HUMAP: Hierarchical Uniform Manifold Approximation and Projection
This work addresses the need for flexible and effective hierarchical visualization in science domains, though it appears incremental as it builds on existing dimensionality reduction approaches.
The authors tackled the problem of visualizing high-dimensional data with multiple granularities by introducing HUMAP, a hierarchical dimensionality reduction technique that preserves both local and global structures and maintains mental map consistency during exploration, showing empirical superiority over existing hierarchical methods.
Dimensionality reduction (DR) techniques help analysts to understand patterns in high-dimensional spaces. These techniques, often represented by scatter plots, are employed in diverse science domains and facilitate similarity analysis among clusters and data samples. For datasets containing many granularities or when analysis follows the information visualization mantra, hierarchical DR techniques are the most suitable approach since they present major structures beforehand and details on demand. This work presents HUMAP, a novel hierarchical dimensionality reduction technique designed to be flexible on preserving local and global structures and preserve the mental map throughout hierarchical exploration. We provide empirical evidence of our technique's superiority compared with current hierarchical approaches and show a case study applying HUMAP for dataset labelling.