Authors: Elizabeth Delmelle*, University of North Carolina at Charlotte
Topics: Geographic Information Science and Systems, Temporal GIS, Urban Geography
Keywords: Neighborhood Dynamics, GIScience
Session Type: Paper
Presentation File: No File Uploaded
In 2003, David Mark defined GIScience as “the development and use of theories, methods, technology, and data for understanding geographic processes, relationships, and patterns”. Seventeen years later, our ability to understand geographic relationships and patterns has flourished, while the use of GIS to understand many geographic processes has lagged. Understanding processes of neighborhood changes is one such area that has confounded traditional GIS procedures as it involves analyzing multiple attribute dimensions that define a neighborhood, over time, for spatially situated units whose boundaries are not static. Recent advances in blending various machine learning algorithms for deciphering processes of change are beginning to make inroads onto this subject matter. In this presentation, I will discuss some of these advancements as well as opportunities for improvement. In addition, I will discuss other areas where GIScience developments can help further our understanding of these processes including the use of text as data. Using a case study on property advertisements, I will analyze the language of neighborhood change and stability in a GIS environment. This analysis will involve an exploratory analysis of text for neighborhoods classified according to mortgage lending trends by race and income and the predictive power of various words in explaining neighborhood transitions. Finally, I will wrap up with a research agenda for furthering our understanding of neighborhood processes.