IsmoLehtiniemi
Level 5
Options
- Subscribe to RSS Feed
- Mark as New
- Mark as Read
- Bookmark
- Subscribe
- Email to a Friend
- Printer Friendly Page
- Report Inappropriate Content
24-07-21
10:11 AM
Status:
Under Consideration
I would like to have some transparency for the decisions which Decipher uses to recognize specific field values.
Currently there are situations where either Decipher selects values that make no sense, or where Decipher does not select some values after countless training documents.
I would suggest to add some sort of easily readable log files, which would tell the user why some specific value was selected or why some field was not filled-in.
The log entries could be for example something like:
"Field 'Invoice number' was recognized based on a header value".
"Field 'Total amount' was not filled-in due to the excluded header -values."
"Field 'xxx' was recognized based on the relative field location".
"Field 'yyy' was recognized by the existing Machine Learning models."
Currently there are situations where either Decipher selects values that make no sense, or where Decipher does not select some values after countless training documents.
I would suggest to add some sort of easily readable log files, which would tell the user why some specific value was selected or why some field was not filled-in.
The log entries could be for example something like:
"Field 'Invoice number' was recognized based on a header value".
"Field 'Total amount' was not filled-in due to the excluded header -values."
"Field 'xxx' was recognized based on the relative field location".
"Field 'yyy' was recognized by the existing Machine Learning models."
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
