Anonymization of sparse multidimensional data

Mar 8, 2016 12:00 AM

Who: Manolis Terrovitis\
When: Tuesday, {{ | date_to_long_string }}, 15:00-12:00\
Where: Room 8103\
Title: {{ page.title }}

Data privacy is of increasing importance as most human activities leave digital traces in some information system. \
In this presentation I will talk about data anonymization and more specifically about protection against identity disclosure in the publication of sparse multidimensional data. The presentation will explain the notion of km-anonymity and how it is applied to collections of high dimensional data like set-values and tree-structured data. I will sketch the techniques that anonymize data through generalization, record splitting (disassociation) and algorithms that work on tree-structured data. The advantages and disadvantages of each approach will be summarized and I will discuss future research directions on data anonymization.

Previous Talks

{: .t60 } {% include list-posts tag=‘csstalk’%}