Machine Learning for GIS and Remote Sensing

Catalog description

This course introduces machine learning and applied computer vision techniques for using GIS and remotely sensed data to solve Earth science problems related to climate resilience and sustainability. Topics include supervised learning, neural networks, convolutional neural networks, dimension reduction, and unsupervised learning (clustering). The course interweaves theory and practice where classes provide theoretical depth into contemporary artificial intelligence approaches. The hands-on labs, assignments, and projects give students experience managing, wrangling, and utilizing geospatial data in machine learning tasks. Students will explore integrations of machine learning methods in Earth, environmental, and geospatial sciences via in-class artificial intelligence applications of contemporary Earth science topics such as food-water nexus, natural hazard detection, and computational sustainability. Students taking GIS 5092 will work on self-guided final projects that apply machine learning methods to a problem in Earth, environmental, or geospatial sciences.