prioritization between relevance and redundancy in correlation-based feature selection techniques for

Chia Huey Ooi1, Madhu Chetty, Shyh Wei Teng

  • 1Gippsland School of Information Technology, Monash University, Churchill, VIC 3842, Australia. Chia.Huey.Ooi@infotech.monash.edu.au

BMC Bioinformatics
|June 27, 2006
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

Editorial: The application of multi-omics analysis in translational medicine.

Frontiers in medicine·2025
Same author

BioBERT based text mining for incorporating prior knowledge in the inference of genetic network models.

Computers in biology and medicine·2025
Same author

BSDR: A Data-Efficient Deep Learning-Based Hyperspectral Band Selection Algorithm Using Discrete Relaxation.

Sensors (Basel, Switzerland)·2024
Same author

Large language model based framework for automated extraction of genetic interactions from unstructured data.

PloS one·2024
Same author

MCSNet+: Enhanced Convolutional Neural Network for Detection and Classification of <i>Tribolium</i> and <i>Sitophilus</i> Sibling Species in Actual Wheat Storage Environments.

Foods (Basel, Switzerland)·2023
Same author

MICFuzzy: A maximal information content based fuzzy approach for reconstructing genetic networks.

PloS one·2023

New feature selection methods improve gene expression classification accuracy by balancing gene relevance and redundancy. These techniques offer better performance on multiclass microarray datasets, including GCM and NCI60, with smaller gene sets.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Microarray datasets contain numerous genes, necessitating effective feature selection for noise reduction and computational efficiency in tissue classification.
  • Existing feature selection methods often fail to account for gene correlations or are computationally expensive and yield unreproducible results.
  • Previous evaluations of gene correlation-aware techniques have been hampered by less rigorous procedures, leading to inflated accuracy estimates.

Purpose of the Study:

  • To introduce and evaluate novel, realistically assessed correlation-based feature selection techniques for multiclass microarray data.
  • To address limitations of existing methods by incorporating a new criterion for balancing gene relevance and redundancy.

Main Methods:

  • Developed two filter-based feature selection techniques incorporating relevance, redundancy, and a novel criterion: degree of differential prioritization (DDP).

Related Experiment Videos

  • DDP allows for flexible optimization of relevance versus redundancy.
  • Evaluated techniques on nine well-known multiclass microarray datasets using rigorous assessment procedures.

    • Main Results:

    • The proposed techniques achieved optimal classification accuracy with reasonably small predictor set sizes across nine multiclass microarray datasets.
    • The degree of differential prioritization (DDP) effectively balanced relevance and redundancy.
    • Demonstrated superior performance compared to previous methods when evaluated realistically.

    Conclusions:

    • The DDP-enhanced filter-based techniques significantly improve classification accuracy for multiclass microarray datasets.
    • Particularly effective for GCM and NCI60 datasets, outperforming prior studies with similar evaluation standards.
    • Offers a more robust and accurate approach to feature selection in gene expression analysis.