[Adapt] First seminar by Xinhui

[Xinhui]

Hi guys,

Tomorrow I will present on our first seminar this semester. I will talk
about something related finding 'quasi-identifier'.

Here is the abstract:
Nowadays micro data sharing are quite normal and this also brings huge risk
of privacy leakages. In a published table from a relational database,
attributes are split into four groups, Explicit identifier,
Quasi-identifier, Non-sensitive attributes and sensitive attributes. While
quasi-identifier is a subset of attributes which can be used to infer
the owner of the records in the table, or used to determine the values of
sensitive attributes, with the knowledge of quasi-identifier values from
external world of public information, attacker can easily perform linkage
attack.
So before publishing, it is very crucial to find quasi-identifier and make
them masked.

If you want to know more, please read this survey:
Privacy-Preserving data publishing: A survey of recent developments
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.150.6812

Thanks,
Xinhui
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.sjtu.edu.cn/pipermail/adapt/attachments/20110920/fbf7dc87/attachment.html>