Degree

Doctor of Philosophy (PhD)

Department

Division of Computer Science and Engineering

Document Type

Dissertation

Abstract

Nonprofit organizations serve a crucial role in tackling a wide range of significant social, environmental, and economic issues. But it is often hard to get a clear picture of their work because their information is spread out and it is difficult to see how they are collaborating. To address this issue we developed a web-based tool to collect scattered data—from a variety of sources, such as the IRS, social media, and the Census, into one easy-to-use resource. The tool begins by taking IRS records and geocoding each nonprofit’s physical address With its coordinates. It then retrieves census tract information from the U.S. Census Bureau’s website. To help researchers and practitioners identify the social media presence of these organizations, we present an algorithm that suggests likely Twitter pages given information from the IRS database. For collecting the candidate pages we use a combination of search engine queries, the Twitter search API, and website scraping. These candidates are then ranked using a machine learning classifier that combines the scores from multiple fuzzy name-matching algorithms and address features to assign a probability score to each match. These probability scores are then used to recommend the possible social media presence of the organizations. In testing with 89 organizations in the Baton Rouge area, our tool was able to geocode and collect their census tract information. For Twitter page recommendation, our model performed significantly better than any individual name-matching method. At the top-10 recommendation level, it failed to find the correct Twitter page for only seven organizations, compared to 18 misses using the best single algorithm. Beyond the matching task, we analyzed the connections between organizations on social media platforms, showing that they form a complex network. This network can be studied using standard graph metrics to understand how nonprofits build social capital online and where there may be opportunities to strengthen collaboration. Our goal with this tool is to allow researchers to combine the various data sources for other cities and to understand how the local nonprofits communicate and plan their strategies online. Beyond that, we also see it as a way for nonprofits and social workers to improve how they reach out, boost their visibility, and connect with potential partners for collaboration using social media.

Date

7-16-2025

Committee Chair

Baumgartner, Gerald

DOI

10.31390/gradschool_dissertations.6891

Download

Available for download on Thursday, July 16, 2026

Share

COinS