I have a document with 50.pages from which one page data needs to be extracted
Shall I upload the whole pdf in the document manager ?
That would be wastage of the AI units. It would be best practice to split the pdf and pass the only page which is required.
Thanks,
Ashok ![]()
For training I can manually split the document and upload it !!
However, when the bot will run it is difficult for it to identify the page because there are no unique keywords!!!
Is there any other way to split the pdf and then check the pdf ??
There should be some keywords to identify that page.
Otherwise no option than uploading the whole file then.
Thanks,
Ashok ![]()
Some ideas on the automation of splitting the document.
If specific keywords on the target page are not available, you can at least try to eliminate pages as much as possible.
- Is it in a specific chapter you can identify.?
- Is there content before and or after on different pages that are consistent.? Again, think of keywords, chapters, patterns etc.
So even if identifying the specific page is not possible, if yo can reduce it from 50 pages to 10 you already save 80% of your AI unit usage…
This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.