The CreatorCon Call for Content is officially open! Get started here.

How to use Predictive Intelligence (PI) similarity solutions.

Marie Akahane
Tera Contributor

Hi,all.
My team is trying to introduce the Predictive Intelligence (PI) Knowledge Similarity, but it's not working as expected.
We've completed the Knowledge Similarity Solution training but PI Knowledge solution is not visible on any platform.
We've also found following document and proceeded to step 1~3 but but SOW Agent Assist does not show anything.
https://www.servicenow.com/docs/ja-JP/bundle/yokohama-platform-user-interface/page/administer/worksp...

 

2025-06-10 103333.png

 

Is there a way to use PI similarity solution in SOW? If so, please let me know the necessary settings.

1 ACCEPTED SOLUTION

@Marie Akahane 

Kindly ensure that the correct Agent Assist is selected in the Table Configuration. Specifically, Agent Assist [incident] should be included in the SOW.

Thank you.
Esh

View solution in original post

4 REPLIES 4

Eshwar Reddy
Kilo Sage

Hi @Marie Akahane 

After training the model in ServiceNow, follow these steps to show recommendations in Agent Assist:

  1. Create a Trend Definition – Go to the navigator, search for Trend Definition, and create one.

  2. Set up a Recommendation – Open  Recommendations and create a new recommendation to link your model.

Thanks
Esh

Hi @Eshwar Reddy 
I've already added suggested settings as shown in following capture, but the situation is not improving.
↓ Target Similarity Definition

MarieAkahane_0-1749605223635.png

MarieAkahane_1-1749605366819.png

1. Create a Trend Definition 

MarieAkahane_2-1749605472313.png

2. Set up a Recommendation

MarieAkahane_3-1749605737926.png

MarieAkahane_4-1749605783507.png

Are there any mistakes or omissions in my settings? Or is there any other approach we can try?

Kindly, please confirm.

@Marie Akahane 

Kindly ensure that the correct Agent Assist is selected in the Table Configuration. Specifically, Agent Assist [incident] should be included in the SOW.

Thank you.
Esh