Location Intelligence and Spatial Data Science: from geographic coordinates to business insights
summary
Demonstrates using Python to turn raw coordinates into a modelable dataset: starts from 50,000 places across Italy, geocodes and fixes missing postal codes, joins Milan postcode polygons, and enriches each point with population density, income, nearest-neighbor and centroid distances and counts of POIs within 300 m radius. Uses geopandas + folium + shapely + pyproj to build interactive maps and compute geometry, then fits regression baselines (linear, regularized, simple neural net, polynomial) to a synthetic revenue target and reports improved RMSE with nonlinear features.
AI-written summary · about
chapters (14)
- 0:00Introduction and speaker intro
- 1:47What is location intelligence
- 6:26Retail site planning challenges
- 8:11Visualizing geospatial data basics
- 10:43Interactive mapping with Folium
- 12:08Geometric calculations with GeoPandas
- 14:58Working with polygon data
- 16:45Enriching data with demographics
- 18:09Geocoding and data cleaning
- 20:16Downloading points of interest
- 22:05Building predictive features
- 23:10Predictive modeling for revenue
- 25:40Summary and key takeaways
- 26:21Q&A session
related talks
30mUnlocking Geospatial Data Science: A Strategic Overview
31mUnpacking the geospatial engineering toolbox – an overview of data science techniques for spatial da
28mA pythonista’s pipeline for large scale geospatial analytics
22mExpert Panel | Location Data in Retail & Real Estate: Delivering an ROI Through Analytics
10mCivic Hacking & Urban Informatics: Mapping Data with GeoPandas
26mSridhar Telidevara: Professor, Great Lakes Institute of Management at Cypher 2017 - Part 1/2
20mCloud Native Patterns for Geospatial Analytics and AI
40mMachine Learning over Small Data: generating new variables with predictive power