Quick Audience:User Behavior

The user behavior table records the browsing, purchase, collection, and purchase behaviors of users. Based on user behavior data, you can perform Audience Filtering. You can also generate custom tags such as AIPL Model and Preference Tags. AIPL models can be used for AIPL User Analysis, AIPL Flow Analysis, and Audience Filtering.

1. Choose Workspace> Configuration Management> Data Import> Table Structure Configuration.

2. In the upper-right corner, choose Create Table Configuration > User Behavior. The Create User Behavior page appears.

3. Select the computing source and data table that you want to import.

   You can select a table or a view for the data table.

   • If you select Table, the system automatically identifies whether the table contains time partitions.

     • If the table contains time partitions, select the Time Partition field and enter the format of the partition value. The following figure shows the supported formats.

       Also, choose how you want to update:

       • Incremental update: saves historical data before each update and saves new data in a new partition. You must select whether to pull historical data for the first scheduling.

       • Full update: The historical data is cleared before each update and is not partitioned.

     • If the table does not contain time partitions, you do not need to select them. Full update will be used.

   • If you select a view, you need to manually select whether to include time partitions. The rest of the configuration methods are the same as those for tables. For more information, see the preceding table configuration.

   If the current account does not have the read permission on the table you selected, an error message appears, as shown in the following figure. In the MaxCompute project, grant the read permissions on the data table to the Quick Audience account. Click Authorize SQL in the following figure to obtain the authorization statement. After the authorization is complete, click Update to continue the configuration.

4. Enter a data table alias to set an alias for the data table.

   Note

   The table alias is used for the behavioral data source information on the Create AIPL Model and Crowd Filter pages.

   

5. Select the fields corresponding to the behavior time, behavior channel, behavior type, statistical indicator, behavior attribute, and user ID, and set the ID type and ID encryption method for the ID field.

   You can also set it in the field list.

6. In the field list, turn off the Import switch for fields that do not need to be imported. When disabled, the field cannot participate in any application and can be opened later during editing.

7. Set the field alias, field type, and field classification for the fields to be imported. Set the ID type and ID encryption method for the ID field.

8. Click Complete.

   If you set the Update Method parameter to Full Update, you are prompted to select whether to pull historical data for the first scheduling when you save the configuration.

   • If you select Yes, all historical partition data will be pulled and saved when you import data for the first time.

   • If you select No, only the latest partition data is pulled when the data is imported for the first time.

After you configure the table structure, you must configure a scheduling task to import data. For more information, see Scheduling Task. For more information about how to manage table structures, see Table Structure Configuration Management.