- Le plus récent
- Le plus de votes
- La plupart des commentaires
Hello,
Is the age
column listed in the User dataset schema? You should have an entry similar to:
{ "name": "AGE", "type": "int" },
For the filter, when you go to build the expression, you should select:
- Action =
Include
- ID =
UserID
This should allow you to select the WHERE property of Users.age
. Screenshot of this below:
Hi, adding to the above.
First: Filters are deterministic, Context is a nudge. A filter will determine the recommendations fully, i.e. nothing that does not match the filter is recommended, while the context will subtly influence the recommendations, i.e. recommendations will be different if your user behaviour is different in different contexts, but there may be overlap in recommended items.
You should use item and user metadata fields if the attribute does not change over (a short) time for the same user or item (user age, user city). You should use context attributes of the interaction that change over time (device type, time of day, ...). Read more about context here
To apply a filter using an user-metadata attribute in the user-personalization recipe, you can add an "IF" clause to the expression:
Click on the "+" to add the condition and select the "IF" operator.
Contenus pertinents
- demandé il y a 6 mois
- demandé il y a un mois
- demandé il y a un an
- AWS OFFICIELA mis à jour il y a 2 ans
- AWS OFFICIELA mis à jour il y a 2 ans
Hi.. thanks for the above answer. I am looking specific to user-personalization recipe. I want the recommended items based on the location/age. Just wondering if the users metadata can be used to filter in this scenario or do I need to use it as a context?