Publicação
Privas: assuring the privacy in database exploring systems
| Resumo: | Currently, given the technological evolution, data and information are increasingly valuable in the most diverse areas for the most various purposes. Although the information and knowledge discovered by the exploration and use of data can be very valuable in many applications, people have been increasingly concerned about the other side, that is, the privacy threats that these processes bring. This document follows an user-role approach within the data exploration process. These users are: Data Provider (provides the data), Data Collector (collects and stores the data provided), Data Publisher (transforms data and publishes it to be explored) and Data Explorer (retrieves information from data). All of them have privacy concerns and can address them with appropriate methods and techniques. In this Master thesis we built a system named Privas that aids the Data Publisher in its publishing process. Currently he can assure the data privacy by adopting, manually choosing and then applying the privacy-preserving data publishing techniques (PPDP). Privas accepts a repository with its description (written in a DSL) and creates a copy maintaining the information to be explored but assuring that involved individuals/organizations cannot be identified by applying PPDP techniques. Privas automatically chooses the privacy models to apply according with the description, and applies the transformation. In the end of the process metrics about the privacy loss are reported. The Domain Specific Language (DSL) – called PrivasL – was developed to easily allow the original repository description, the identification of the data entities that one wants to explore and the definition of the privacy level to be assured. To visually help end-users to describe their repositories, a web platform was developed – where after describing the repository, the correspondent PrivasL description is generated. In the end, an analysis on different kind of repositories, with different information using the Privas tool, was made – conclusions were drawn about transformations and in privacy loss. |
|---|---|
| Autores principais: | Miguel, Joana Margarida |
| Assunto: | Anonymization DSL PPDP Privacy Repositories |
| Ano: | 2020 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | Currently, given the technological evolution, data and information are increasingly valuable in the most diverse areas for the most various purposes. Although the information and knowledge discovered by the exploration and use of data can be very valuable in many applications, people have been increasingly concerned about the other side, that is, the privacy threats that these processes bring. This document follows an user-role approach within the data exploration process. These users are: Data Provider (provides the data), Data Collector (collects and stores the data provided), Data Publisher (transforms data and publishes it to be explored) and Data Explorer (retrieves information from data). All of them have privacy concerns and can address them with appropriate methods and techniques. In this Master thesis we built a system named Privas that aids the Data Publisher in its publishing process. Currently he can assure the data privacy by adopting, manually choosing and then applying the privacy-preserving data publishing techniques (PPDP). Privas accepts a repository with its description (written in a DSL) and creates a copy maintaining the information to be explored but assuring that involved individuals/organizations cannot be identified by applying PPDP techniques. Privas automatically chooses the privacy models to apply according with the description, and applies the transformation. In the end of the process metrics about the privacy loss are reported. The Domain Specific Language (DSL) – called PrivasL – was developed to easily allow the original repository description, the identification of the data entities that one wants to explore and the definition of the privacy level to be assured. To visually help end-users to describe their repositories, a web platform was developed – where after describing the repository, the correspondent PrivasL description is generated. In the end, an analysis on different kind of repositories, with different information using the Privas tool, was made – conclusions were drawn about transformations and in privacy loss. |
|---|