Abstract: Spatial data science uses many of the same techniques and algorithm as traditional data science, but the spatial component can add a large amount of additional information by combining with other sources at the same location (e.g., census, geolocated tweets), using real-time routing services, or using the spatial structure of the distribution of the data.
In this talk, I will highlight work we have done in linear optimization, genetic algorithms, and constraint-based clustering that take special advantage of the spatial part of the data. For example, using the Python package CVXOPT, we solved a linear optimization problem that optimally distributes an asset from a source to a drain according to the road network and constraints that the drains cannot be over capacity, occasionally have fixed assignments, and all the asset has to be moved.
Bio: Andy Eschbacher is a spatial data scientist at CARTO, where he solves problems using convex optimization, machine learning, and statistics that often take advantage of the properties of location. He also builds tools such as cartoframes, a data science package for interacting with CARTO, analyses for CARTO's infrastructure, and custom solutions for clients.