- Selecting multiple overlapping fields or segments lowers extraction quality.
- Selecting a field outside a segment, when the entity for that field is extracted from that segment, prevents the field from being trained.
- Fields used in training cannot be nested more than one level deep. You can train a field inside a group of fields, but not a field inside a group that is itself part of another group. The nesting level is counted from the source field.
- You cannot create more than one segmentation model within a single document section.
- Multi-lingual NLP models are not supported.
Using NLP to process unstructured documents
Known limitations
Known limitations of NLP technology in ABBYY FlexiCapture, covering overlapping fields, field nesting depth, segmentation models, and language support.
The NLP technology in ABBYY FlexiCapture currently has the following limitations:
