A prepackaged Trankit NER component is available to extract Named Entities from text.
While Kairntech is well suited to start your own entity extraction models from scratch, for some common scenarios, you don’t have to do that but can instead rely on prepackaged named entity recognition models that do the job your you. If you need entities such a person names, locations or company/organisation names, the Trankit NER component may be the solution for you.
In order to use it, the following steps need to be performed:
- Go to the Processing view
- Create a new pipeline.
- Create a processing component and select “Off-the-shelf annotator” in the drop-down list
- Select “trankitner” annotator
- Adjust available parameters with the settings icon.
- Save the pipeline, test in the Test Page or annotate automatically your documents by selecting your pipeline.
