Poster #39 - Evolving Language in Disaster Recovery: A Longitudinal NLP Analysis of CDBG-DR Action Plans

• In Event: Friday Poster Reception

Abstract

The Community Development Block Grant–Disaster Recovery (CDBG-DR) program delivers vital federal funding to support long-term recovery in communities affected by major disasters, with particular emphasis on low- and moderate-income populations. These action plans serve as blueprints for how jurisdictions intend to rebuild infrastructure, restore housing, and address social vulnerability. Yet, little is known about how the language, priorities, and funding narratives in these plans have evolved over time. This study analyzes 180 CDBG-DR Action Plans issued between 2010 and 2023 to examine how federal, state, and local governments frame recovery and resilience. Using a multi-method natural language processing (NLP) approach, we assess shifts in discourse related to housing, infrastructure, equity, and funding strategies. By tracing changes in policy language over more than a decade, this work contributes to a deeper understanding of how disaster recovery planning adapts to emerging challenges. We conclude by suggesting how NLP can be leveraged to enhance transparency, support more inclusive planning processes, and promote greater alignment between recovery policy and community needs.

Author

• Sanam Aksha, University of Central Florida
  Presenting Author