With the massive increase in personal information stored in various electronic databases, such as medical records, financial records, web search histories, and social networks, the preservation of privacy is growing exponentially. At first glance, it may appear that simple anonymization of private information is enough to preserve privacy. However, this is often not the case; even if obvious identifiers, such as names and addresses are removed from the data, the remaining fields can still form unique “signatures” that can help re-identify individuals. To meet the privacy requirements while maintaining data utility, privacy-preserving machine learning techniques have been developed by the iDASH research team.
iDASH's privacy-preserving research targets data analysis, anonymization, and sharing along four axes of applications:
- Quantification of privacy risk
- Institution-to-institution data access
- Institution-hosted data access
- Public release of data