Re-identification of De-identified Documents with Autoregressive Infilling
Preprint
Abstract
Documents revealing sensitive information about individuals must typically be de-identified. This de-identification is often done by masking all mentions of personally identifiable information (PII), thereby making it more difficult to uncover the identity of the person(s) in question. To investigate the robustness of de-identification methods, we present a novel, RAG-inspired approach that attempts the reverse process of re-identification based on a database of documents representing background knowledge. Given a text in which personal identifiers have been masked, the re-identification proceeds in two steps. A retriever first selects from the background knowledge passages deemed relevant for the re-identification. Those passages are then provided to an infilling model which seeks to infer the original content of each text span. This process is repeated until all masked spans are replaced. We evaluate the re-identification on three datasets (Wikipedia biographies, court rulings and clinical notes). Results show that (1) as many as 80% of de-identified text spans can be successfully recovered and (2) the re-identification accuracy increases along with the level of background knowledge.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Evaluating the disclosure risk of anonymized documents via a machine learning-based re-identification attackData Mining and Knowledge Discovery, 2024
- In-Context Retrieval-Augmented Language ModelsTransactions of the Association for Computational Linguistics, 2023
- Anonymisation Models for Text Data: State of the art, Challenges and Future Directions2021
- Enabling Language Models to Fill in the Blanks2020
- Synthea: An approach, method, and software mechanism for generating synthetic patients and the synthetic electronic health care recordJournal of the American Medical Informatics Association, 2017
- De-identification of patient notes with recurrent neural networksJournal of the American Medical Informatics Association, 2016
- A unified framework for evaluating the risk of re-identification of text de-identification toolsJournal of Biomedical Informatics, 2016
- C‐sanitized: A privacy model for document redaction and sanitizationJournal of the Association for Information Science and Technology, 2015
- The Probabilistic Relevance Framework: BM25 and BeyondFoundations and Trends® in Information Retrieval, 2009
- Automated de-identification of free-text medical recordsBMC Medical Informatics and Decision Making, 2008