Τομέας Υπολογιστικών Συστημάτων και Εφαρμογών
Library of the School of Science

2016-10-24

2016

Γεωργόπουλος Βασίλειος
Τουρπέτας Γεώργιος

Αλωνιστιώτη Αθανασία

ανωνυμοποίηση και τεχνικές εξόρυξης δεδομένων σε περιβάλλον έξυπνης κατοικίας

Greek

data anonymization and data mining in smart living

This thesis specifically refers to the presentation of the elements that
characterize a residence as smart, the benefits provided by such a residence,
and the risks that may arise.
A smart home can be characterized as such when by using artificial intelligence
it is able to recognize what is happening within it and when it is required to
act appropriately voluntarily. But to achieve this, it requires the collection
and proper processing of data relating both to the circumstances and the people
who move within the housing. While the utilization of these data may be
beneficial for the occupants of a smart home, it poses the risk of malicious
usage by people who can gain access to them when they should not.
One of the advantages of proper data utilization of a smart home is the
identification of "abnormal" behavior of the occupants, which may indicate the
existence of risks. Such data recovery can be done by data mining algorithms.
In this thesis, algorithmic data mining in intelligent housing data was
presented and executed to draw conclusions about the mental health of its
occupants.

Smart Living, Data Mining, Data Anonymization, Decission Tree, k-Anonymity

Yes

6-10

Yes

17

91

File access is restricted.

https://pergamos.lib.uoa.gr/uoa/dl/object/1324092