Most businesses have a geolocation component, think of tracking a runners club to show progress, optimized delivery routes or finding the perfect location for your next store. When implemented effectively, geospatial data can drive growth and open up new strategic opportunities.

The talk will walk you through the full lifecycle of geospatial data developments with open-source tools—from preprocessing GPS signals with Kalman filters to SQL processing using PostGIS in PostgreSQL databases. You will learn how to analyze and transform data in Python using GeoPandas, the geospatial extension of Pandas. When the data is ready, you’ll see how to visualize geospatial data using Python libraries such as Folium and the open-source tool QGIS. You will also learn how to use unsupervised machine learning models to segment geospatial time-series and classify, for example, the types of movement.

In the demo, we will show step-by-step a geospatial pipeline that transforms a the GPS trace of a runner & cyclist into processed insights. We’ll learn how to visualize and manipulate geospatial data in QGIS.