Data Stories

169  |  Data Conversations with Vidya Setlur


Listen Later

We have Vidya Setlur on the show to talk about the role language, and natural language processing (NLP) play in data visualization and analytics.

Vidya is the director of research at Tableau and has a background in natural language processing and visualization. She is one of the main drivers behind Eviza, a research-based prototype and the corresponding product Ask Data, developed within Tableau to interact with data visualizations through natural language.

She is also the co-author, with Bridget Cogley, of Functional Aesthetics for Data Visualization, a new book on data visualization with a lot of information about semantics and language in data visualization.

In the episode, we talk about the challenges of going from a research prototype to an actual product, research vs. engineering, speech and natural language interfaces, the many ways language plays a role in visualization, the advent of language models, and much more.

Enjoy the show!

 

Links

  • Winners of Information is Beautiful Awards 
  • Book: Functional Aesthetics for Data Visualization
  • Paper: Automatic Generation of Semantic Icon Encodings for Visualizations
  • Paper: A Linguistic Approach to Categorical Color Assignment for Data Visualization
  • Eviza: A Natural Language Interface for Visual Analysis
  • Paper: Snowy: Recommending Utterances for Conversational Visual Analysis
  • Remember: our podcast is listener-supported. Please consider donating using Patreon or Paypal. Thanks! ?

    Related episodes
    • Text Visualization: Past, Present and Future with Chris Collins
    ...more
    View all episodesView all episodes
    Download on the App Store

    Data StoriesBy Enrico Bertini and Moritz Stefaner

    • 4.6
    • 4.6
    • 4.6
    • 4.6
    • 4.6

    4.6

    388 ratings


    More shows like Data Stories

    View all
    99% Invisible by Roman Mars

    99% Invisible

    26,182 Listeners

    Up First from NPR by NPR

    Up First from NPR

    56,107 Listeners

    DataFramed by DataCamp

    DataFramed

    267 Listeners