If you want to ensure that media recommendations results returned from a single recommendation request are more diversified rather than looking similar, you can turn on diversification. This feature is available only for media recommendations, not generic recommendations or search.
Generally, diversification reduces the likelihood that similar content is shown in a recommendation panel, at the risk of removing some good recommendations.
Diversification is set on the serving config level. You can have multiple serving configs per media recommendations app, each with a different diversification setting. For more about serving configs, see Create and manage serving configs.
Two types of diversification are available: rule-based diversity and data-driven diversity.
Rule-based diversity
Rule-based diversity relies on the categories that are uploaded with your content. Use rule-based diversity to recommend content from a variety of categories. Diversification is configured by level, with higher levels of diversification causing fewer items to be displayed per category. This diversification type works best if your content has high-quality categories.
| Diversification level | Maximum items per category |
|---|---|
| None | Unlimited |
| Low | 3 |
| Medium | 2 |
| High | 1 |
| Auto | Depends on content |
In general, setting rule-based diversity and using filter recommendations is not advised. For more information, see Filters and diversification settings.
Data-driven diversity
Use data-driven diversity to produce recommendation results that balance relevance and diversity. Data-driven diversity learns from content metadata such as titles or categories. Instead of relying on words from a title or category, data-driven diversity uses semantic similarity to produce better-performing diversification.
| Diversification level | Maximum similar items |
|---|---|
| None | Unlimited |
| Low | 3 |
| Medium | 2 |
| High | 1 |
| Auto | Depends on content |
Diversify recommendations
To diversify your media recommendations:
In the Google Cloud console, go to the Agent Builder page.
Click the name of the app that you want to configure recommendations for.
Go to the Configurations > Serving tab.
If you have more than one serving config, select the serving config that you plan to set diversification for.
In the Result diversification section, choose how to diversify your recommendations in the Diversification type field.
Turn on the Enable diversification level toggle to see the available diversification levels.
Select a diversification level for your recommendations.
Click Save and publish.