How to Get Instant Geospatial Insights Using Natural Language
At the Spatial Data Science Conference in New York, Danielle Gewurz from Deloitte demonstrates how to instantly transform a natural language question into a rich, interactive map and a comprehensive analysis of pedestrian safety data. This is the future of democratized spatial analytics, turning hours of work into seconds of insight. Inspired by this clip? Watch the full talk and all sessions from SDSC NYC 2025 on demand: https://spatial-data-science-conference.com/2025/newyork#Watch-on-demand #SDSC25 #geospatial #spatialdatascience #deloitte Transcript: 00:05 I'll put a question in on the right: "Show me the pedestrian crashes by month and time of day in 2023 in a district." I'm showing California. On the back end, a series of agents parse this question to determine the intent and what data sources would best address the user's query. 00:31 Then, code is written to pull the data, and the generative AI agents respond, choosing how to visualize the data that answers the question. You'll see a full description of what's happening and an analysis of that data. At the bottom of the chat response on the right, the agent has chosen a symbol to indicate pedestrian crashes. 00:57 If you mouse over any of those, you can see the details. Additionally, on the right, there are responsive widgets generated by agents to show the crash severity distribution and the most frequent cause of those crashes from underlying police reports. 01:16 That's a situational awareness question: tell me about the crashes that have happened. Now, we're putting up a more forward-looking question: "What countermeasures and changes do I need to make to my safety structure to improve pedestrian crashes and increase pedestrian safety in Modesto?" Here, the agent does a live analysis of, for example, the time of day to produce responsive countermeasures. 01:44 One of the suggestions that comes up is lighting. You'll also see the highest-risk locations. These are the places where you should focus your pedestrian countermeasures.