Does pre trained model of pdf data extraction and document template uses the same technology or not?

[pavankumar007]

Hello experts,

As I was trying to extract the data from pdf using pretrained model but the data extracted was not accurate whereas when I tried it with building a document template for that pdf separately, it did extracted accurate data.

So I wanted to know whether the sap irpa uses the same technology with pre trained bot and manually building document template.

microsoftteams-image-4.png microsoftteams-image-3-ed.png screenshot-79.png

[Bob0001 · Accepted solution]

There is no general answer to your question. The document extraction internally uses various techniques to identify and extract fields. In your case a different "technology" (as you called it) is used, which might differ in the way of character recognition used, which would describe the different results.
You may report this as incident to SAP to let them check whether this can be improved.