[D] Working with PII data (documents) in Machine Learning applications

Hi everyone!

I have been working on a project on information extraction + document management. It appears that the vast majority of the documents are PII (Personal Identifiable Information). The end goal of the project does not involve any "direct" access to the PII data, however, it requires running inferences on them (for example: classifying a document as a passport or inferring the the name of the banks from a financial statement).

It would be fantastic if anyone points me …

applications data document management documents extraction information information extraction machine machine learning machinelearning machine learning applications management pii project running