Document Understanding: Splitting in Classic project

Help AI Center
, , , ,
1

I am currently working on creating a Document understanding classic project and would like to know how I can split a document. I currently have 1 pdf and it contains 4 document types that i would like to separate.

2

@ashton.lalchan

Extract pdf range activity should extract the docs

Cheers

3

hi @Anil_G,

  1. does this mean I would have to annotate in document manager for those documents and then split in the studio process for that specific document type?
  2. the reason I as is becuase within the 1 pdf it contains 4 different document types, but there is no way of knowing where those document types are located in the 1 pdf
4

@ashton.lalchan

Basically you would have trained with 4 doc types…so you would split the pdf into 1 page each and send each file separately to classify and depending on which type it is it would be clasified

Cheers

5

@Anil_G but how will i annotate…wouldnt that make the model misleading?

6

@ashton.lalchan

If each page is a different document type then you would annotate each type itself right…that is what classification does

cheers