On March 21, 2024, Dr. Esteban Lopez Ochoa, an Assistant Professor of Urban and Regional Planning at the University of Texas at San Antonio, presented his research on the application of data science, machine learning, and artificial intelligence to address equity concerns stemming from rapid urban growth and development in U.S. cities. This lecture was hosted by the Urban Spatial Analytics program in collaboration with PennIUR. Dr. Ochoa’s work focuses on leveraging newly available data sources to quantify and analyze issues such as displacement, gentrification, and low-income housing demolitions, which have been challenging to measure in the past. In doing so, he aims to empower communities and level the playing field in data-driven equity advocacy for urban issues.
Dr. Ochoa used two of his projects to illustrate his work. In his first project, Dr. Ochoa developed a property-level gentrification index using proprietary micro-data from Data Axel. This approach allowed him to identify gentrification patterns at a more granular level compared to traditional methods that rely on aggregated data. By tracking household movements and comparing income levels of incoming and outgoing residents, he was able to provide a more detailed picture of gentrification processes. The results, visualized through maps, will be made available to the public via a dashboard, enabling communities to access this information and inform their decision-making processes.
The second part of Dr. Ochoa's presentation focused on predicting housing demolition orders in San Antonio using machine learning techniques. He trained models using Google Street View images of homes that had received demolition orders and those that had not. While the models achieved high accuracy in predicting at-risk homes, Dr. Ochoa also discovered that the perceptions and biases of code enforcement officers significantly influenced the number of homes identified for demolition. This finding stresses the need for bias training and the development of more objective assessment methods in code enforcement practices.
Central to Dr. Ochoa's research is the importance of community engagement. He collaborated with local residents to create a housing conditions survey using GIS and smartphones, empowering them to collect detailed data on the state of homes in their neighborhood. He hopes that this data will serve to refine the demolition prediction models and support advocacy efforts for targeted resources to improve housing conditions.
Dr. Ochoa also discussed an ongoing project involving the installation of air quality sensors in homes to investigate the relationship between housing conditions and indoor environmental quality. His research demonstrates the growing role of data-driven approaches in understanding and addressing urban equity concerns, and throughout his presentation, he emphasized the potential for data science to address critical social justice issues related to housing inequalities. Dr. Ochoa stressed the necessity of using these tools in service of the community, highlighting the importance of collaborative efforts between researchers and local residents in tackling the challenges posed by rapid urban development and growth.
Providing expert commentary on urban topics and highlighting Penn IUR's research in the context of pressing urban issues.
