Model Development

[kunnark]

Goal: develop anonymization model and training scripts

Epic consist of following stories:

• Label existing new Bürokratt corpus #70
• Train a NER model using new labeled Bürokratt dataset #50
• Create and upload clean training script #51
• Improve search based component #52
• Improve Regex based component #53
• Validate manually 100 outputs of the training #54
• Improve Pseudonymization Logic #55
• Improve Anonymization process performance #56
• Improve trained model results #57

[turnerrainer]

@kunnark user story and AC title does not correspond to the style we have agreed upon.

[kunnark]

@turnerrainer This is an epic not a story. Should I use a story format here?

[turnerrainer]

@turnerrainer This is an epic not a story. Should I use a story format here?

Yes, let's have the same format throughout all issues, be they epics or not.