Unanswered

PDF data extraction - pdfs do not have consistent structure

(1) Share

Report

Posted on by Presnjmat

Hi everyone, I'm new to the AI builder.

I receive pdf documents from different sources about properties. They don't all follow the same format (different creators etc) but they all have fields of information I extract and transfer to an excel document (Name, address, sqft, etc). Is it possible to use the AI builder to do this even though the documents would be multiple pages and all have their own unique layout and structure? I'd greatly appreciate any advice. So far I have watched videos where something similar was done but by extracting data from a form that always had a consistent structure.

Categories:

AI Builder

I have the same question (0)

All responses (3)

Answers (0)

Sort by

samiak Microsoft Employee on at

Like (0)

Report
Copy link

Link copied!

Hello @Presnjmat,

You can use the Document Processing model for your specific scenario by creating multiple collections.

You'll find more details here: Overview of document processing model - AI Builder | Microsoft Learn

Samia

Was this reply helpful? Yes No
Presnjmat 4 on at

Like (0)

Report
Copy link

Link copied!

Hi @samiak
Thanks for the response. Will the same process apply even if it is not a standard form though? And it is a pdf with consistent data but laid out in different ways. Thanks

Was this reply helpful? Yes No
takolota1 4,980 Moderator on at

Like (0)

Report
Copy link

Link copied!

@Presnjmat

You may have more luck using a set up that just OCRs all the text in a given document & passes it to a GPT prompt for data extraction like this: https://powerusers.microsoft.com/t5/Power-Automate-Cookbook/Extract-Data-From-PDFs-and-Images-With-GPT/td-p/2201345

Was this reply helpful? Yes No