Search Documents from the Flint Water Crisis
Explore over 300,000 documents released under Freedom of Information Act requests from the Flint Water Crisis in Flint, MI (USA) using a semantic search interface.
About
Digital correspondence can be a rich source of information on the structure and interactions of social networks, including the often obscure dynamics of government decision-making.
Thanks to the Freedom of Information Act (FOIA) and other public records laws, troves of intragovernmental correspondence are publicly available for analysis.
However, while publicly available, these datasets are often published in practically inaccessible formats (e.g., scanned printouts of email chains).
Data science tools have a significant role to play in revealing the insights locked away in these datasets.
We see a need for a research toolkit that facilitates:
- Batch processing of inaccessible large-scale data dumps of digital correspondence into time-indexed, queryable graph representations amenable for analysis
- An integrated approach to Natural Language Processing (NLP) and network analysis to reveal social network structure and quantify the spread and diffusion of different types of information (e.g., ideas, facts, beliefs, questions, requests, demands)
This toolkit is currently under development, but we intend to release it as an open-source project. We are testing our development toolkit on a 300,000+ image dataset of Freedom of Information Act responses related to the Flint Water Crisis in Flint, MI (USA). Click the button above to try our search tool.