Title | ||
---|---|---|
Semantic Concept Spaces: Guided Topic Model Refinement using Word-Embedding Projections. |
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)
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">understand</italic>
the semantic space of the model, (2)
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">identify</italic>
regions of potential conflicts and problems, and (3)
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">readjust</italic>
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
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">concept regions</italic>
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. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/TVCG.2019.2934654 | IEEE transactions on visualization and computer graphics |
Keywords | Field | DocType |
Semantics,Analytical models,Computational modeling,Visual analytics,Machine learning,Task analysis | Computer science,Theoretical computer science,Natural language processing,Artificial intelligence,Word embedding,Topic model | Journal |
Volume | Issue | ISSN |
26 | 1 | 1077-2626 |
Citations | PageRank | References |
4 | 0.42 | 17 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mennatallah El-Assady | 1 | 120 | 13.73 |
Rebecca Kehlbeck | 2 | 4 | 0.76 |
Christopher Collins | 3 | 1037 | 49.74 |
Daniel A. Keim | 4 | 7704 | 1141.60 |
Oliver Deussen | 5 | 2852 | 205.16 |