'mbd generates scores for items based on on a user's interaction data and metadata. These scores represent the relative certainty that 'mbd has in whether the user will interact with the item next. Higher scores represent greater certainty.

📘

Note

'mbd doesn't show scores for similar endpoint. For information on scores for ranking endpoint recommendations, see How personalized ranking scoring works.

'mbd generates scores for items relative to each other on a scale from 0 to 1 (both inclusive). At each model call 'mbd generates scores for a subset of relevant items.

If you apply a filter to recommendations, depending on how many recommendations the filter removes, 'mbd might add placeholder items. It does this to meet the top_k for your recommendation request. These items are popular items, based on amount of interactions data, that satisfy your filter criteria. They don't have a relevance score for the user.

The total of all scores equals 1. For example, if you're getting quests recommendations for a user and there are three quests appearing the Items dataset and Interactions dataset, their scores might be 0.6, 0.3, and 0.1. Similarly, if there have been 10,000 social posts in the training dataset, the highest-scoring post might have very small scores (the average score would be 0.001), but, because scoring is relative, the recommendations are still valid.

In mathematical terms, scores for each user-item pair (u,i) are computed according to the following formula, where exp is the exponential function, w̅_u and w_i/j are user and item embeddings respectively, and the Greek letter sigma (Σ) represents summation over all items with scores:

Depicts the formula used to calculate scores for each item in recommendations.

Recommendation reasons

If you use for-you endpoint, items the model wouldn't normally recommend include a reason list. These reasons explain why the item was included in recommendations. Possible reasons include the following:

• Promoted item – Indicates the item was included as part of a promotion that you applied in your recommendation request.
• Exploration – Indicates the item was included with exploration. With exploration, recommendations include items with less interactions data or relevance for the user. For more information about exploration, see Exploration (documentation pending).
• Popular item – Indicates the item was included as a placeholder popular item. If you use a filter, depending on how many recommendations the filter removes, 'mbd might add placeholder items to meet the top_k for your recommendation request. These items are popular items, based on interactions data, that satisfy your filter criteria. They don't have a relevance score for the user.