It was necessary, although it has taken too long to arrive. This is what I thought when I saw how in your last press release released by the marketing department of Google it is announced that the company is currently developing a new artificial intelligence system capable of moderating all kinds of comments "toxic”Present in any type of forum or community.
This new project has been baptized under the name of Perspective and it is nothing more than a new platform, I insist that it is still in the development phase, capable of using machine learning to process all kinds of comments completely automatically, giving them a score from 0 to 100 depending on whether they are truly constructive or everything. contrary. Undoubtedly, a tool that will surely help, at least, to limit the power that a user can have when expressing himself behind a computer screen.
Perspective, Google's new artificial intelligence platform for moderating offensive comments.
Now, tell you that Perspective is just a backup artificial intelligence systemIn other words, once the platform has evaluated and classified a comment, it is up to each editor to decide what to do with it. As a detail, tell you that this decision can also be automated on each platform so that only messages with higher scores can be published, messages with lower scores, launch a notice to the author indicating that his comment may be offensive ... this This decision will have to be made by each administrator.
Undoubtedly a very interesting tool, especially for communities that are too large whose volume of comments makes them impossible to moderate by anyone, in Google's own words:
Perspective analyzes the comments and evaluates them based on how similar they are to the comments that people have indicated as “toxicOr that they can probably cause someone to drop out of a conversation. To learn how to spot potentially toxic language, Perspective examines tens of thousands of comments tagged by dedicated reviewers. Each time Perspective finds new examples of toxic comments or receives corrections from users, it improves its ability to evaluate future comments.