Douyin, the Chinese mainland version of ByteDance’s short video sharing platform TikTok, has recently added features aimed at curbing cyberviolence.
Within the digital realm, publishing comments and sending direct messages are the most common ways for people to express their opinion on major social events and public figures. However, due to the low cost and high dissemination scope of such channels, they are also prone to generate harassment, personal attacks, insults, abuse, slander and false rumors.
In order to deal with inappropriate comments, Douyin will launch a “post warning” function to intercept them. In the recent small-scale test, this function has reduced about 30,000 inappropriate comments every day.
For improper direct messages, Douyin will launch functions such as a “risk reminder” and “negative content filtering.”
When the system identifies a risk in a stranger’s direct message received by the user, it will push a pop-up asking the receiver whether or not they know the other user. If the user chooses “I don’t know,” the system will allow them to perform operations such as “blocking, reporting and removing from follower list” to ensure his or her own safety. At present, this function has sent risk alerts to 250,000 messaging sessions every day, guiding users to manage their followers independently.
SEE ALSO: Douyin Upgrades its Open Music Platform
Using the “negative content filtering” function, when the model determines that a private message received by a user is “abuse or harassment,” the message will be directly categorized as a “stranger message – spam.” The feature has served 70,000 users every day, significantly reducing the negative messages they receive.
In addition, Douyin launched the “emotions warmer” assistant to relieve the negative user feelings with an interactive message dialogue.