Semantic Concept Spaces: Guided Topic Model Refinement using Word-Embedding Projections

TL;DR

This paper introduces a visual analytics framework that allows users to refine topic models by manipulating semantic concept regions in word-embedding spaces, improving model quality with minimal feedback.

Contribution

It presents a model-agnostic, interactive system for semantic refinement of topic models using word-embedding projections, guided by user input and recommendations.

Findings

User studies show improved topic model quality.

Semantic concept manipulation enhances model interpretability.

Guided interactions reduce feedback needed for refinement.

Abstract

We present a framework that allows users to incorporate the semantics of their domain knowledge for topic model refinement while remaining model-agnostic. Our approach enables users to (1) understand the semantic space of the model, (2) identify regions of potential conflicts and problems, and (3) readjust the semantic relation of concepts based on their understanding, directly influencing the topic modeling. These tasks are supported by an interactive visual analytics workspace that uses word-embedding projections to define concept regions which can then be refined. The user-refined concepts are independent of a particular document collection and can be transferred to related corpora. All user interactions within the concept space directly affect the semantic relations of the underlying vector space model, which, in turn, change the topic modeling. In addition to direct manipulation, our system guides the users' decision-making process through recommended interactions that point out potential improvements. This targeted refinement aims at minimizing the feedback required for an efficient human-in-the-loop process. We confirm the improvements achieved through our approach in two user studies that show topic model quality improvements through our visual knowledge externalization and learning process.