|
DP:
|
Programari relacionat amb privacitat de dades.
-
Implementation in Python of tools for data privacy (masking methods, information loss, disclosure risk).
For a description of the methods use:
-
Implementation of attribute disclosure for k-anonymity and cells.
For a description of the methods use:
- V. Torra, G. Navarro-Arribas (2023) Attribute disclosure risk for k-anonymity: the case of numerical data. Int. J. Inf. Sec. 22:6 2015-2024.
Springer link
-
Implementation and experiments in Python related to database integration and trade-off with data privacy.
Files:
- prog.me.def.web.txt Definitions needed for the experiments.
- prog.choquet.web.txt Definitions for games (also known as non-additive measures), including calculation of Shapley values of a given game.
- prog.sdc.web.txt Definitions for Masking methods, information loss, and disclosure risk measures.
- prog.vectors.matrices.web.txt Definitions for vector and matrices used in the other files
- prog.dbintegration.web.txt Experiments themselves, and plots used to write the paper. Note: some of the results are saved in the files, in variables, to avoid running the experiments again. Do not run the full file.
The results of our research are published in the paper:
- L. Jiang, V. Torra, Data Protection and Multi-Database Data-Driven Models Future Internet 2023, 15(3), 93;
Open access paper here
-
Implementation and experiments in Python related to explainability (Shapley value) and trade-off with data privacy.
Files:
- prog.me.def.web.txt Definitions to do the experiments, comparison of
Shapley values for pairs of original and masked files (given a ML model).
- prog.me.experiments.web.txt Experiments themselves, and plots used to write the paper. Note: some of the results are saved in the files, in variables, to avoid running the experiments again.
- prog.choquet.web.txt Definitions for games (also known as non-additive measures), including calculation of Shapley values of a given game.
- prog.sdc.web.txt Definitions for Masking methods, information loss, and disclosure risk measures.
- prog.vectors.matrices.web.txt Definitions for vector and matrices used in the other files
The results of our research are published in the paper:
- A. Bozorgpanah, V., Torra, L., Aliahmadipour, Privacy and Explainability: The Effects of Data Protection on Shapley Values, Technologies 2022, 10, 125. Open access paper here
|
|
AGOP:
|
Programari relacionat amb l'agregació d'informació.
-
Last version of the implementation of Python for non-additive measures / fuzzy measures. Including Choquet and Sugeno integrals, measure identification
from examples, Shapley and Upsilon values, interaction indices, Möbius and (max,+)-transforms, etc
-
Fuzzy measure identification from examples.
This software is described in:
- E. Turkarslan, V. Torra, Measure Identification for the Choquet integral: a Python module, I. J. of Comp. Intel. Systems 15:89 (2022) Open access here
- V. Torra, Y. Narukawa (2007) Modeling Decisions, Springer. Springer link here
-
Implementation of the transport problem for non-additive measures (also known as fuzzy measures, monotonic games) through the (max,+)-transform.
This software corresponds to this paper:
- V. Torra, The transport problem for non-additive measures, European Journal of Operational Research, 2023
Open access here
|
