Spatial Machine Learning in ArcGIS Pro – ESRI Workshop

This workshop is designed to introduce spatial machine learning and data science Geoprocessing tools in ArcGIS Pro. User familiarity with ArcGIS Pro is required.  This workshop is offered both in person in the Tisch Data Lab and online via zoom.  If you are planning to join remotely, you need to have access to ArcGIS Pro. If you are a current Tufts affiliate, you can install ArcGIS Pro on your personal windows computer (instructions here: https://tufts.box.com/v/ArcGIS-Pro2-8) or you can access ArcGIS Pro via the TTS Remote Lab (adobe). Instructions for the remote lab can be found here: https://sites.tufts.edu/datalab/covid-19-data-lab-tools-preparing-to-work-remotely/

If you have any questions about this event or how to access the software, please feel free to reach out to DataLab-support@elist.tufts.edu

Problems related to physical and human geography involve working with data from diverse sources from a variety of spatial scales. Modeling patterns, relationships, and groups in rich multidimensional data is common in both physical and human geography workflows with the soaring use of machine learning methods in spatial problems.

This workshop will introduce spatial machine learning methods in ArcGIS Pro and showcase their uses in a variety of workflows. Learners will have the ability to use the tools in a hands-on, applied setting. Regression, clustering, and classification methods in ArcGIS Pro will be presented.  Metrics for model performance assessment alongside workflows for choosing optimal machine learning models will be discussed for various geographic problems.

Data Engineering & Visualization (1 Hour)

1. Encoding
2. Transformation
3. Visualizing Relationships
4. Visualizing Time
5. Visualizing Space-Time Cubes

1. Finding Patterns (0.5 Hour)
   1. Spatial Patterns
   2. Space-Time Patterns
   3. Finding Clusters in Space and Time
2. Modeling Relationships (1.5 Hour)
   1. Linear Regression
   2. GWR
   3. Bivariate Association
   4. Forest-Based Classification and Regression