Pattern or Artifact? Interactively Exploring Embedding Quality with TRACE
This addresses the challenge for data analysts in evaluating embedding quality to avoid misinterpretations, though it is incremental as it builds on existing quality metrics with a new interactive tool.
The paper tackles the problem of assessing the quality of 2D embeddings from dimensionality reduction, where visual structures can be misleading if objectives like preserving local or global structure are not uniformly achieved, by presenting TRACE, a tool that provides an interactive interface to compute and explore local and global quality measures, enabling analysts to choose suitable methods.
This paper presents TRACE, a tool to analyze the quality of 2D embeddings generated through dimensionality reduction techniques. Dimensionality reduction methods often prioritize preserving either local neighborhoods or global distances, but insights from visual structures can be misleading if the objective has not been achieved uniformly. TRACE addresses this challenge by providing a scalable and extensible pipeline for computing both local and global quality measures. The interactive browser-based interface allows users to explore various embeddings while visually assessing the pointwise embedding quality. The interface also facilitates in-depth analysis by highlighting high-dimensional nearest neighbors for any group of points and displaying high-dimensional distances between points. TRACE enables analysts to make informed decisions regarding the most suitable dimensionality reduction method for their specific use case, by showing the degree and location where structure is preserved in the reduced space.