Attribute Weighting in User-Based Recommendations Algorithm

[annief]

I want to be able to tweak the Recommmended for you Algorithm to downrank certain types of content. I know the feature exists, but when I select "User-Based algorithm" the option disappears. Is there any alternatives that I could use to downrank certain attributes without having to remove them from recommendations all together?

[Rite18]

@annief i think it would not work right away, rather 1st  you need to prep a User-Based algorithm for certain data that are similar in experience 

[Gaureshk_Kodag]

Hi @annief ,

You can use the Inclusion rules option from the criteria, as demonstrated in the below screenshot -
 
Algorithm type - User based
Algorithm - Recommended for you

Filtering Key - Current Item
Here I am using a static filter where I want to exclude products where the brand is equal to "mybrand"

Let me know if this helps!

[annief]

Hello,
Thank you for the reply.

When I apply the static filter use the exclude operator, it will remove entities that matches the static filter completely. I would like to find a way to keep entities that match, include them in the recommendations, but if they match the static filter, they will show up lower in the ranking compared to entities that do not match the static filter.

I hope that makes sense?

[Gaureshk_Kodag · Accepted solution]

Hi @annief ,

I don't think there's any provisioning for your requirement while using the "recommended for you" criteria. I'll look into it further and see if there's any workaround.
Thanks!