Exctract invoice for different departments


I need to extract invoices for two different departments. One department requires data such as invoice number, due date, invoice date, and total amount.
The other department also needs to retrieve vendor name, vendor address, and items from a table.

If human validation will be required, I would like each department to display only the fields they need.

At the same time, I would like it to be one machine learning model so that it can be trained from two sources.

Is it possible to do this? What is the best way to accomplish it?


Here is my proposal:

  • Create a single ML model that includes every required field from both departments.
  • Develop two separate taxonomies, one for each department.
  • Employ distinct extractors for each department, utilizing the same ML model and the corresponding taxonomy.

There will be only one problem with training: each document validation action has missing fields for the model. You can connect the training activities with the same dataset, but it is recommended to use a document manager in the AI center and label the missing fields.