What can you ask the Textual Agent to do?

You can ask the Textual Agent to provide information about the content of the dataset files and to update the entity handling configuration.

The following are just a few examples:

Task	Example
Ask questions about the number and type of files	How many files are in the dataset? How many PDFs are in the dataset? What is the largest file in the dataset? Are there any images in the dataset?
Ask questions about the detected entity types in the file	How many files contain name entities? Which file contains the largest count of different entity types? Are there any occupation entities? How many credit card numbers are in the files?
Ask questions about specific values	Is the value John present in any of the files, and is it detected as a given name? Do any of the files contain values that start with ABC?
Ask questions about the current configuration	Are any of the entity types ignored or synthesized?
Set the entity type handling for the dataset	Ignore all location-based entity types. Synthesize name values.
Configure details for synthesized values	Synthesize names and maintain the same capitalization. Always replace the first name John with Michael. Replace all addresses with HIPAA compliant values.
Configure handling for specific document types	Ignore images in docx files. Block out tables in docx files.