fbpx
Getting the Most Out of Spatial Data Getting the Most Out of Spatial Data
Key Takeaways: GIS is an art and a science, requiring sophisticated software and a keen focus on aesthetics Organization and visualization... Getting the Most Out of Spatial Data

Key Takeaways:

  • GIS is an art and a science, requiring sophisticated software and a keen focus on aesthetics
  • Organization and visualization of data from online platforms like Twitter can provide insights into geographic and geological phenomena
  • Spatial data can be subject to optimization processes in ways that make businesses more efficient, as in the case of a UK waste company seeking the best truck routes to deliver refuse

For Andy Eschbacher, getting the most out of spatial data is both a science and an art. In his role as a map scientist at CARTO, a leading provider of location intelligence software based in New York City, Eschbacher uses geographical information systems (GIS) to help companies and organizations make decisions.  At ODSC East 2018, Eschbacher talked about some of the use cases for GIS in the public and private sectors, as well as tools, tips, and tricks for working with spatial datasets.

Slide copyright Andy Eschbacher, ODSC East 2018

Spatial data can provide novel approaches to understanding geographic, demographic, and even geological events and trends. In one instance, Eschbacher organized Twitter data around mentions of the word “earthquake” to create impact maps detailing the extent of earthquakes in the American Midwest. Analyzing clusters and outliers of activity helped paint accurate pictures of major seismic events. This helps communities better understand the effects of earthquakes in their areas.

Applying GIS Principles to Business Problems

CARTO applies GIS principles to business problems for companies that rely on spatial data or networks. To illustrate such real-world GIS applications, in his talk, Eschbacher described an engagement with a UK waste management company that sought more efficient truck routes to collect and deliver refuse.

By mapping UK road networks in the company’s area of operation, and applying methods for linear optimization, Eschbacher and his team were able to devise more efficient routes for the company’s drivers. Eschbacher has also used similar processes in the retail world, helping merchandisers decide where to open new stores based on spatial data for demographics and past store performance.

Slide copyright Andy Eschbacher, ODSC East 2018

Eschbacher concluded his talk by highlighting several of what he considers to be the best tools for GIS and working with spatial data. For Python users, the open source project, GeoPandas, enables geospatial operations that would otherwise require spatial database software like PostGIS, including mapping, geocoding, and geometric manipulations. Similarly, Folium provides Python users with an easy-to-use interface for creating Leaflet maps.

Finally, Eschbacher is one of the creators of CARTOframes, a new Python package for integrating CARTO into data science workflows. It allows you to build CARTO maps, read/write/query datasets, perform spatial operations, and augment data from Data Observatory. It also easily ties into the Python ecosystem because it speaks the language of the de facto data science standards pandas, Jupyter, matplotlib, and SQL/PostGIS.

 


Interested in hearing more in-depth discussions from leaders in the data science and artificial intelligence space? Check out ODSC’s upcoming conferences in San Francisco and London and register to be in the know.

Stay up to date on all things data science here.

Alex Amari

I’m a graduate student at Oxford University pursuing an MSc in Social Data Science with the ultimate goal of working in tech entrepreneurship.

1