Using Document Understanding I want to extract amount from pdf

I have 50+ documents that has amount values.
I need to assess this document and find “Amount” values from those document.

The header changes for every document. It is not the same. The position changes.

Those are W-2 TYPE FORMS

Can someone share a flow?

1 Like

@BHUSHAN_NAGAONKAR1

There are specific models for this in ai center cna try using them and train the documents

Cheers

1 Like

Hi, can you share a flow. Im using keyword classifier. The result I’m getting from it is not accurate.
I have very limited knowledge about storage bucket to what it does.

1 Like

@BHUSHAN_NAGAONKAR1

  1. You need to use ml classifier if the format is not constant
  2. Storage bucket is no where in picture…
  3. Ai center training is not a workflow but you have to do it on your dataset

Cheers

2 Likes

Here is my flow, that I was working on.
I was using Intelligent Keyword Classifier.

Can you check and let me know if it is correct




Here is the second flow that I created. Here, I’m using Keyword based classifier and Form Extractor


Here is the type of selection areas, where the label changes in every document as well as location



I will take a look into action center currently unexplored from my end.

Can you share a link or where I can learn this properly

Thankyou.

1 Like

@BHUSHAN_NAGAONKAR1

The flow is correct inplace of intelligent extractor you would use machine learning extractor with skill crwated in ai center

Go through academy as well even that would help

The keywords are changing right so this might not work…you need to train a model in ai center…

Cheers

1 Like

Thankyou so much for your time. Will definitely check

1 Like