Implementation and evaluation of microaggregation algorithms for categorical data
Tipus de documentProjecte Final de Màster Oficial
Data2022-09-14
Condicions d'accésAccés obert
Tots els drets reservats. Aquesta obra està protegida pels drets de propietat intel·lectual i
industrial corresponents. Sense perjudici de les exempcions legals existents, queda prohibida la seva
reproducció, distribució, comunicació pública o transformació sense l'autorització del titular dels drets
Abstract
In a growingly digitalised world, the need for data privacy is apparent. Data scientists have contributed much previous work to ensure privacy regarding numerical data attributes in published datasets. However, work with categorical data tends to significantly affect the data utility concerning information loss, and less feasible research is available. The thesis aims to describe, implement and compare multiple microaggregation algorithms for categorical data. To achieve the goals of the thesis, and provide valuable output, multiple new proposals to handle categorical data based on the Mondrian algorithm were presented as part of the thesis. It was found that the proposals fared well compared to some previously presented algorithms, both in terms of algorithm execution time, potential information loss and reidentification risk.
Descripció
Different data anonymization algorithms have been proposed in the literature, but sometimes it is not easy for the practitioners to understand which one is better for different situations.
TitulacióMÀSTER UNIVERSITARI EN CIBERSEGURETAT (Pla 2020)
Col·leccions
Fitxers | Descripció | Mida | Format | Visualitza |
---|---|---|---|---|
arnold_veltmann_thesis.pdf | 2,267Mb | Visualitza/Obre | ||
all_files.zip | 57,04Mb | application/zip | Visualitza/Obre |