Configure AI Search Relevancy Rules to boost Result Score based on match between User profile value and Taxonomy Topic

Gavin Jolly2
Tera Contributor

I would like to improve relevancy based on a users data. Specifically, if a user has a an attribute which matches a Topic related to an article can we boost those results

The scenario is a user has a "skill level" defined in a field in sys_user. This skill level matches a Topic with the same skill level. Example, a User has the skill level "Level 2" and the topic has the Topic Path "Expertise > Level 2"

Is there a way to configure a Relevancy rule to boost results where the Topic and sys_user.skill_level match (Note this is an abstract example to illustrate my design objective. We are not using SKILLS or other features.)

Looking at the Relevancy Rules we need to select specific content for each Relevancy Rule. In our case I would like

  1. Rules are defined as part of search. If one rule needed for each value we could do that, would need about 20 rules to match the possible Topic/Skill values
  2. Content, mainly knowledge, can be defined, related to topics 
  3. AI Search will apply the Relevancy Rules against the content

I think this is a valid AI Search use case and could be part of the core product. I have looked through the DOCS, AI Search FAQs, and this forum and cannot find anything matching this scenario.  

2 REPLIES 2

Gerard Dwan
ServiceNow Employee
ServiceNow Employee

Hi Gavin, 

 

I think there are a couple of features missing to have this work end-to-end, but it is a great use case. Do you mind posting it to the idea portal?

Gavin Jolly2
Tera Contributor

Gerard, I have a list of missing features from Employee Centre as a result of an ongoing project. I had a meeting with someone from the Product team and presented this and other items to them.

 

I assume they have added it to the idea portal.