Same Template for all PDF - Document Understanding

Hi Team,

Requirement: Extract Data from PDF
Method : Document Understanding
PDF Template : Fixed 5 table’s
Resolution : Not Fixed
File Size : Not Fixed

Example: 10 input PDF
Template1 : PDF 1,2,3,4,5
Template2: PDF 6,7,8,9,10
Note: Template1&2 can’t able identify by human eyes.

But Bot can able to differentiate like if PDF 1 is used for Development then bot extracting proper data only from Template1(PDF-1,2,3,4,5)
and for Template2 need to use of one these (PDF 6,7,8,9,10) for Development.

In future, we might get more Template’s by that time Bot should work properly.

if any of you have idea please assist.

Thanks in advance.

Hi Balaji,

If bot is supplied with a new PDF template that was not configured before, it may not be able to identify the fields.
If you are using DU ML model, then you need to train all the possible templates. In future if you get a new template, you will be able to retrain it.