Using machine learning

The Document Processing configuration contains an option to enable machine learning. When this option is ticked, ETA uses manually marked up text references and connections to create a machine learning dictionary and machine learning entity extraction script, then uses them in document processing. This enables you to correct entity extraction errors and add new classes of extracted text references and connections.

The dictionary and entity extraction script can be copied into a custom configuration and used as a base for custom rules.

To use machine learning:

  1. Open the Document Processing configuration you want to use for your collectionA collection is a container for storing and organising ingested files and documents. Only the textual content is stored in collections, not the original files and documents.. Tick ‘Enable Machine Learning’ then click Save.
  2. Open a document with markup mistakes or omissions then click Mark Up.
  3. A message at the top of the Document pane confirms that Machine Learning is active.

  4. Add text references and connections as required.
  5. ETA does the following: