Description
Proposal
Topic of the session
Demo of geoda spatial analysis modules in kepler, deck and AI (openassistant)
Type of the session
- Collaborate
- Workshop
- Talk
Estimated duration of the session
20 min (open for discussion)
Date and Time of the session
TBD
Level
- Beginner
- Intermediate
- Advanced
Pre-requisite knowledge
No pre-requisite knowledge needed.
Describe the session
For a long time, tools for analyzing geospatial data have been specialized, closed applications that were isolated from modern tech stacks. In this demo, we introduce new efforts to make geospatial analytics, such as local cluster statistics, more broadly accessible to the open visualization community by integrating with widely used open web-based visualization platforms (kepler.gl and deck.gl). Further, by combining traditional spatial analysis software with large language models (LLMs), analysts are now able to obtain spatial analysis results in response to questions they ask in natural language rather than based on code or menus. Data remains private as the raw data stays within the existing software system and is not uploaded or shared with LLMs. We demo this AI Assistant with an in-browser exploration of geospatial data that combines table queries with spatial analytics and descriptive statistics to answer spatial questions. More broadly, the AI Assistant is designed as a software library that can be used as a plugin to turn existing software systems into AI-powered software.
Session facilitator(s), Github handle(s) and timezone(s)
Xun Li (lixun910), MST or in person -- Julia Koschinsky (jkoschinsky), in person
Meeting notes and Virtual Meeting Link
Follow-up / Set-up sessions (if any)
Additional context (optional)
Session in person; no specific time needed