Anonymizing datasets with demographics and diagnosis codes in the presence of utility constraints

详细信息    查看全文

• 作者：Giorgos Poulis^a ; ^{poulis@uop.gr} ; Grigorios Loukides^b ; ^{gloukides@acm.org} ; Spiros Skiadopoulos^a ; ^{spiros@uop.gr} ; Aris Gkoulalas-Divanis^c ; ^{gkoulala@us.ibm.com}
• 关键词：Privacy ; Demographics ; Diagnosis codes ; Utility constraints ; Generalization ; Suppression
• 刊名：Journal of Biomedical Informatics
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：65
• 期：Complete
• 页码：76-96
• 全文大小：1947 K
• 卷排序：65

文摘

Patients re-identification may be performed based on their demographics and diagnosis codes. Existing approaches are unable to prevent the attack while preserving data utility. A new approach that is based on (k; km)-anonymity and utility constraints is proposed. Extensive experiments on healthcare data highlight the effectiveness of our approach.