Prototüüp privaatsust säilitava andmete avalikustamise õppimiseks
Date
2017
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Erinevad organisatsioonid, valitsusasutused, firmad ja üksikisikud koguvad andmeid, mida on võimalik hiljem uute teadmiste saamiseks andmekaeve meetoditega töödelda. Töötlejaks ei tarvitse olla andmete koguja. Sageli ei ole teada andmetöötleja usaldusväärsus, mistõttu on oluline tagada, et avalikustatud andmetest poleks enam võimalik tagantjärgi privaatseid isikuandmeid identifitseerida. Selleks, et isikuid ei oleks enam võimalik identifitseerida, tuleb enne andmete töötlejatele väljastamist rakendada privaatsust säilitavaid meetodeid. Käesolevas lõputöös kirjeldatakse erinevaid ohte privaatsusele, meetodeid nende ohtude ennetamiseks, võrreldakse neid meetodeid omavahel ja kirjeldatakse erinevaid viise, kuidas andmeidanonümiseerida. Lõputöö teiseks väljundiks on õpitarkvara, mis võimaldabtudengitel antud valdkonnaga tutvuda. Lõputöö viimases osas valideeritakse loodud tarkvara.
Our data gets collected every day by governments and different organizations for data mining. It is often not known who the receiving part of data is and whether data receiver can be trusted. Therefore it is necessary to anonymize data in a way what it would be not possible to identify persons from released data sets. This master thesis will discuss different threats to privacy, discuss and compare different privacy-preserving methods to mitigate these threats. The thesis will give an overview of different possible implementations for these privacy-preserving methods. The other output of this thesis is educational purpose software that allows students to learn and practice privacy-preserving methods. The final part of this thesis is a validation of designed software.
Our data gets collected every day by governments and different organizations for data mining. It is often not known who the receiving part of data is and whether data receiver can be trusted. Therefore it is necessary to anonymize data in a way what it would be not possible to identify persons from released data sets. This master thesis will discuss different threats to privacy, discuss and compare different privacy-preserving methods to mitigate these threats. The thesis will give an overview of different possible implementations for these privacy-preserving methods. The other output of this thesis is educational purpose software that allows students to learn and practice privacy-preserving methods. The final part of this thesis is a validation of designed software.