Exploring Gentrification and Displacement Through User-Generated Geographic Information

Gentrification and displacement are pressing issues for many cities today, as urban populations continue to grow and neighborhoods change rapidly in response. Our Capstone study utilizes two such novel data sources, Twitter and Foursquare, to explore gentrification and displacement risk for neighborhoods within the 31-county NY metro region. Using methodology established by the UC Berkeley Urban Displacement Project, we utilize both administrative census data and user-generated social media data, to model the gentrification phenomena and in changing neighborhoods.

Visualization

Project Design

Data Exploration

An initial round of exploratory analysis was done on the datasets to gain an understanding of the profile of the data, distribution of the data across our target geography, and identify attributes of the data that can be feature engineered to include into our models. Foursquare data was initially clustered by income level. Similarly, Twitter data was clustered using several different metric combinations.

Data Modeling

The initial round of models focused on attempting to model the spectrum of gentrification, using multiclass classification to predict all eight gentrification typologies in the original UDP index. Based on these results, we re-evaluated the modeling approach of the project through additional literature review and feedback from our project sponsor/mentor. Two major changes were done to address the poor performance of the models. First, we shifted from a multiclass classification approach to a binary classification approach Second, we started to delve into different metrics of gentrification, namely the distinction between people gentrification, place gentrification, and hybrid gentrification approaches.

Our Team

Our Mentor

Karen Chapple, Ph.D., is Professor and Chair of City & Regional Planning at the University of California, Berkeley, where she holds the Carmel P. Friesen Chair in Urban Studies. Chapple studies inequalities in the planning, development, and governance of regions in the U.S. and Latin America, with a focus on economic development and housing.

Tiffany Patafio

Tiffany graduated from Lafayette College with a degree in Mechanical Engineering and focus in Psychology. In her time as a consultant, she has worked with clients to host training courses, such as Building your Analytics Acumen, leverage techniques such as data visualization and machine learning, and most recently helping police departments leverage social media data to better understand public sentiment. She is currently pursuing an Advanced Certificate from CUSP while working with Alliance for Downtown NY to support in their efforts to become more data driven.

Kent Pan

Kent graduated from Cornell University with a degree in environmental engineering. With experience working in NYC government on engineering design and project management, he is interested in using his studies at CUSP to help make cities more livable and sustainable through data-driven decision making.

Manrique Vargas

Manrique graduated from the University of Hong Kong with a degree in engineering, and has industry experience in data engineering, numerical modeling, and remote sensing. He is pursuing his master's degree at CUSP specializing in Urban Informatics, with a focus in using data science techniques to foster sustainable development.

Jiawen Wan

Jiawen graduated from New York University. She is currently pursing her master degree at NYU CUSP.

Tiancheng Yin

Tiancheng graduated from University of California, Davis with a degree in computer science. Before joining NYU to pursue his master degree for applied data science, he worked closely with Airbnb China as a product data scientist. He is passionate about bringing his programming skills as well as machine learning techniques into the field of data analytics to extract valuable insights.