Modeling and Leveraging Analytic Focus During Exploratory Visual Analysis

Zhou, Zhilan; Wen, Ximing; Wang, Yue; Gotz, David

doi:10.1145/3411764.3445674

Cited by 11 publications

(3 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Researchers also visualized interactions to allow users to recover their own / others' reasoning processes [18,35,42]. Moreover, interactions were analyzed to help guide users along the analytic process [16], predict personality traits [8], support the recommendation of relevant resources for further exploration [20,62], etc. As a primary goal of visualization is to support insight, we attempt to explicitly relate interactions to the resulting insight to help better understand insight generation.…”

Section: Interaction Analysis With Entitiesmentioning

confidence: 99%

Characterizing Visualization Insights through Entity-Based Interaction: An Exploratory Study

He¹,

Vuong²

2022

Preprint

View full text Add to dashboard Cite

One of the primary purposes of visualization is to assist users in discovering insights. While there has been much research in information visualization aiming at complex data transformation and novel presentation techniques, relatively little has been done to understand how users derive insights through interactive visualization of data. This paper presents a crowdsourced study with 158 participants investigating the relation between entity-based interaction (an action + its target entity) and the resulting insight. To this end, we generalized the interaction with an existing CO 2 Explorer as entity-based interaction and enabled users to input notes and refer to relevant entities to assist their narratives. We logged interactions of users freely exploring the visualization and characterized their externalized insights about the data. Using entity-based interactions and references to infer insight characteristics (category, overview versus detail, and prior knowledge), we found evidence that compared with interactions, entity references improved insight characterization from slight/fair to fair/moderate agreements. To interpret prediction outcomes, feature importance and correlation analysis indicated that, e.g., detailed insights tended to have more mouse-overs in the chart area and cite the vertical reference lines in the line chart as evidence. We discuss study limitations and implications on knowledge-assisted visualization, e.g., insight recommendations based on user exploration.

show abstract

Section: Interaction Analysis With Entitiesmentioning

confidence: 99%

Characterizing Visualization Insights through Entity-Based Interaction: An Exploratory Study

He¹,

Vuong²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…For example, prior tools [2,24,33,34,41] have considered the relevance of previously explored visualizations based on similarities in their data content or data patterns. Zhou et al [43] monitored the development of the exploration focus (based on which attributes were being explored) and developed a model to infer the next focus. Qian et al [24] recommend visualizations based on the data patterns that could be of interest to the users.…”

Section: Processes Of Exploratory Visual Analysismentioning

confidence: 99%

Exploration and Practice of Cultivating Mode of Top Talents in Chemistry of Nankai University

WANG¹

2019

University Chemistry

View full text Add to dashboard Cite

Domain generalization (DG) aims to train a model from limited source domains, allowing it to generalize to unknown target domains. Typically, DG models only employ large-scale pre-trained models during the initialization of fine-tuning. However, large-scale pre-trained models already possess the ability to resist domain shift. If we reference pre-trained models continuously during fine-tuning to maintain this ability, it could further enhance the generalization ability of the DG model. For this purpose, we introduce a new method called Fine-Tune with Large-scale pre-trained Priors (FT-LP), which incorporates the pre-trained model as a prior into the DG fine-tuning process, ensuring that the model refers to its pre-trained model at each optimization step. FT-LP comprises a theoretical framework and a simple implementation strategy. In theory, we verify the rationality of FT-LP by introducing a generalization error bound with the pre-trained priors for DG. In implementation, we utilize an encoder to simulate the model distribution, enabling the use of FT-LP when only pre-trained weights are available. In summary, we offer a new fine-tuning method for DG algorithms to utilize pre-trained models throughout the fine-tuning process. Through experiments on various datasets and DG models, our proposed method exhibits significant improvements, indicating its effectiveness.Preprint. Under review.

show abstract

“…To better understand the design space for the content recommendation within visual analytics platforms, and motivated by our own work on this topic [83], we conducted a formal survey of the visual analytics literature to identify prior efforts at content recommendation. We searched the literature for adaptive visualization systems that algorithmically identify and surface new content with the aim of helping users discover additional relevant information given their analytical context.…”

Section: Introductionmentioning

confidence: 99%

A Design Space for Surfacing Content Recommendations in Visual Analytic Platforms

Zhou¹,

Wang²,

Guo³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Recommendation algorithms have been leveraged in various ways within visualization systems to assist users as they perform of a range of information tasks. One common focus for these techniques has been the recommendation of content, rather than visual form, as a means to assist users in the identification of information that is relevant to their task context. A wide variety of techniques have been proposed to address this general problem, with a range of design choices in how these solutions surface relevant information to users. This paper reviews the state-of-the-art in how visualization systems surface recommended content to users during users' visual analysis; introduces a four-dimensional design space for visual content recommendation based on a characterization of prior work; and discusses key observations regarding common patterns and future research opportunities.

show abstract

Modeling and Leveraging Analytic Focus During Exploratory Visual Analysis

Cited by 11 publications

References 43 publications

Characterizing Visualization Insights through Entity-Based Interaction: An Exploratory Study

Characterizing Visualization Insights through Entity-Based Interaction: An Exploratory Study

Exploration and Practice of Cultivating Mode of Top Talents in Chemistry of Nankai University

A Design Space for Surfacing Content Recommendations in Visual Analytic Platforms

Contact Info

Product

Resources

About