大数据时代的隐私保护的挑战跟机遇.pdfVIP

大数据时代的隐私保护： 挑战和机遇 张振杰博士 伊利诺伊大学高等数字科学中心 Outline • Why privacy? • Privacy Attacking Examples • Old models and limitations – K-anonymity – L-diversity • Di erential Privacy Publishing sensitive data about individuals. • Medical research – What treatments have the best outcomes? – How can we recognize the onset of disease earlier? – Are certain drugs better for certain phenotypes? • Web search – What are people really looking for when they search? – How can we give them the most authoritative answers? • Public health – Where are our outbreaks of unpleasant diseases? – What behavior patterns or patient characteristics are correlated with these diseases? Today, access to these data sets is usually strictly controlled. Only available: • Inside the company/agency that collected the data Society would • Or after signing a legal contract benefit if we could – Click streams, taxi data • Or in very coarse-grained summariespublish some useful – Public health form of the data, • Or after a very long wait without having to – US Census data details • Or with definite privacy issues worry about privacy. – US Census reports, the AOL click stream, old NIH dbGaP summary tables, Enron email • Or with IRB (Institutional Review Board) approval – dbGaP summary tables Why is access so strictly controlled? No one should learn who had which disease. “Microdata” What if we “de-identif