An analysis of racial segregation using topological data analysis

UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
Jakini Auset Kauba (Creator)
The University of North Carolina at Greensboro (UNCG )
Web Site:
Thomas Weighill

Abstract: In recent years, Topological Data Analysis (TDA) has been used to analyze complex data and provide insights that other research techniques cannot. TDA is a newer form of data analysis which analyzes trends of data from a topological perspective by way of the main visualization tool of persistence diagrams. TDA has been used to measure breast cancer transcriptional DNA, voting patterns in precincts, gerrymandering, and even texture representation. In this paper, we apply TDA to geospatial data from the census to more accurately describe racial segregation among the Black and Hispanic demographics across one hundred cities in America. Our goal was to complete city to city comparisons in 2010 and 2020 as well as compare city similarities over the course of ten years for each race and note the respective trends. Generally, we were able to conclude that larger cities were more likely to exhibit racial shifts in demographic data, while smaller cities did not. However, it was also noted that both demographics contained large city representatives which behaved as smaller cities indicating little shift in racial demographics. In summary, this project represents a first step in uncovering trends in demographic data using TDA. We hope to continue exploring this data set in an effort to expand our understanding of racial segregation in America.

Additional Information

Language: English
Date: 2022
Geospatial Data, Homology, Persistence, TDA, Topology
Segregation $z United States $x Mathematical models
Demographic surveys $z United States
Algebra, Homological

Email this document to