DimenFix: a Novel Meta-Dimensionality Reduction Strategy for Feature Preservation

Simple item page
dc.contributor.authorLuo, Qiaodanen_US
dc.contributor.authorChristino, Leonardoen_US
dc.contributor.authorMilios, Evangelosen_US
dc.contributor.authorPaulovich, Fernando V.en_US
dc.contributor.editorEl-Assady, Mennatallahen_US
dc.contributor.editorSchulz, Hans-Jörgen_US
dc.date.accessioned2024-05-21T08:29:58Z
dc.date.available2024-05-21T08:29:58Z
dc.date.issued2024
dc.description.abstractDimensionality Reduction (DR) methods have become essential tools for the data analysis toolbox. Typically, DR methods combine features of a multi-variate dataset to produce dimensions in a reduced space, preserving some data properties, usually pairwise distances or local neighborhoods. Preserving such properties makes DR methods attractive, but it is also one of their weaknesses. When calculating the embedded dimensions, through usually non-linear strategies, the original feature values are lost and not explicitly represented in the spatialization of the produced layouts, making it challenging to verify the features' contribution to the attained representations. Some strategies have been proposed to tackle this issue, such as coloring the DR layout or generating explanations. Still, they are post-processes, so specific features (values) are not guaranteed to be preserved or represented. This paper proposes DimenFix, a novel meta-DR strategy that explicitly preserves the values of a particular feature or external data (e.g., class, time, or ranking) in one of the embedded dimensions. DimenFix works with virtually any gradient-descent DR method and, in our results, has shown to be capable of representing features without heavily impacting distance or neighborhood preservation, allowing for creating hybrid layouts joining characteristics of scatter plots and DR methods.en_US
dc.description.sectionheadersVisual Analytics Methods and Approaches
dc.description.seriesinformationEuroVis Workshop on Visual Analytics (EuroVA)
dc.identifier.doi10.2312/eurova.20241110
dc.identifier.isbn978-3-03868-253-0
dc.identifier.pages6 pages
dc.identifier.urihttps://doi.org/10.2312/eurova.20241110
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/eurova20241110
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Mathematics of computing → Dimensionality reduction; Computing methodologies → Visual analytics
dc.subjectMathematics of computing → Dimensionality reduction
dc.subjectComputing methodologies → Visual analytics
dc.titleDimenFix: a Novel Meta-Dimensionality Reduction Strategy for Feature Preservationen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
04_eurova20241110.pdf
Size:
3.75 MB
Format:
Adobe Portable Document Format
Download
Collections
EuroVA2024