Hi everyone,
I’m working on a process that needs to compare specific data points from invoices with information in customs declarations. I’ve set up a Document Understanding workflow with Digitizing, Classifying, and Data Extraction, following videos on YouTube and the UiPath Academy.
The challenge is that I need to handle many different invoice formats. I figured I couldn’t do this efficiently without using the ML Extractor. We might have to settle with processing most of them automatically and handle the rest manually. Ideally, I’d like a solution that is cost-effective but still capable of handling the majority of the work on its own.
My main issue with the ML Extractor for Invoices is that it doesn’t allow me to extract the Country of Origin, which is one of the data fields I need to compare. Is there any way to retrieve this information without building a custom Machine Learning model (that is more expensive, as I understand it)? The standard Invoices model works well for all the other data points I need.
I’ve also tried using the Form Extractor and creating templates for the most common invoices, but I still can’t extract data from the Country of Origin field. The issue is that the Country of Origin can appear in various formats and in different locations in the invoice—for example, in the items table or elsewhere on the page.
I’d really appreciate any guidance or suggestions. I’m quite frustrated with Document Understanding at the moment, though I believe it’s the right approach for this process.
Thanks in advance,
Kristine