Using Natural Language Processing to Examine State Child Maltreatment Policies and Associations With Outcomes: Multistate Cross-Sectional Study
Zhidi Luo, Richard A Epstein, Nethra Sambamoorthi, Lutfiyya N Muhammad, Neil Jordan · 2026
Read Using Natural Language Processing to Examine State Child Maltreatment Policies and Associations With Outcomes: Multistate Cross-Sectional Study by Zhidi Luo, Richard A Epstein, Nethra Sambamoorthi, Lutfiyya N Muhammad, Neil Jordan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13628039.txt.
Description
Zhidi Luo and four co-authors use language processing to study child protection policy in a paper from two thousand twenty-six. States differ widely in how they define, report, and answer child maltreatment. Past work looked at single policy pieces, not the whole policy mix a state runs.
The team coded four hundred eleven policy items from all fifty states plus two territories, using six language models in a zero-shot setup. One model, DeBERTa, performed best overall. Factor analysis found three policy domains: definitions, mandated reporting, and alternative response. Clustering grouped the jurisdictions into four policy profiles. The study population covered over seventy million children and millions of reports. Outcome gaps across clusters were then compared.
Catalog record
| CREATOR | Zhidi Luo, Richard A Epstein, Nethra Sambamoorthi, Lutfiyya N Muhammad, Neil Jordan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 57 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |