Validation Station showing depending on Confidence

Hi Forum,

I would like to know if I can access the confidence value of a certain field.

In the Validation Station I can see the field name, the confidence % and the scraped value.

What if I only want to show the Validation Station if the confidence of a certain field is < 50% ?

Like this, I could show the Validation Station only for unclear scrapings, not for all scraped documents.

I know we can set the Confidence threshold for the whole document, but I would like to do this on a field level.

Thanks

@Eric_Parijs I guess you’ll need to use Export Extraction Results Before Present Validation Station with Include Confidence Checked as suggested in this psot below. You can then Access the field value which you need then you can Compare it with the Confidence value :
Post

1 Like

confidence1

Thanks superman,

The thing is that this doesn’t give me the confidence percentage. I guess it just shows if a value has been scraped.

For example, in the first image, you can see the percentage for End Date is 69%. But in column AN the confidence = 1.

What I want to achieve is to only show the Validation Station if for example the < 70% (so in this case it would show the Validation Station)

Any suggestion ?

@Eric_Parijs Are you sure you have used the Validation Station After Export Results and Write Range?

This is part of my flowchart.

@Eric_Parijs The Extraction Result that you get in Export Extraction Results after Validation Station is always 100 percent Since it is validated and accepted as the correct data, Hence you get it as 1. Use Export Extraction Results and Export Customer Data Before Present Validation Station. The corresponding confidences should be reflected in the Excel File.

The Input to Export Extraction Results should be the Output of Data Extraction Scope Not the Output of Present Validation.

3 Likes

confidence3

Awesome !

Thanks superman, you’re explanation made perfect sense of course.

2 Likes

What you can do, is “mess around” with the taxonomy by copying it to a new object and removing the fields that you do not want the user to validate.

Please note this is not recommended for massive validation (many files) as knowledge workers tend to remember order of things and position of things on the screen after processing a few files by hand…

1 Like

Thanks!

I think I could check the confidence % of a number of fields, as well as the document itself, and not process the document if any confidence % is lower than a certain %, and move them to another folder.

And than let another version of the bot have the Validation Station open for all of these documents in that folder.

Like this all the documents with a high confidence will be processed automatically, and only the uncertain ones be validated with the Validation station.

1 Like

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.