微软azure ai鉴黄
总览 (Overview)
The use of an ethical language on web platforms and in social networks is one of the most common themes regarding the behavior that a user should have in an online conversation; too often, in fact, we read offensive and hateful words, without any control (or very limited). For those who develop web applications that give the possibility that users may leave comments or that they can write their opinions, limiting these behaviors as much as possible must be a goal to be pursued.Developing a language verification feature to identify hateful or offensive phrases is very expensive, especially if it is to be effective.
在网络平台和社交网络中使用道德语言是有关用户在在线对话中应具有的行为的最常见主题之一。 实际上，我们经常在没有任何控制(或非常有限)的情况下阅读令人反感和仇恨的言论。 对于那些开发Web应用程序的人来说，他们可能会留下用户发表评论或发表意见的可能性，因此，必须尽可能限制这些行为。开发一种语言验证功能来识别可恨或令人反感的短语是非常昂贵，尤其是要有效时。
Microsoft comes to our rescue by providing a service on Azure which, among its many functionalities, also has the ability to identify content that is potentially offensive, risky, or otherwise undesirable: Azure Content Moderator.
Microsoft通过在Azure上提供一项服务来拯救我们，该服务在其许多功能中还具有识别可能令人反感，有风险或其他不受欢迎的内容的能力：Azure Content Moderator。
Azure内容主持人 (Azure Content Moderator)
From the Microsoft documentation (https://docs.microsoft.com/en-us/azure/cognitive-services/content-moderator/overview):
从Microsoft文档( https://docs.microsoft.com/zh-cn/azure/cognitive-services/content-moderator/overview )：
Azure Content Moderator is a cognitive service that checks text, image, and video content for material that is potentially offensive, risky, or otherwise undesirable. When this material is found, the service applies appropriate labels (flags) to the content. Your app can then handle flagged content in order to comply with regulations or maintain the intended environment for users.
Azure内容主持人是一项认知服务，可检查文本，图像和视频内容中是否有令人反感，有风险或其他不良材料。 找到该材料后，该服务将适当的标签(标记)应用于内容。 然后，您的应用可以处理标记的内容，以便遵守法规或维护用户预期的环境。
The following are a few scenarios in which a software developer or team would use Content Moderator:
以下是一些软件开发人员或团队将使用Content Moderator的方案：
- Online marketplaces that moderate product catalogs and other user-generated content. 调节产品目录和其他用户生成内容的在线市场。
- Gaming companies that moderate user-generated game artifacts and chat rooms. 负责管理用户生成的游戏工件和聊天室的游戏公司。
- Social messaging platforms that moderate images, text, and videos added by their users. 社交消息传递平台，用于审核用户添加的图像，文本和视频。
- Enterprise media companies that implement centralized moderation for their content. 企业媒体公司，对其内容实施集中审核。
- K-12 education solution providers filtering out content that is inappropriate for students and educators. K-12教育解决方案提供商过滤掉不适合学生和教育者的内容。
Very well, all this looks very interesting, but how does it work in detail? Let’s try to implement a simple solution that extracts the words it considers offensive or inappropriate from a text.
很好，所有这些看起来都很有趣，但是它如何详细工作？ 让我们尝试实现一个简单的解决方案，从文本中提取它认为令人反感或不合适的单词。
