# NLP-based Automated Compliance Checking of Data Processing Agreements against GDPR

**Authors:** Orlando Amaral, Muhammad Ilyas Azeem, Sallam Abualhaija, Lionel C Briand  
**arXiv:** [2209.09722](https://arxiv.org/abs/2209.09722)  
**Date:** 2022-09-20  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2209.09722v2/

## Key points

- Authors address regulatory challenges posed by GDPR in Europe
- Proposed solution aims to streamline compliance checking process mandated by GDPR for software development involving personal data processing
- Approach leverages NLP technologies and collaboration with legal experts
- System extracts relevant requirements from GDPR provisions and creates glossary table defining key legal concepts
- Automated system successfully identifies violations and satisfied requirements with high precision and recall rates
- Demonstrates significant improvements in accuracy compared to baseline NLP tools
- System's accuracy can be further enhanced with limited manual verification efforts
- Solution enhances efficiency in ensuring DPA compliance and showcases potential for technology-driven advancements in navigating complex regulatory landscapes like GDPR within software development processes.

## Summary

In their paper titled "NLP-based Automated Compliance Checking of Data Processing Agreements against GDPR," authors Orlando Amaral, Muhammad Ilyas Azeem, Sallam Abualhaija, and Lionel C Briand address the regulatory challenges posed by the General Data Protection Regulation (GDPR) in Europe. The  solution proposed by the authors aims to streamline the process of ensuring , which is mandated by GDPR for software development involving personal data processing. This approach leverages  technologies and collaboration with legal experts to extract relevant requirements from GDPR provisions and create a glossary table defining key legal concepts. By generating phrasal-level representations of DPAs' textual content and comparing them against predefined "shall" requirements, the automated system successfully identifies violations and satisfied requirements with high precision and recall rates. Compared to baseline NLP tools, this approach demonstrates significant improvements in accuracy. The authors highlight that with limited manual verification efforts, the system's accuracy can be further enhanced. This innovative solution not only enhances efficiency in ensuring DPA compliance but also showcases the potential for technology-driven advancements in navigating complex regulatory landscapes like GDPR within software development processes.