embComp: Visual Interactive Comparison of Vector Embeddings

Florian Heimerl, Christoph Kralj, Torsten Möller, Michael Gleicher

View presentation:2021-10-29T16:00:00ZGMT-0600Change your timezone on the schedule page
2021-10-29T16:00:00Z
Exemplar figure, but none was provided by the authors
Fast forward

Direct link to video on YouTube: https://youtu.be/eU_KgecHTQE

Keywords

I.6.9.c Information visualization < I.6.9 Visualization < I.6 Simulation, Modeling, and Visualization < I Computing Methodologie, L.3.0.f Human- computer interaction < L.3.0 Integrating touch-based interactions into various domains Assistive technology < L.3, H.5.2 User Interfaces < H.5 Information Interfaces and Representation (HCI) < H Information Technology and Systems

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.