Identification of Sensitive Content in Data Repositories to Support Personal Information Protection
This article presents a two-step approach focusing on the identification of sensitive data within documents. The proposed pipeline first detects the domain of a document, then identifies the sensitive information it contains. Detection of domains allows to better understand the context of a documents, hence supports the disambiguation of potentially sensitive information. The prototype considers three domains: health, business and “other”. The system developed for the domain detection step is built and evaluated on a corpus composed of clinical notes, and articles about business or art from Forbes, Reuters, and The New York Times. The identification of sensitive information relies on a Conditional Random Fields (CRF) model.
KeywordsCompliance Domain detection Named-Entity Recognition Natural language processing Personal Health Information Sensitive information
As part of this work, the Deidentified Clinical Records used were provided by the i2b2 National Center for Biomedical Computing funded by U54LM008748 and were originally prepared for the Shared Tasks for Challenges in NLP for Clinical Data organized by Dr. Ozlem Uzuner, i2b2 and SUNY.
- 1.Bodnari, A., Deleger, L., Lavergne, T., Neveol, A., Zweigenbaum, P.: A Supervised named-entity extraction system for medical text. In: CLEF (Working Notes) (2013)Google Scholar
- 2.Centers for Medicare & Medicaid Services and others: The Health Insurance Portability and Accountability Act of 1996 (HIPAA) (1996). http://www.cms.hhs.gov/hipaa
- 3.Finkel, J.R., Grenager, T., Manning, C.: Incorporating non-local information into information extraction systems by gibbs sampling. In: Proceedings of the 43rd annual meeting on association for computational linguistics, Association for Computational Linguistics, pp. 363–370 (2005)Google Scholar
- 5.Manning, C., Surdeanu, M., Bauer, J., Finkel, J., Bethard, S., McClosky, D.: The stanford CoreNLP natural language processing toolkit. In: Proceedings of 52nd Annual Meeting of the Association for Computational Linguistics: System Demonstrations, pp. 55–60 (2014)Google Scholar
- 8.Stenetorp, P., Pyysalo, S., Topić, G., Ohta, T., Ananiadou, S., Tsujii, J.: BRAT: a web-based tool for NLP-assisted text annotation. In: Proceedings of the Demonstrations at the 13th Conference of the European Chapter of the Association for Computational Linguistics, Association for Computational Linguistics, pp. 102–107 (2012)Google Scholar