Development and evaluation of a semi-automatic anonymization tool for textual data
Year: 2019
Type: dissertation
Abstract: Despite its undeniable advantages, the exponential growth of data analytic capabilities implies a significant increase in the risk of personal privacy loss and re-identification of the individuals appearing in databases. A tool capable of treating the data in a way that would avoid re-identification of the user and, thus, could be shared to other companies or to study groups would be undoubtedly effective. The main objective of this project focuses on anonymization of textual data. That is, the goal is to create a tool that enables the anonymization of an input text to prevent identity disclosure while conserving as much utility as possible. This is achieved with the help of Information Theory and Natural Language Processing techniques. Additionally, the user interaction is required to make certain decisions, having as a consequence a semi-automatic anonymization tool for textual data. Because of that, we have developed a Graphical User Interface. The evaluation of the tool in automatic mode has been carried out in order to check its performance in function of certain parameters.
Author: Hèlia Brull Corretger
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2975922902