EUVIMEDObjevte EUVIMED
Crossref

Statistical distances between multivariable distributions for predicting performance of clustering algorithms based on gene expression data of cancers

Oleksiński, Artur, Mateusz Kania, Andrzej Polański

Digital Library of the Silesian University of Technology (Silesian University of Technology) · 2024-01-01

DOI: 10.34918/89991 ↗

At a glance

Journal article in Digital Library of the Silesian University of Technology (Silesian University of Technology), published on 2024-01-01. Written by Oleksiński and 3 co-authors.

AbstraktSubjectsBibliografické údajeCiteRelated studies

Abstract (original language)

This study investigates how the choice of distance metric affects the clustering quality of cancer gene expression data using the Expectation-Maximization (EM) algorithm with Independent Multidimensional Gaussian Mixture Models (IMGMMs).The research focuses on three specific distance metrics: Bhattacharyya distance, Jeffrey divergence, and Jensen-Shannon divergence.Experiments revealed a promising moderate correlation between these distance metrics and the accuracy of the clustering results.This finding suggests a potential link between the separability of the underlying data distributions (measured by distance) and the effectiveness of clustering (measured by accuracy).These results pave the way for further exploration of how distance metrics can be used to optimize cancer gene expression data clustering using EM and IMGMMs.

Subjects

statistical distancedistance measurescancer gene expression dataexpectation-maximization algorithmBhattacharyya distanceJensen-Shannon divergenceclustering accuracy

Cite

APA

Oleksiński, Artur, Kania, M., & Polański, A. (2024). Statistical distances between multivariable distributions for predicting performance of clustering algorithms based on gene expression data of cancers. Digital Library of the Silesian University of Technology (Silesian University of Technology). https://doi.org/10.34918/89991

Vancouver

Oleksiński, Artur, Kania M, Polański A. Statistical distances between multivariable distributions for predicting performance of clustering algorithms based on gene expression data of cancers. Digital Library of the Silesian University of Technology (Silesian University of Technology). 2024. doi:10.34918/89991

BibTeX
@article{Oleksiski2024,
  author = {Oleksiński and Artur and Kania, Mateusz and Polański, Andrzej},
  title = {Statistical distances between multivariable distributions for predicting performance of clustering algorithms based on gene expression data of cancers},
  journal = {Digital Library of the Silesian University of Technology (Silesian University of Technology)},
  year = {2024},
  doi = {10.34918/89991}
}
Start for free →