Conversation: Intro to BayesDB: Open-source AI that can help journalists search messy multivariate data

  • Event: 2018 CAR Conference
  • Speaker: Sara Rendtorff-Smith of MIT
  • Date/Time: Thursday, Mar. 8 at 9:00am
  • Location: Halsted
  • Audio file: No audio file available.

Over the past ten years, researchers at MIT have been developing an open-source probabilistic AI software called BayesDB. BayesDB has the potential to allow users to answer a broad class of data science questions interactively, in seconds or minutes, without requiring statistical programming. Our team is currently working to understand how BayesDB can help journalists, with or without programming skills, identify and build their story. Specifically, we are in the process of developing tools for spreadsheet search, predictive modeling and time series forecasting.

Come and learn more about our open-source probabilistic AI software and help provide feedback on how we can adapt BayesDB tools to better suit the needs and workflow of investigative journalists.

Speaker Bios

  • Sara Rendtorff-Smith is the applied research lead for MIT's Public Interest AI Project. Her work supports public officials, nonprofits & journalists in leveraging probabilistic AI to advance public interest & social equity. Prior to joining MIT she had an extensive career at the United Nations dedicated to advancing inclusive approaches to development, peace & security. She is interested in criminal justice, equity & the integrity of the political process @globalinquirer

Related Tipsheets

No tipsheets have yet been uploaded for this event.