From simple methods for summarizing and describing spatial patterns to advanced machine learning clustering techniques, this workshop will introduce you to the power of spatial statistics and equip you with the knowledge needed to get started exploring your data in new and useful ways. Concepts covered include describing your data’s shape and spatial distribution; comparing datasets in meaningful defensible ways; identifying spatial clusters; and mining for multivariate patterns. We will discuss how the tools work and provide examples to demonstrate the range of questions that can be answered.
Spatial Data Mining: Cluster Analysis and Space-time Pattern Mining
Whenever we look at a map, we naturally organize, group, differentiate, and cluster what we see to help us make better sense of it. This workshop will explore the powerful spatial statistics techniques designed to do just that in space and time: We’ll start with hot-spot analysis and cluster and outlier analysis. Through discussions and demonstrations, we will learn how these techniques work and how they can be used to identify significant patterns in our data. We will explore the types of questions each tool can answer, best practices for running the tools, and strategies for interpreting and sharing results. We will then present advanced techniques for analyzing your data in the context of both space and time, covering space-time pattern mining techniques including aggregating temporal data into a space-time cube, emerging hot spot analysis, local outlier analysis, best practices for visualizing your space-time cube, and strategies for interpreting and sharing your results.
Beyond Where: Modeling Spatial Relationships and Making Predictions
Once we’ve identified where patterns are present, the next logical question is “why?” This workshop will cover techniques for examining, modeling, and exploring our spatial data to uncover relationships and predict spatial outcomes. Application and use of generalized linear regression (GLR), geographically weighted regression (GWR), and Forest-based Classification and Regression (FBCR) will be demonstrated. You will learn how to build a model and how to effectively interpret the results and diagnostics.
You can register for any of our 4 pre-workshops at this link:
http://merltech.org/merl-tech-dc-pre-workshops-2019/