In an effort to promote our research in the area of privacy preserving data analysis, at UT Dallas Data Security and Privacy Lab, we complied our implementation of various anonymization methods into a toolbox for public use by researchers. The algorithms can either be applied directly to a dataset or can be used as library functions inside other applications.
The toolbox currently contains 6 different anonymization methods over 3 different privacy definitions:
- Datafly
- Mondrian Multidimensional k-Anonymity
- Incognito
- Incognito with l-diversity
- Incognito with t-closeness
- Anatomy
Together with the anonymization toolbox, we also release the source code of our recent study on classifying anonymized data. In this study, we proposed methods for building distance-based classification models over anonymized data. More specifically, investigated methods include instance-based classifiers and support vector machines.
This work was partially supported by National Institutes of Health Grant 1R01LM009989, National Science Foundation Grant Career-0845803 and Air Force Office of Scientific Research MURI Grant FA9550-08-1-0265.
UTD Anonymization ToolBox is open source software issued under the GNU General Public License.
Download
Anonymization ToolBox for Windows
Anonymization ToolBox for Linux_32
Anonymization ToolBox for Linux_64
Documentation
Manual for Anonymization ToolBox
Manual for Classification with Anonymous Data
FAQ
Toolbox gives Sqlite exceptions related with platform. What should we do?
Embedded sqlite database is not platform independent. If it gives arhitecture related problems, you may need to compile sqlite driver on your platform. For compilation, follow the steps given at http://www.ch-werner.de/javasqlite/overview-summary.html. After driver compilation, replace the sqlite native library and sqlite.jar in the toolbox directory with the new ones.
What is the installation process for UTD Anonymization ToolBox?
There is no installation process for using UTD Anonymization ToolBox, some scripts are provided in main toolbox directory. Methods provided by the toolbox can be used by running those scripts after adjusting the provided configuration file according to requirements. The details can be found in README file that is located in toolbox main directory. In order to use UTD Anonymization Toolbox, your system should have Java Runtime Environment(JRE)1.5 or later.
News
- Beta version of UTD Anomymization ToolBox is released (02/22/2010).
- A new version of the Toolbox is released. This version provides graphical user interface for researchers to easily arrange parameters of the available anonymization algorithms. In additon, the new version introduces application programming interface (API) for developers. Provided API enables integration of anonymization algorithms into various privacy preserving data processing applications (03/01/2012).