Document Understanding for Line items

As I use document understanding for extract the data from line items. The issue is the extra data are getting extracted.

In the screenshot you can see that upto 140 i need to extract but remaining field “Total No.Lines” and “total no.of quantity” that is also getting extracted.
please help!!!

Hello @agathiyanv Which extractor you are using. If you are using Intelligent Form Extractor or Form Extractor then it is better to create a template with the maximum number of line items.

Ex: If you create a template with 5 line items and you are trying to extract the table data with invoice of 6 line items which might causes this issue.

Also, another cause would be the layout of the document or position of the data changes apart from the template that you have created

Below video for ref

Extract Line Items using DU

1 Like

I am using FORM EXTRACTOR and i have totally 5 pages(total 140 line item data), in that the above screenshot is end of 5th page and also that last page have minimum no.of line items.
If I use that last page as template , then the other pages have omits some of the line items data.(around 25 line item data are missing)
If I use the other page having maximum number of line item, then at the last the extra data getting extracted.

@agathiyanv Form extractor is fixed template based extractor.
It will extract data basis how you have trained it.
Only the same template will be extracted properly, which you have trained.
If you have dynamic template invoices coming , better use machine learning extractor.

Also , answering your next query
If you are having fixed template , then train table for all 5 pages.
Do not use only one page template for all pages.

1 Like

Thank you for that info. But in machine learning extractor I use the invoice end point. But the ML skills are not show anything.

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.