Multi-Layer Privacy-Preserving Record Linkage with Clerical Review based on gradual information disclosure
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909675294818304 |
|---|---|
| author | Rohde, Florens Christen, Victor Franke, Martin Rahm, Erhard |
| author_facet | Rohde, Florens Christen, Victor Franke, Martin Rahm, Erhard |
| contents | Privacy-Preserving Record linkage (PPRL) is an essential component in data integration tasks of sensitive information. The linkage quality determines the usability of combined datasets and (machine learning) applications based on them. We present a novel privacy-preserving protocol that integrates clerical review in PPRL using a multi-layer active learning process. Uncertain match candidates are reviewed on several layers by human and non-human oracles to reduce the amount of disclosed information per record and in total. Predictions are propagated back to update previous layers, resulting in an improved linkage performance for non-reviewed candidates as well. The data owners remain in control of the amount of information they share for each record. Therefore, our approach follows need-to-know and data sovereignty principles. The experimental evaluation on real-world datasets shows considerable linkage quality improvements with limited labeling effort and privacy risks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_04178 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Multi-Layer Privacy-Preserving Record Linkage with Clerical Review based on gradual information disclosure Rohde, Florens Christen, Victor Franke, Martin Rahm, Erhard Cryptography and Security Databases Machine Learning Privacy-Preserving Record linkage (PPRL) is an essential component in data integration tasks of sensitive information. The linkage quality determines the usability of combined datasets and (machine learning) applications based on them. We present a novel privacy-preserving protocol that integrates clerical review in PPRL using a multi-layer active learning process. Uncertain match candidates are reviewed on several layers by human and non-human oracles to reduce the amount of disclosed information per record and in total. Predictions are propagated back to update previous layers, resulting in an improved linkage performance for non-reviewed candidates as well. The data owners remain in control of the amount of information they share for each record. Therefore, our approach follows need-to-know and data sovereignty principles. The experimental evaluation on real-world datasets shows considerable linkage quality improvements with limited labeling effort and privacy risks. |
| title | Multi-Layer Privacy-Preserving Record Linkage with Clerical Review based on gradual information disclosure |
| topic | Cryptography and Security Databases Machine Learning |
| url | https://arxiv.org/abs/2412.04178 |