A prepackaged component is available to keep or delete some selected segments. It can be used to clean documents or to provide a representation of documents that focuses on only those places where something has been extracted, facilitating the information consumption for the reader.
- Go to the Processing view
- Create a new pipeline
- First, select a model in your pipeline (Entity detection model that will provide the annotations)
- Then, select “off-the-shelf annotator” component in your pipeline
- Select “Filter sentences based on annotations” component
- Adjust available parameters with the settings icon.
- Save the pipeline
- Use the Test page to give a try