Visual Exploration of Relationships and Structure in Low-Dimensional Embeddings

Eckelt, Klaus; Hinterreiter, Andreas; Adelberger, Patrick; Walchshofer, Conny; Dhanoa, Vaishali; Humer, Christina; Heckmann, Moritz; Steinparz, Christian; Streit, Marc

doi:10.31219/osf.io/ujbrs

Cited by 3 publications

(10 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To increase the comprehensiveness of pattern recognition through both topology-and attribute-driven layout, we added a feature to show both layouts simultaneously in a multiple-coordinated view. Additionally, as outlined in Section 4.3.3, we use summary visualizations as suggested by Eckelt et al [13], wherefore we implemented custom visualizations for both user studies, see Figure 5a and 10. All sessions described in this paper can be explored online 3 .…”

Section: Methodsmentioning

confidence: 99%

“…We derive the following requirements for single and multiples sessions from the existing literature [26], [36] and our prior research experience [13], [22], [52]. To support singlesession investigation, Provectories is designed to: S1 show the entire analysis sequence from beginning to end, following the temporal order;…”

Section: Requirementsmentioning

confidence: 99%

“…To let meta-analysts inspect the underlying high-dimensional data for specific points, the Provectories visualization features so-called summary visualizations [13]. Upon hovering over a point, the summary visualization of the corresponding single application state is shown.…”

Section: Visual Encoding and Interactivitymentioning

confidence: 99%

“…We additionally enrich the Provectories visualization with meta-attributes to understand the embeddings in more detail, following the high-dimensional data summary visualization from Eckelt et al [13]. Meta-attributes for this user study are the user ID, the task ID, the accuracy per task on a user level, the task difficulty level, the Boolean attribute of auto-complete used, and the selected rank of the prediction that was used by a user.…”

Section: Summary Visualizationmentioning

confidence: 99%

See 3 more Smart Citations

Provectories: Embedding-based Analysis of Interaction Provenance Data

Walchshofer¹,

Hinterreiter²,

Xu³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Understanding user behavior patterns and visual analysis strategies is a long-standing challenge. Existing approaches rely largely on time-consuming manual processes such as interviews and the analysis of observational data. While it is technically possible to capture a history of user interactions and application states, it remains difficult to extract and describe analysis strategies based on interaction provenance. In this paper, we propose a novel visual approach to meta-analysis of interaction provenance. We capture single and multiple user sessions as graphs of high-dimensional application states. Our meta-analysis is based on two different types of two-dimensional embeddings of these high-dimensional states: layouts based on (i) topology and (ii) attribute similarity. We applied these visualization approaches to synthetic and real user provenance data. From our visualizations, we were able to extract patterns for data types and analytical reasoning strategies.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Requirementsmentioning

confidence: 99%

Section: Visual Encoding and Interactivitymentioning

confidence: 99%

Section: Summary Visualizationmentioning

confidence: 99%

See 2 more Smart Citations

Provectories: Embedding-based Analysis of Interaction Provenance Data

Walchshofer¹,

Hinterreiter²,

Xu³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We derive the following requirements for single and multiple sessions from the existing literature [27], [37] and our prior research experience [13], [23], [53]. To support singlesession investigation, Provectories is designed to: S1 show the entire analysis sequence from beginning to endin temporal order; S2 include the user interaction and/or system state information, such as the changes between two consecutive steps in the analysis sequence;…”

Section: Requirementsmentioning

confidence: 99%

Provectories: Embedding-Based Analysis of Interaction Provenance Data

Walchshofer¹,

Hinterreiter²,

Xu³

et al. 2023

IEEE Trans. Visual. Comput. Graphics

Self Cite

View full text Add to dashboard Cite

Understanding user behavior patterns and visual analysis strategies is a long-standing challenge. Existing approaches rely largely on time-consuming manual processes such as interviews and the analysis of observational data. While it is technically possible to capture a history of user interactions and application states, it remains difficult to extract and describe analysis strategies based on interaction provenance. In this paper, we propose a novel visual approach to the meta-analysis of interaction provenance. We capture single and multiple user sessions as graphs of high-dimensional application states. Our meta-analysis is based on two different types of two-dimensional embeddings of these high-dimensional states: layouts based on (i) topology and (ii) attribute similarity. We applied these visualization approaches to synthetic and real user provenance data captured in two user studies. From our visualizations, we were able to extract patterns for data types and analytical reasoning strategies.

show abstract

DRAVA: Aligning Human Concepts with Machine Learning Latent Dimensions for the Visual Exploration of Small Multiples

Wang¹,

L’Yi²,

Gehlenborg³

2022

Preprint

View full text Add to dashboard Cite

This paper proposes Drava, a novel method that utilizes Disentangled Representation learning as A Visual Analytics approach for concept-driven exploration of small multiples. While latent vectors extracted by machine learning models are widely used to organize and explore data (e.g., layout data items based on their latent vectors in a 2D space using t-SNE), they usually suffer from a lack of interpretability. Disentangled representation learning (DRL) alleviates this issue by learning vectors that encode concepts in separated dimensions and are therefore more interpretable. Even though promising, it can be challenging to apply these disentangled vectors due to the imperfections of algorithms and our limited understanding of how to support DRL-based visual exploration. To address this problem, we propose a three-step workflow where users 1) understand ML-learned concepts, 2) refine and align ML concepts with human concepts, and 3) generate new knowledge about the analyzed data through concept-driven exploration. We support this three-step workflow by providing a set of interactions based on visual piles and enhancing the DRL model with a concept adaptor that fine-tunes concept quality based on human feedback. Interactive visual piles enable users to effectively organize, summarize, and compare groups of items based on human-readable concepts. The concept adaptor assists users in modifying potential imperfect semantic vectors and better facilitates concept-driven exploration. We demonstrate the usability of Drava using four usage scenarios with different datasets. Our experiments show that the semantic latent vectors can capture human-readable concepts and the concept adaptor can further improve these semantic concepts using human feedback.

show abstract

Visual Exploration of Relationships and Structure in Low-Dimensional Embeddings

Cited by 3 publications

References 35 publications

Provectories: Embedding-based Analysis of Interaction Provenance Data

Provectories: Embedding-based Analysis of Interaction Provenance Data

Provectories: Embedding-Based Analysis of Interaction Provenance Data

DRAVA: Aligning Human Concepts with Machine Learning Latent Dimensions for the Visual Exploration of Small Multiples

Contact Info

Product

Resources

About