Hello everyone, I have a question regarding document understanding.
So what I have is many pdf files that have different formats, from what I understand document understanding can use machine learning extractor to extract specific values. For example invoice data (invoice number), from the invoice model.
So does machine learning extractor uses predefined models only created by Uipath, or I can add my own model? Or third party ML models? If I can have a custom model what are the requirements to do so?
And Is document understanding the best option to do so, or there are other tools that can aid me?
If you have documents of different template and it’s better to use Document understanding with AI Center to create ML model, there are various pre-built models that you can give a try under Out of the box - Document Understanding Model that can extract various types of data.
If required you can also create your own model.
you can use machine learning extractor with various ML models. There are Out of the box ML models eg for invoices, you can use document understanding extractor for custom document with many variations and you can deploy your own code/algorithm and make a custom ML package.
For your use case, if it’s invoices you can use the pre built invoices ML model which has a pre defined schema to extract. If else you can choose document understanding extractor.