runck014 / day-of-data

01.12.18 - Notebook provides basic introduction to spatial data science in python for UMN Day of Data 2018

Getting Started with Spatial Data Science in Python

01.12.18 - Notebook provides basic introduction to spatial data science in python for UMN Day of Data 2018 website 10.24.18 - University of Minnesota Geocomputing Group

Bryan C. Runck // [e-mail] // Department of Geography, Environment and Society

Overview

How can we use python to do spatial data science? This jam session will provide a hands-on overview of basic mapping in Python with GeoPandas and how to perform basic spatial analysis using PySAL. No programming experience is required.

Supporting Information

Google Slides

Objectives

1. Make simple maps with GeoPandas and AirBnB data
   • Data I/O
   • Make chloropleth maps
   • Make scatterplots
   • Rate mapping
   • Recognize the importance of projections
2. Perform an exploratory visual analysis of the data to identify potential places you would want to hone an AirBnB stay
3. Use PySAL to compute global spatial autocorrelation
   • Constructing spatial weights
   • Moran's I (Global)
   • Visually check result
4. Use Moran's I to determine which AirBnB variables have high levels of spatial autocorrelation

Data

Data files are from GeoDa Data and Lab files (details).