Content Moderation:Features

Feature

Service

Description

References

Image Moderation 2.0

Common baseline moderation

You can moderate images for illegal visual content and textual content, such as pornographic, suggestive, politically sensitive, terrorism, prohibited, and malicious content (videos and players hidden in images).

Set the scopes of image moderation

You can set the scopes of image moderation. Over 80 labels and over 100 moderation items can be configured.

Copy image moderation services

You can copy a moderation service and configure different moderation scopes based on the needs of different services to meet the moderation requirements of these services.

Set the labels that hit custom image libraries

A label is returned if the associated custom image library is hit.

Set trustworthy images

You can specify one or more custom image libraries as moderation-free image libraries. The nonLabel_lib label is returned for images that hit the moderation-free image libraries.

Query the moderation results of images

In the Content Moderation console, you can query moderation results within the last 30 days.

Query the number of image moderations

Content Moderation collects statistics on the number of API calls. The statistics are stored for one year. You can query and export data that spans two months.

Text Moderation 2.0

Moderate chat interactions in international business scenarios

Considering international business characteristics, Text Moderation 2.0 adopts separate moderation policies and a separate labeling system for social networking and Internet entertainment services in Southeast Asia, the Middle East, Europe, and other regions.

-

Add terms to and manage terms in custom term libraries

In the Content Moderation console, you can create and manage custom term libraries and add and import terms to meet your business requirements.

-

Copy text moderation services

You can copy a moderation service and configure different moderation scopes based on the needs of different services to meet the moderation requirements of these services.

-

Use Content Moderation 2.0 in the decision engine of risk moderations

You can access Text Moderation 2.0 by using the decision engine of risk moderations. You can also configure custom labels for moderation results.

-

Video Moderation 2.0

Moderate video files

You can moderate images and voices in video files for violation information. We recommend that you perform moderations on all video files that are accessible over the Internet.

-

Moderate live streams

You can moderate images and voices in live streams for violation information. We recommend that you perform moderations on all live streams that are accessible over the Internet.

-

Adjust the policies of video moderation

You can adjust the moderation scopes and the moderation results to be returned based on common industry characteristics or your business requirements.

-

Copy video moderation services

You can copy a moderation service and configure different moderation policies to meet different moderation requirements of multiple services.

-

Query video moderation results

After a video is moderated, you can query the detailed moderation results by taskId, dataId, service, image label, and voice label.

-

Query the number of video moderations

You can view the statistics on the number of moderated videos in the console. Based on the statistics, you can develop further moderation or governance measures for specific video content.

-

Voice Moderation 2.0

Multi-language voice moderation

Moderations of voice content in Chinese, English, or a mixture of Chinese and English are supported. An internationalization labeling system is adopted and risk labels such as profanity and region labels are added.

Feature

Service

Description

References

Image Moderation 1.0

Moderate images for pornographic content

You can moderate images for pornographic and vulgar content.

• /green/image/scan
• /green/image/asyncscan and /green/image/results

Moderate images for politically sensitive and terrorism content

You can moderate images for politically sensitive, terrorism, and bloody risks, such as characters, logos, and events.

-

Moderate images for undesirable content

You can moderate images for undesirable scenes, such as smoking, drinking, and gambling.

-

Moderate images for ad violations

You can moderate text in images for pornographic content, politically sensitive content, abusive content, ad violations, and QR codes.

-

Moderate images for specific logos

You can moderate images for specific logos, such as trademarks and TV station logos.

-

Custom image library

Custom black-and-white image libraries are supported.

Text Moderation 1.0

Moderate text for pornographic content

You can moderate text for pornographic and vulgar content.

Moderate text for politically sensitive content

You can moderate text for politically sensitive content, such as characters and events.

-

Moderate text for terrorism content

You can moderate text for terrorism content.

-

Moderate text for ad violation content

You can moderate text for spam ads.

-

Moderate text for abusive content

You can moderate text for abusive content.

-

Moderate text for prohibited content

You can moderate text for prohibited content, such as contraband and illegal transactions.

-

Moderate text for undesirable content

You can moderate text for undesirable content, including content related to mammonism, negative enticement, and online violence.

-

Custom term library

Custom black-and-white term libraries are supported.

Voice Moderation 1.0

Moderate voices for abusive content

You can moderate voices for abusive content.

-

Moderate voices for undesirable content

You can moderate voices for undesirable content, including content related to mammonism, negative enticement, and online violence.

-