All Products
Search
Document Center

Artificial Intelligence Recommendation:Overview

Last Updated:Mar 18, 2022

Terms related to a data source

Artificial Intelligence Recommendation (AIRec) uses only MaxCompute (formerly known as ODPS) as its data source. Other data sources such as Object Storage Service (OSS) and ApsaraDB RDS will also be available for use in the future.

  • MaxCompute: MaxCompute is a fast and fully managed computing platform for large-scale data warehousing. It can process terabytes or petabytes of data. For more information, see What is MaxCompute?

  • Table: A table is a data storage unit in MaxCompute. A table is a two-dimensional data structure composed of rows and columns. Each row represents a record, and each column contains fields of the same data type. One record can contain one or more columns. The column names and data types form the schema of the table. For more information about the schema of full data that is required by AIRec, see data specifications in the content, news, and item industries.

  • Partition: A partitioned table is a table with partitions. You can specify one or more columns as partition key columns to create a partitioned table. For more information, see Partition.

  • Project: A project is a basic organizational unit of MaxCompute. A project in MaxCompute is similar to a database or schema in a traditional database management system. Projects are used to isolate users and manage access requests.

  • Client: The client allows you to use MaxCompute based on a command-line tool. For more information, see MaxCompute client.

Procedure to prepare full data

1. Activate MaxCompute

If MaxCompute is activated, skip this step. Otherwise, activate MaxCompute in advance. After you activate MaxCompute, you can use it to upload full data to AIRec. For more information about how to activate MaxCompute, see What is DataWorks?

2. Upload full data to MaxCompute

MaxCompute allows you to use multiple methods to upload data. If your data is stored in Alibaba Cloud services such as OSS or ApsaraDB RDS, you can use the Data Integration service provided by DataWorks to upload the data. For more information about how to upload full data from a local machine to MaxCompute, see Overview.

3. Grant permissions to a RAM user

You can synchronize full data from a MaxCompute project to AIRec after you are granted the required permissions. You can create a RAM user and authorize the RAM user to upload full data from MaxCompute to AIRec. When you use the RAM user to upload the data, its AccessKey pair is required. For more information about how to authorize a RAM user, see Grant permissions on offline storage to AIRec.

4. Create the required data version in the AIRec console

For more information, see Use historical data to start an instance.

5. Push subsequent incremental data

AIRec synchronizes the full data only when the AIRec instance is started. Subsequent incremental data is pushed to AIRec by using SDKs rather than MaxCompute. If you do not need to create a data version or change full data, you can freeze and delete the MaxCompute project that is created for the data upload.