Semantic interaction for visual analytics : : inferring analytical reasoning for model steering /
By: Endert, Alex [author.].
Material type:![materialTypeLabel](/opac-tmpl/lib/famfamfam/BK.png)
Item type | Current location | Call number | Status | Date due | Barcode | Item holds |
---|---|---|---|---|---|---|
![]() |
PK Kelkar Library, IIT Kanpur | Available | EBKE727 |
Mode of access: World Wide Web.
System requirements: Adobe Acrobat Reader.
Part of: Synthesis digital library of engineering and computer science.
Includes bibliographical references (pages 79-87).
1. Introduction -- 1.1 The role of visual analytics in a data-driven era -- 1.2 Semantic interaction -- 1.3 Outline --
2. Fundamentals -- 2.1 Sensemaking and analytical reasoning -- 2.2 The use of analytic models in visual analytics -- 2.3 User interaction -- 2.3.1 Modeling user interest from user interaction -- 2.4 Model steering -- 2.5 Mixed-initiative system principles --
3. Spatializations for sensemaking using visual analytics -- 3.1 The value of manually organizing data in spatializations -- 3.2 Computationally generating spatializations --
4. Semantic interaction -- 4.1 Designing for semantic interaction -- 4.1.1 Capturing the semantic interaction -- 4.1.2 Interpreting the associated analytical reasoning -- 4.1.3 Updating the underlying model -- 4.2 Exploring the semantic interaction design space -- 4.2.1 The interaction-feedback loop -- 4.2.2 Approximating and modeling user interest -- 4.2.3 Choice of mathematical model -- 4.2.4 Relative and absolute spatial adjustments --
5. Applications that integrate semantic interaction -- 5.1 ForceSPIRE: semantic interaction for spatializations of text corpora -- 5.1.1 Constructing the spatial metaphor -- 5.1.2 Semantic interaction in ForceSPIRE -- 5.1.3 Model updates -- 5.2 Semantic interaction for dimension reduction models -- 5.2.1 Probabilistic principal component analysis (PPCA) -- 5.2.2 Multi-dimensional scaling (MDS) -- 5.2.3 Generative topographic mapping (GTM) -- 5.2.4 InterAxis: steering scatterplot axes --
6. Evaluating semantic interaction -- 6.1 Methodology considerations -- 6.1.1 Evaluation of analytic process -- 6.1.2 Evaluation of analytic product -- 6.2 Example: evaluating semantic interaction in ForceSPIRE -- 6.2.1 Method -- 6.2.2 Results --
7. Discussion and open challenges -- 7.1 User interaction for visual analytics -- 7.2 Effects of semantic interaction on analytic process -- 7.3 Differentiating bias from intuition -- 7.4 Additional visual representations and interactions --
8. Conclusion -- References -- Author's biography.
Abstract freely available; full-text restricted to subscribers or individual document purchasers.
Compendex
INSPEC
Google scholar
Google book search
User interaction in visual analytic systems is critical to enabling visual data exploration. Interaction transforms people from mere viewers to active participants in the process of analyzing and understanding data. This discourse between people and data enables people to understand aspects of their data, such as structure, patterns, trends, outliers, and other properties that ultimately result in insight. Through interacting with visualizations, users engage in sensemaking, a process of developing and understanding relationships within datasets through foraging and synthesis. This book discusses a user interaction methodology for visual analytic applications that more closely couples the visual reasoning processes of people with the computation. The methodology, called semantic interaction, affords user interaction on visual data representations that are native to the domain of the data. These interactions are the basis for refining and updating mathematical models that approximate the tasks, intents, and domain expertise of the user. In turn, this process allows model steering without requiring expertise in the models themselves.instead leveraging the domain expertise of the user. Semantic interaction performs incremental model learning to enable synergy between the user's insights and the mathematical model. The contributions of this work are organized by providing a description of the principles of semantic interaction, providing design guidelines for the integration of semantic interaction into visual analytics, examples of existing technologies that leverage semantic interaction, and a discussion of how to evaluate these techniques. Semantic interaction has the potential to increase the effectiveness of visual analytic technologies, and opens possibilities for a fundamentally new design space for user interaction in visual analytic systems.
Also available in print.
Title from PDF title page (viewed on September 18, 2016).
There are no comments for this item.