Can UiPath AI resolve item name variations and map them to the correct item code?

Help AI Center
1

Hello everyone,

I’m wondering if it’s possible to use UiPath’s AI capabilities to resolve variations in item names and map them to the correct item codes.

For example, let’s say we have the following item master data:

Item Code, Item Name
SCREW001, Hex Head Screw M8
SCREW002, Phillips Pan Head Screw M4
SCREW003, Flat Head Machine Screw M6
SCREW004, Socket Head Cap Screw M5
SCREW005, Self-tapping Screw M3
SCREW006, Wood Screw M10
SCREW007, Round Head Wood Screw M4
SCREW008, Shoulder Screw M6
SCREW009, Countersunk Head Screw M8
SCREW010, Thread Forming Screw M5

Now, imagine that the order data is received not as a CSV but in an email message, like this:

===========================================
Dear Sir,
Thank you for your continued support.
Please provide us with 5 woodworking screws.
Thank you for your cooperation.

Best Regards

  • Search for similar item names within the text.
  • From the search results, match “woodworking screws” to the item name “Wood Screw M10” in the master data.
  • Find the corresponding item code, which is “SCREW006.”

I want to extract the term “woodworking screws” from this email and map it to the correct item code, which is SCREW006.

Is it possible to accomplish this with UiPath’s AI tools? If so, I would appreciate any guidance on how to set up the workflow or configuration.

Thank you for your support!

2

@mkt.scott4

It more looks like a text classification use case

This should solve

Cheers

3

I read your article. The example in the article seems to be able to classify emotions from sentences.
Happiness, anger, Positive, etc. etc.

{
“class”: “Positive”,
“confidence”: 0.9422031841278076
}

Is it possible to classify products from sentences?

4

@mkt.scott4

that is just an example…if you see further you can train on your own data to get the classification you need…so you can train few phrases so that it would infer

cheers

5

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