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

1 code implementation • 1 Aug 2023 • Hyeon Jeon, Yun-Hsin Kuo, Michaël Aupetit, Kwan-Liu Ma, Jinwook Seo

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.

Dimensionality Reduction

ClustML: A Measure of Cluster Pattern Complexity in Scatterplots Learnt from Human-labeled Groupings

no code implementations • 1 Jun 2021 • Mostafa M. Abbas, Ehsan Ullah, Abdelkader Baggag, Halima Bensmail, Michael Sedlmair, Michaël Aupetit

We propose a new VQM for visual grouping patterns in scatterplots, called ClustML, which is trained on previously collected human subject judgments.