A Toolbox for Scalable Spatial Data Science in the Cloud
summary
CARTO presents a cloud‑native SQL library that installs inside warehouses (BigQuery Redshift Snowflake Databricks) to run spatial analytics at scale without ETL, claiming terabyte‑scale speed. The Analytics Toolbox exposes core open‑source spatial index and geometry transforms plus advanced modules: statistics (local Moran’s I, geographically weighted regression), routing, and a retail pipeline that prepares data, trains an XGBoost model, and predicts revenue; outputs include model, feature importance tables and inference maps. Many methods require an H3 index for scalability.
AI-written summary · about
chapters (7)
- 0:00Introduction and context
- 2:37Analytics Toolbox overview
- 4:44Local Moran's I demo
- 7:54Geographically weighted regression
- 10:02Revenue prediction functions
- 13:37Key takeaways and closing
- 14:21Q&A session
related talks
30mUnlocking Geospatial Data Science: A Strategic Overview
24mManaging Heterogeneous Spatial Data with Analysis Ready Cloud Optimized formats
07mSmarter Cities through Geotab
43mData warehousing reinvented for today's needs
46mGeoSpatially enabling your Big Data Apache clusters with LocationTech
33mBuilding Data Science Tools for Sustainable Transformation
32mIntroduction to geospatial analysis for uninitiated SQL Data Engineer
25mA quack of all trades: DuckDB, a versatile analytical database to keep in your toolkit