For full text search please use the '?' prefix. e.g. ? Onboarding

Wednesday - Day 3

Wednesday - 2 short, 2 full

  1. VIS Full Papers: Transforming Tabular Data and Grammars - Alex Lex
  2. VIS Full Papers: Understanding and Modeling How People Respond to Visualizations - Carolina Nobre
  3. VIS Short Papers: Visualization Systems and Graph Visualization - Kat Isaacs
  4. VIS Short Papers: Visual Analytics, Decision Support, and Machine Learning - Matthew Berger (handled qa)

Highlights

  • Tabular data and gramars
    • visualizations + recommendations can help w/ both graph config, data transforms, and query formula definitions (e.g verb previews)
      • HiTailor: "recipes" can fit into categories (change, distro)
    • Animated Vega Lite for presentation quality / DSLs for API design on core-viz
  • Responding / Evaluation
    • Examine frameworks for how we decide whether a viz is "working" (both for aesthetics and function)
    • Think about epilepsy risk with animations/color choice
    • Progressive loading coms (multistep queries, changing the bin size by timeframe size)
  • Graph Systems
    • nl4dv python package: build graphs in stages to learn VL. See tweet
    • Vega Fusion: for expensive vega lite specs
    • resampler: a visualization-aware sampling algorithm - could use in experimental form with tiny buckets for big timeseseries that need more detail over long time periods.
  • Decision Support / VA / Machine Learning
    • TimberTrek, Visual Auditor, Fuzzy Spreadsheets: all ways to fit into existing workflows (notebooks, sheets)
      • TimberTrek in particular: make use of sticky panes + an attribute-based tree slicer?
      • May need to make use of funnels, service level meatdata, etc
    • FairFuse: Do we have different ranking algos to compare?
    • ML explainability: explaining rule classifiers + compare behavior in subgroups
    • "Active Search": Recommend follow-up items in "real time"?
  • Hallway Track

Raw Notes

Tabular Data + Grammars

VIS Full Papers: Understanding and Modeling How People Respond to Visualizations

VIS Short Papers: Visualization Systems and Graph Visualization

VIS Short Papers: Visual Analytics, Decision Support, and Machine Learning

Thank you for these additional reflections and references 🙂 . I assume Quan's work on diminishing returns will show up on this page if it isn't already here: https://krisnguyen135.github.io/publications/

I didn't ask this last question, but it made me think of the general risk of humans being less capable if they become over-dependent on automated systems (simplified example is when people forget to spell when using a plain text editor, because they're so used to typing with spellcheck on).

(digest version + link to the short paper, 6 page) https://blog.acolyer.org/2020/01/08/ironies-of-automation/

  • Parametric Dimension Reduction by Preserving Local Structure

  • UMATO ( Uniform Manifold Approximation with Two-phase Optimization )

    • Hyeon Jeon
    • #demo with python: https://github.com/hyungkwonko/umato
    • https://virtual.ieeevis.org/year/2022/paper_v-short-1047.html
    • Compete with T-sne, UMAP at preserving local structure and global structure at same time
    • Is pipeline on top of UMAP, so is slightly slower
    • Qualitatively... global metrics succeed, though t-sne/umap still won at pure local
    • Seems like it works with a pass that picks "hub spots" before going to final
      • TBD what problem domain this would be useful for outside of visualizing geometric hypercube tests, although he did test it with 3 real-world algos.

Bonus side track: Uncertainty snippet

  • Learned from Fuzzy Column Spreadsheets in other track on Uncertainty viz
    • #to-read / #demo https://jku-vds-lab.at/fuzzy-spreadsheet/
    • Fuzzy Spreadsheet: Understanding and Exploring Uncertainties in Tabular Calculations - Viashali Dhanoa
    • You can write custom Spreadsheet plugins using Typescript!
    • See slides for more pictures... lesson in fittin ginto existing workflows
    • Fits into what-if analysis, and puts context in sidepanel in a friendly slideshow environment
    • Spreadsheet-based tools provide a simple yet effective way of calculating values, which makes them the number-one choice for building and formalizing simple models for budget planning and many other applications. A cell in a spreadsheet holds one specific value and gives a discrete, over precise view of the underlying model. Therefore, spreadsheets are of limited use when investigating the inherent uncertainties of such models and answering what-if questions. Existing extensions typically require a complex modeling process that cannot easily be embedded in a tabular layout. In Fuzzy Spreadsheet, a cell can hold and display a distribution of values. This integrated uncertainty-handling immediately conveys sensitivity and robustness information. The fuzzification of the cells enables calculations not only with precise values but also with distributions, and probabilities. We conservatively added and carefully crafted visuals to maintain the look and feel of a traditional spreadsheet while facilitating what-if analyses. Given a user-specified reference cell, Fuzzy Spreadsheet automatically extracts and visualizes contextually relevant information, such as impact, uncertainty, and degree of neighborhood, for the selected and related cells. To evaluate its usability and the perceived mental effort required, we conducted a user study. The results show that our approach outperforms traditional spreadsheets in terms of answer correctness, response time, and perceived mental effort in almost all tasks tested.

Hallway Track

Hack Ideas

Scratchpad

Backlinks