Anonymisation of personal data
Protecting personal data such as name, address and date of birth as well as other identifiers play an important role in every company, especially with regards to the new EU General Data Protection Regulation (GDPR). If documents have to be processed outside the company for procedures such as external audits, archiving or the like the personal information in data has to be anonymised beforehand.
Simple keyword-based methods aren’t able to carry out automatic anonymisation, especially if the quality of the OCR-documents is fluctuating. The Glanos DataSphere combines semantic and linguistic methods with extensive dictionaries, which makes it possible to identify and anonymise sensitive information such as name, address and date of birth with just a small amount of training data anonymsed by hand, even in documents with bad OCR quality.
Advantages of our solutions:
- Customers get access to the Glanos DataSphere locally or online to control the automatic automatisation on their own.
- Reliable anonymisation with the help of linguistic and AI-based methods from the first data set.
- Improvement of results after providing only a few documents as an example.
- Option of additional pseudonymisation (backwards encoding possible) of sensitive data.
- Continuous improvement of quality thanks to self-learning algorithms integrated in the DataSphere.