Extract Document Data with Generative Extractor returning Null

Help Document Understanding
, , , ,
1

Hello all!

I want to create an automation that extracts the Principal Address and Mailing Address of bankruptcy filings reports (PDF files) in the US. Although these files are fairly structures and standardized, the section containing the address might differ a little bit from document to document.

Here’s how I’m configuring this activity in Studio Community Edition:
image
*FileReference is just a variable containing the file path string.
The prompt is:
Field Name: Principal Address // Generative Prompt: “What is the Principal Address?”
Field Name: Mailing Address // Generative Prompt: “What is the Principal Address?”

However, the result of this extraction is always NULL ( ‘’), and I can’t figure out why. Is it because the Predefined doesn’t contain this type of file?

Appreciate any help,

2

@LeoF

May be it is not able to wxtract…try using any out of box model

Cheers

3

@Anil_G Thanks, but I was able to solve this yesterday. I just had to “reset to text builder” in the Prompt column, and it started extracting information from the PDF.