embComp: Visual Interactive Comparison of Vector Embeddings

TL;DR

embComp is a visual analysis system designed to compare and understand differences between vector embedding spaces, aiding interpretation of similarities in various applications.

Contribution

It introduces a comprehensive visual analysis approach for comparing embedding spaces, including local and global metrics, and supports diverse analysis workflows.

Findings

Effective in analyzing corpus differences via word embeddings

Helps understand algorithmic variations in embedding generation

Provides intuitive visual tools for embedding comparison

Abstract

This paper introduces embComp, a novel approach for comparing two embeddings that capture the similarity between objects, such as word and document embeddings. We survey scenarios where comparing these embedding spaces is useful. From those scenarios, we derive common tasks, introduce visual analysis methods that support these tasks, and combine them into a comprehensive system. One of embComp's central features are overview visualizations that are based on metrics for measuring differences in the local structure around objects. Summarizing these local metrics over the embeddings provides global overviews of similarities and differences. Detail views allow comparison of the local structure around selected objects and relating this local information to the global views. Integrating and connecting all of these components, embComp supports a range of analysis workflows that help understand similarities and differences between embedding spaces. We assess our approach by applying it in several use cases, including understanding corpora differences via word vector embeddings, and understanding algorithmic differences in generating embeddings.