This problem studies data inference control in massive data sets. There are two sub-projects in this project.
A prelimiary extended abstract appeared in Yi-Ting Chiang, Yu-cheng Chiang, Tsan-sheng Hsu, Churn-Jung Liau and Da-Wai Wang, "How Much Privacy? --- A System to Safe Guard Personal Privacy While Releasing Databases", Proceedings of the 3rd International Conference on Rough Sets and Current Trends in Computing (RSCTC), Springer-Verlag LNCS/AI# 2475, Pages 226--233, 2002.
Extended abstract in Proc. Asia Pacific Medical Informatics Conference (APAMI-MIC), 2000, under the title "Preserving Confidentially When Sharing Medical Data."
To estimate the value, we define the information state of the data user by a class of probability distributions on the set of possible confidential values. We further define the usefulness of information based on how easy the data user can locate individuals that fit the description given in the queries. These states and the usefulness of information can be modified and refined by the user's knowledge acquisition actions. The value of information is then defined as the expected gain of the privacy receiver and the privacy is protected by imposing costs on the answers of the queries for balancing the gain.