All Products
Search
Document Center

Artificial Intelligence Recommendation:PAI-Rec recommendation development platform-data schema in content recommendation scenarios

Last Updated:Apr 01, 2025

Note

The following section describes the fields in the user table, item table, and behavior log table that are recommended for content recommendation scenarios. The more fields you specify, the better recommendation results you will obtain. You can also provide more fields that are not listed in the following tables to obtain better recommendation results.

User table

We recommend that the user table contains information about all users registered in your system. We recommend that you create a partition every day and synchronize daily data of all users to the partition.

Field

Required

Description

user_id

Required for users who have logged on

The unique ID of a user.

imei

Required for users who have not logged on

International Mobile Equipment Identity (IMEI)

register_time

No. We recommend that you specify this field.

The time when the account was registered. Unit: seconds. Example: 1520017038.

gender

No

The gender of the user.

age

No

The age of the user.

country

No

The country.

province

No

The region, state, or province.

city

No

The city.

ip

No

The IP address of the last logon.

education

No

The education background of the user.

career

No

The occupation of the user.

item_cnt

No

The number of content pieces that the user created from the time when the account was registered.

favorite_cnt

No

The number of favorites.

follow_cnt

No

The number of users that the user follows.

follower_cnt

No

The number of followers of the user.

last_login_time

No

The last logon time.

tags

No

The tags of the user.

Item table

We recommend that the item table contains information about all the content in your system. We recommend that you create a partition every day and synchronize daily data of all the content to the partition.

Field

Required

Description

item_id

Yes

The ID of the content.

item_type

Required for content that contains multiple types of information.

Content type, such as article and video.

status

Yes

This field indicates whether the item can be recommended.

duration

No. It is required for videos.

The duration of the video.

pub_time

Yes

The time when the item is published.

title

No. We recommend that you specify this field.

The title of the item.

category

No. We recommend that you specify this field.

The level-1 tag.

tags

No

The tags. You can configure multiple tags and separate them with semicolons (;).

author

No. We recommend that you specify this field.

The author.

abstract

No

The abstract of the content.

content

No

The body part of the content.

image_url

No

The image URL used for extracting image features.

video_url

No

The video URL used for extracting video features.

pv_count

No

The total number of times the content is exposed.

click_count

No

The total number of times the content is clicked.

praise_count

No

The total number of times the content is liked.

comment_count

No

The total number of times the content is commented.

collect_count

No

The total number of times the content is added to favorites.

share_count

No

The total number of times the content is shared.

download_count

No

The total number of times the content is downloaded.

tip_count

No

The total number of times the content is rewarded.

Behavior log table

The behavior log table contains the behavioral data of the app or the behavioral data in specific scenarios in the most recent period. We recommend that the most recent period is at least 30 to 60 days.

Field

Required

Description

user_id

Required for users who have logged on

The unique ID of a user.

imei

Required for users who have not logged on

IMEI

item_id

Required

The ID of the content.

item_type

Required for content that contains multiple types of information.

The content type, such as article and video.

request_id

No

The request ID, which is the unique identifier of each recommendation. If you leave this field empty, the accuracy of samples is affected and real-time features cannot be added. This field is optional when you create a recommendation solution, and you must configure this field, modify the training sample code, prepare training samples again, and then perform model training after you create the recommendation solution.

request_info

No. We recommend that you specify this field.

The request tracking information, such as experiment ID.

exp_id

Yes

The experiment ID returned by the PAI-Rec recommendation interface. If the result is not recommended by PAI-Rec, set this parameter to default or other values.

scene

Required in case of multiple scenarios.

The scenario.

page

No. We recommend that you specify this field in case of multiple pages.

The page.

source_page

No. We recommend that you specify this field in case of multiple pages.

The previous page.

position

No

The position of the content.

event

Yes

The behavior type, such as exposure, stay, click, favor, download, buy, cart, and evaluate.

event_time

Yes

The time when the behavior occurred.

playtime

No. We recommend that you specify this field.

The playback duration or the reading duration. Unit: seconds.

comment

No

The comment content.

net_type

No

The type of the network.

device_platform

No. We recommend that you specify this field.

The platform of the client.

device_brand

No

The brand of the client device.

device_model

No

The model of the client device.

device_system

No

The operating system of the client device.

app_version

No

The app version.

longitude

No

The longitude.

latitude

No

The latitude.

ip

No

Used to extract the country and city features.