Mini Program
取消
Security
中文
EN
取消
  • Security solutions

    • # Content Security Solutions

    In order to improve the ecological security of WeChat open platform, Platform open API provides content security solutions to developers to help developers cope with issues such as sensitive content identification under text, picture, audio content types, yellow content recognition, and violent terrorist content identity, so that developers can maintain order and security of Mini Program operations.

    # I. Introduction to Safety Testing

    # 1. Text Content Security Detection:

    Function description: The text audit interface can identify pornography, current political violations, violence and other illegal and harmful content in text information, and help users to prevent violation risk in a timely and accurate manner. It can be used for content audit, sensitive information filtering, public opinion monitoring and other scenarios.

    • The function is based on a massive sensitive thesaurus of 100 thousand level, combined with a variety of text countermeasures, policy authority requirements, etc., and the use of deep learning technology, efficient identification of high-risk and harmful content. At the same time, we will continuously update and iterate according to large-scale corpus and real-time anti-manslaughter system to ensure continuous improvement of the effect.

    Application scenarios: User Profile Text Content DetectionMedia news users published articles, community comments content detectionGame users edit and upload material(Such as questions and answers uploaded by the user of the answer class Mini Program)Detection, etc.

    # 2. Photo Content Security Testing:

    Function description: Image content security based on Tencent massive data resources and deep learning technology, Providing intelligent review services for image content for developers can not only help users reduce the risks of pornography, political violations, violence and terrorism, but also greatly save manual review costs and protect the healthy development of business.

    • By learning and analyzing the skin color, posture, and scene dimensions of the picture image, it is possible to recognize pornography in the picture.
    • Provide scene recognition including facial recognition of sensitive people and sensitive events.
    • Based on public opinion analysis, it provides a more rigorous model of violence and terror, and intelligently identifies violent and bloody scenes and suspected prohibited picture content such as terrorism and extremism.

    Application scenarios: User-defined avatar detection, tool applications involving taking photos(Such as P chart, self-timer application)User Photo Upload DetectionE-Commerce Goods On The Shelf Picture DetectionImage Detection in Media User ArticlesDetection of images uploaded by social users.

    # 3. Audio Content Security Detection:

    Function description: Identify illegal content such as pornography, politics, and abuse in audio, so as to reduce labor costs and improve audit efficiency. Application scenarios: Speech Detection in Game Chat ChannelsAnchor Speech Detection in Live BroadcastAudio detection of related media content published by the forum community.

    # II. Developing Access Guidelines

    • Text Content Security Interface Document:[msgSecCheck ](https://developers.weixin.qq.com/miniprogram/dev/OpenApiDoc/sec-center/sec-check/msgSecCheck .html)
    • audio frequency/Image Content Secure Asynchronous Interface Documentation:mediaCheckAsync

    # III. Frequently Asked Questions

    # 1, please do not rely entirely on content security services
    • The Mini Program UGC content is connected to the content security service, which can effectively alleviate the manual audit, Reduce the risk of violations, but access to content security services does not mean that all problems are solved once and for all. In order to further ensure content security, we still recommend setting up manual review and confirmation in some links to make up for some shortcomings of AI algorithms.
    • For example, the API determines that the content is REVIEW, indicating that there may be risks and that manual confirmation is required.The API determines that the content of the PASS may contain missed violations, which can be checked in accordance with a certain proportion.
    # 2) If you have questions or needs for content security solutions, how should you respond?
    • WeChat open community opened a safety and risk control zone, developers can go to[Security Center Area](https://developers.weixin.qq.com/community/Minihome /question/1591986099080445956)Post interactive communication.