What is the API interface of the recommendation system?

This article is to share with you what the API interface of the recommendation system is, the editor thinks it is very practical, so I share it with you to learn. I hope you can get something after reading this article.

The API of the recommendation system needs to complete the input of user information and environment information, calculate the probability that the user likes the recalled items in this environment, and return the recommendation list in reverse order. The following describes the input parameters and returns of the recommendation service interface.

The whole recommendation service generally exposes only one large interface.

The input parameters of the API need to include these contents:

-user ID, the premise of personalization

-Environmental information, the recommended list will be different for different users in different environments.

-scene information, distinguishing functional locations, and recommendation strategies

-item ID, related recommended requirements

The API return includes these contents:

-list of items, item information and source of recall for each item

-split bucket KEY, which is used to track AB to test the effect of split bucket.

-recommended ID to track the follow-up effect of a single recommendation

The above is what the API interface of the recommendation system is, and the editor believes that there are some knowledge points that we may see or use in our daily work. I hope you can learn more from this article. For more details, please follow the industry information channel.