Extracting Components from Privacy Statements with Text Mining

Abstract

The effect of the increasing awareness about privacy and new emerging legislation results in an inviting and agitated field of research. This project focuses on an automated approach for analyzing privacy statements. More specifically, text mining is used to first pre-process privacy statements into privacy requirements, followed by extracting components from those privacy requirements. The components, which are aligned with the definitions from the new General Data Protection Regulation (GDPR) that will be enforced 25 May 2018, are identified by a literature study and extracted with text mining techniques. Dependency parsing and text chunking are found to be the best combination of text mining techniques to achieve the goal of this project to extract components from privacy statements.