User experience tests
After you have confirmed that start data is uploaded, you can perform a user experience test. The test must be completed before you use real traffic to verify the recommendation effect. After you specify Scenario ID, Test User, and Recommended Commodities, you can click Request Results to obtain recommendation results. Then, you can perform other operations based on your preferences and interests, request recommendation results, and perform tests to check whether the recommendation results meet interests.
During the test, behavioral data is automatically pushed to the system. You can view detailed information on the Behavior Messages tab of the Update Message Query page. If you deploy AIRec on an app, a mini program, or a PC, real-time behavioral data of users must be synchronized to AIRec. In the return result of a user experience test, you can click the following operations: Click, Favorite, Add, and Buy. When you click one of the operations, a behavior entry is automatically generated in the background to facilitate effect testing. When you request test results again, the return results vary based on the operation that you clicked. 
To view details about test results, go to the Update Message Query page.
Possible causes of an empty result returned in a test:
1. The values of some request parameters are invalid.
When you specify Scenario ID, make sure that the value you entered exists in the reported data.
When you specify Test User to a value that does not exist in the user table, a new user is created.
The value of Recommended Commodities must be less than the number of items that can be recommended in the scene.
In a user experience test, the maximum number of recommendation items is 10.
Note: In the recommendation results that are obtained by using a server SDK, the maximum number of recommendation items is 50.
In a scene with the service type set to related recommendations, you must also specify Existing Commodity in the format of item_id:item_type, as shown in the following figure.
2. The number of items that meet recommendation conditions is small. No available items in the item pool can be pushed to users and the result is empty.