The social network Odnoklassniki has implemented artificial intelligence (AI) to combat offensive comments. The new solution has reduced the number of toxic comments by 47%.
The new ML model automatically identifies negative messages left under videos, photos, and posts.
We see from user requests that our audience has a need for a space for safe and respectful communication: a third of all complaints on the platform are related to negative comments, according to users.
Now, toxic commenters are blocked from writing messages for a period of one hour to one day. The model was trained on the basis of 400,000 negative reviews, and the process of improving it continues. Thanks to the introduction of AI, complaint processing has become three times faster.
In addition, algorithms are used to detect profanity, images of an erotic nature, and photographs with elements of violence. All unacceptable data is automatically hidden.
Earlier, www1.ru reported that the Odnoklassniki audience grew for the first time in five years.
Read more on the topic:
Cover it up immediately!: Odnoklassniki starts hiding erotic images
Removed the annoying factor: Odnoklassniki allowed to refuse gifts
