“VKontakte” began testing a neural network to combat hostile statements

The social platform VKontakte is developing a neural network to combat hostile statements. This was announced by the press center of the company. The technology is designed to protect users from intolerance and aggressive behavior.

In contact with

The neural network will search for and calculate manifestations of intolerance, xenophobia, racism, insults and much more. Users can help train the neural network: for this, the developers have created a special category of complaints – “Hostile statements”. So far, moderators will be engaged in moderation, but this will help in the development of technology.

“We share the position of our users and partners regarding hate speech and the glorification of violence. If you notice an offensive post, please report it. It is more effective to fight against manifestations of intolerance together – and we urge users to join us in this fight “– said the head of the VKontakte moderation team Ivan Korneev.

According to the VKontakte security center, the platform rules prohibit:

  • Direct threats to health and life;
  • Suicide calls;
  • Hostile language that encourages violence and attacks on a person (or group of people) in order to humiliate their dignity;
  • Calls for segregation and incitement to violence;
  • Justification of violence, discrimination and segregation based on religious beliefs, ethnicity or sexuality, etc.

The social network also shared statistics on blocking for hostile user behavior. For six months, the service has blocked more than 520 thousand pieces of content, 2,400 communities and 1,340 accounts.

If you notice an error, select it with the mouse and press CTRL + ENTER.

Leave a Comment