URL: | http://userpage.fu-berlin.de/semu/software/weka/ | |
Author: | Matthias Niemann <matthias.niemann{[at]}fu-berlin.de>,Daniel Rotar <daniel.rotar{[at]}fu-berlin.de>,Maximilian Schr�der <schroedm{[at]}zedat.fu-berlin.de> | |
Maintainer: | Matthias Niemann <matthias.niemann{[at]}fu-berlin.de>,Daniel Rotar <daniel.rotar{[at]}fu-berlin.de>,Maximilian Schr�der <schroedm{[at]}zedat.fu-berlin.de> |
This package contains an anonymization-filter, that is used as a preprocessing-tool. To make sure, no human being can be identified by the quasi-identifiers that occur in the data, one can perform an anonymization. The kind of anonymization is variable, as well as the strength. Therefore the datafly-algorithm is used. Mostly, it will not find the optimal solution, but it is the fastest. For l-diversity, there is a modified datafly-algorithm applied, that generalizes data as long as the l-diversity-criterion is not fulfilled.