Machine Learning

This help page describes the "Machine learning" method for metadata extraction in plusmeta.

workflows

Content of this topic

Machine learning is a method in plusmeta for metadata extraction.

When using machine learning, the metadata is assigned on the basis of a trained model. You can find out how to train an AI model in plusmeta here.

To extract metadata using a trained ML model, the model must be assigned to the corresponding property.

Open the Properties view .
Select the metadata to which you want to assign the ML model.
Click on the button to open the Edit properties dialog.
The Edit properties dialog opens.
Open the Relations tab.
Click on the Add button to add a relation.
Select the relation has associated AI model from the drop-down list.
In the Target of the relation field, select the desired AI model to be used for the property.
Click CLOSE.
All your changes will be saved automatically.

Machine learning can optionally be customized with the configuration object ML prediction.

Open the objects view.
Click on the Add button to create a new object.
Select Configuration object as the object type.
Open the lower tab of the Create object dialogue.
Select the template ML-based prediction.

Note: If no template is selected during creation, it cannot be added later.
Customize parameters in the configuration object.
Enter a display name in the top tab .

Note: The drop-down list suggests titles according to the object type and template selected.
Click on CREATE OBJECT.
Select the metadata to which the ML model is to be assigned and open the Edit properties dialogue via the button.
Open the tab Relations.
Click on the Add button to add a relation.
Select the relation uses configuration from the drop-down list.
In the Relation target field, select the ML-based prediction configuration object.
Click CLOSE.
All your changes will be saved automatically.