07-10, 11:25–11:55 (US/Pacific), Room 317
The representation, synthesis, modeling, and visualization of neighborhoods is a fundamental pursuit across a range of social sciences. In recent decades, recogni
Quantitative research focusing on cities, neighborhoods, and regions is in high demand.
On the practical side, city planners, NGOs, and commercial businesses all have increasing
access to georeferenced datasets and spatial analyses can turn these data into more efficient, equitable, and profitable, decision-making. On the scientific/academic side, many of the era's pressing problems are spatial and or urban in form, thus requiring bespoke statistical methods and simulation techniques to produce accurate inferences. Thus, while data science teams in industries across the planet race to conquer 'geospatial data science' few practitioners have expertise in the appropriate data sources, modeling frameworks, or data management techniques necessary for wrangling and synthesizing urban data.
To address these challenges we introduce geosnap, the geospatial neighborhood analysis package, which is a unique Python package sitting at the intersection of spatial science and practical application. In doing so, it provides a bridge between formal spatial analysis/spatial econometrics and the applied fields of neighborhood change, access to opportunity, and the social determinants of health. The package provides extremely fast and efficient access to hundreds of socioeconomic, infrastructure, and environmental quality indicators that allow researchers to move from zero data to an informative model in a few short lines of code. It is organized into layers for data acquisition, analysis, and visualization and it includes tools for harmonizing disparate datasets into consistent geographic boundaries, creating "geodemographic typologies" that summarize and predict multidimensional segregation over time, and network analysis tools that rapidly generate (locally computed) service areas and travel isochrones. We expect the tool will be immediately useful for any analyst studying urban areas.
Eli is a Senior Research Scientist and the Associate Director of the Center for Open Geographical Science at San Diego State University. He is a spatial data scientist trained in stratification sociology, urban economics, and quantitative geography, whose research focuses on social inequality and spatial structure in neighborhoods, cities, and regions. Eli is a core developer for PySAL, QuantEcon, and OTURNS