skip to main content
Language:
Search Limited to: Search Limited to: Resource type Show Results with: Show Results with: Search type Index

Classes are not Clusters: Improving Label-based Evaluation of Dimensionality Reduction

IEEE transactions on visualization and computer graphics, 2024-01, p.1-11 [Peer Reviewed Journal]

ISSN: 1077-2626 ;EISSN: 1941-0506 ;DOI: 10.1109/TVCG.2023.3327187 ;CODEN: ITVGEA

Digital Resources/Online E-Resources

Citations Cited by
  • Title:
    Classes are not Clusters: Improving Label-based Evaluation of Dimensionality Reduction
  • Author: Jeon, Hyeon ; Kuo, Yun-Hsin ; Aupetit, Michael ; Ma, Kwan-Liu ; Seo, Jinwook
  • Subjects: Clustering ; Clustering Validation Measures ; Degradation ; Dimensionality reduction ; Dimensionality Reduction Evaluation ; Distortion measurement ; Extraterrestrial measurements ; Nonlinear distortion ; Reliability ; Scalability
  • Is Part Of: IEEE transactions on visualization and computer graphics, 2024-01, p.1-11
  • Description: A common way to evaluate the reliability of dimensionality reduction (DR) embeddings is to quantify how well labeled classes form compact, mutually separated clusters in the embeddings. This approach is based on the assumption that the classes stay as clear clusters in the original high-dimensional space. However, in reality, this assumption can be violated; a single class can be fragmented into multiple separated clusters, and multiple classes can be merged into a single cluster. We thus cannot always assure the credibility of the evaluation using class labels. In this paper, we introduce two novel quality measures- Label-Trustworthiness and Label-Continuity (Label-T&C)-advancing the process of DR evaluation based on class labels. Instead of assuming that classes are well-clustered in the original space, Label-T&C work by (1) estimating the extent to which classes form clusters in the original and embedded spaces and (2) evaluating the difference between the two. A quantitative evaluation showed that Label-T&C outperform widely used DR evaluation measures (e.g., Trustworthiness and Continuity, Kullback-Leibler divergence) in terms of the accuracy in assessing how well DR embeddings preserve the cluster structure, and are also scalable. Moreover, we present case studies demonstrating that Label-T&C can be successfully used for revealing the intrinsic characteristics of DR techniques and their hyperparameters.
  • Publisher: IEEE
  • Language: English
  • Identifier: ISSN: 1077-2626
    EISSN: 1941-0506
    DOI: 10.1109/TVCG.2023.3327187
    CODEN: ITVGEA
  • Source: IEEE Open Access Journals
Back to results list
Go to Previous pagePrevious Result  4 NextGo to Next page
INSPIRE LIBRARY - TON DUC THANG UNIVERSITY (84-028) 37 755 057 Feedback
19 Nguyen Huu Tho St. Dist.7, HCM thuvien@tdtu.edu.vn

Copyright © 2017 INSPIRE LIBRARY - TON DUC THANG UNIVERSITY

Searching Remote Databases, Please Wait