Twitch is introducing a new machine learning feature to help streamers protect their channels from people trying to avoid bans. The “Suspicious User Identification” dubbed tool will automatically flag people who suspect it may be “potential” or “potential” Ban Dodgers.
In the case of ex-partners, Twitch will prevent any messages they send from being shown in the chat. It will detect any mode for streamers to help those people and their channel. At that point, they can decide if they want to ban the person. By default, potential repeat trolls may send messages to the chat, but they will also be flagged by the system. Additionally, Twitch says manufacturers have the option to prevent them from sending a message in the first place.
“This tool is powered by a machine learning model that takes into account a number of cues – including but not limited to user behavior and account features – and compares data with accounts that were previously banned from a Creator’s channel to evaluate the probability of the account. Avoiding previous channel-level bans, “a Twitch spokesman told Engadget when we asked about the signals used by the system to identify potential culprits.
While Twitch plans to introduce suspicious user identification for everyone, the tool will not automatically ban users for streamers. This is by design because it is impossible to create 100 percent accurate machine learning tools in every context. “You are the expert in your community, and you should make the final call on who can participate,” the company said in a statement. . “This tool will learn from your actions and the accuracy of its predictions should improve over time.”
The tool was launched a summer after Twitch fought to capture an event called “Hate Campaign.” Attacks have seen malicious individuals use thousands of bots to spam channels with hate speech. In many cases, they have targeted marginalized community builders. Hate campaigns have become such a frequent feature of the platform that some developers have moved away from Twitch for a day. Of the company.
All products offered by Engadget are selected by our editorial team, independent of our parent company. Some of our stories include affiliate links. If you purchase something through one of these links, we may receive an approved commission.