Forum Discussion
Sharepoint Syntex - Take specific line
Hello everyone, i currently try to extract infos from our invoices but struggle with the name of the customer. The name is always in the second row and the full second row. However i cant fin...
I would like to elaborate on the question a bit.
It would also be great if i can say "amongst other possible results please use Michael B. Smith, Jordan C. Kepler or John T. Miller) as these names will appear quite often in the contracts".
Is that possible as well?
It would also be great if i can say "amongst other possible results please use Michael B. Smith, Jordan C. Kepler or John T. Miller) as these names will appear quite often in the contracts".
Is that possible as well?
This is a tough one because the format of a name can vary so much. For example, Michael B. Smith or M.B. Smith or Mike Smith ...
If there any consistency to the line(s) that follow the name. If so, you can use an "after label" evaluation. If not, is this a good fit for AI Builder?
If there any consistency to the line(s) that follow the name. If so, you can use an "after label" evaluation. If not, is this a good fit for AI Builder?
- Well the phrase after is also defined - yet the program stops at the first form of punctuation (. , &) or after 2 "words" e.g. "Company Coca" instead of "Company Coca Cola".
I was however unable to figure out how to say "take the whole line as company name" instead of just "two words" that generally get returned.
Since I am also unable to "re-adjust" the model properly i am kinda at a lost there ^^'- There is no way to force a "take the whole line" unless it fits a RegEx pattern. You may want to try that. Make sure you label lots of samples that differ (some with 3 part names and some with 2 and some with punctuation and ...). The AI needs to "infer" that you need the entire line.
- I just labeled 20 different files and the AI still stops at the very for punctuation... This is really frustrating that you cannot tell the program "ignore punctuation" or "take line"...
I tried to give the program another explanation saying that part of the address (zip code) always comes about 15 Tokens after the name but that didnt help either.. ._.