Sign In
 
 
 
New ItemNew Item

Full name

Anna Monreale 

Title

Privacy and Security Issues in Data Mining 

Start Time

16:00 

Location

Gerace 

Abstract

The field of data mining is gaining importance thanks to the availability of large amounts of data, easily collected and stored via computer systems.
Recently, the large amount of data, gathered from various channels, contains much personal information. When personal and sensitive data are published and/or analyzed, one important question to take into account
is whether the analysis violates the privacy of individuals whose data is referred to. The importance of privacy is growing constantly. For this reason, many research works have focused
on privacy-preserving data mining, proposing novel techniques that allow to extract knowledge while trying to protect the privacy of users. Some of these approaches aim at individual privacy while others aim at corporate privacy.
A body of research related to coorporate privacy is privacy-perserving outsourcing of data mining tasks.
In my research, I address the following two problems:
(a) anonymization of various forms of data and/or data mining results
(b) efficient and secure outsourcing of various forms of data mining and analytical querying.
A common characteristic of these two problems is that the data and/or analysis results mined
from the data are intended to be shared with parties other than the data owner. Instead, the key distinction between them is that, in the context of secure outsourcing of data mining both the underlying data and the mined results are not intended for sharing and must remain secure.
I would like to address the problem of point (a) for different kinds of data, such as sequential data,
spatio-temporal data, social networks data, etc. Specifically, I want to provide
frameworks to allow data owners to mine their data and to share these data and/or the exacted knowledge without to compromise the respondents’ privacy.
Concerning the problem of point (b), I focus on efficiency and privacy issues related to the model of outsourcing of data mining and analysis tasks. To protect sensitive information (data and data mining and analysis results) in the outsourced database, privacy policies must be defined and enforced.

Keywords

Privacy, Security, Data Mining 

Supervisor(s)

Dino Pedreschi e Fosca Giannotti 

Notes

 

Session

Attachments
AnnaMonreale.ppsx    
Created at 8/20/2009 2:12 PM  by  
Last modified at 8/31/2009 12:55 PM  by Cristian Dittamo