“Suspicious User Detection, powered by machine learning, is here to help you identify and restrict suspected channel ban evaders from chatting before they can disrupt your stream.”

Twitch is an American live streaming service. Which offers various opportunities to connect and develop with people. It provides streaming services like games, music, talk shows, sports, travel and outdoors, just chatting, food and drink, special events. It can be considered as another platform for creators. Some famous creators are xQc (Félix Lengyel), Disguised Toast (Jeremy Wang), Valkyrae (Rachell Hofstetter), and many others.

Twitch to use ‘Machine learning’ to detect banned user
Twitch to use ‘Machine learning’ to detect banned user

“Twitch is where millions of people come together live every day to chat, interact, and make their own entertainment together.” Twitch states.

But, where Twitch offers opportunities to millions, it also offers to those who promote inappropriate content, abusive behavior, or even some cases life-threats to others. Up till now Twitch has taken appropriate actions on them and has banned their account. But, with a simple signup process, the platform is again at their service. Even though banned again they can re-create another one. To put an end to this vicious cycle Twitch has developed a machine learning tool named “Suspicious User Detection” (SUD).

On Wednesday, December 1, Twitch posted an article titled “Battling Ban Evasion with Machine Learning.” On which, it stated, “Back in September we launched Phone Verified Chat to help you keep hate and harassment out of your channel. Today, in our ongoing quest to make Twitch safer for everyone, we’re introducing Suspicious User Detection, a tool we built to help Creators and Mods catch and take action against users who are trying to evade channel-level bans.”

Adding, “Suspicious User Detection, powered by machine learning, is here to help you identify those users based on a number of account signals. By detecting and analyzing these signals, this tool will flag suspicious accounts as either “likely” or “possible” channel-ban evaders, so you can take action as needed.”

This tool will help Creators and Mods identify suspicious users, who have met certain conditions to qualify as a likely or a possible previously banned user. The company has left it up to the streamers to either restrict them to chat or directly ban them from the channel.

Those labeled as ‘Likely’, by default they will be restricted to participate in chat. For those labeled as ‘Possible’, will be able to chat. But, in both cases, it will be the Creators’ and Mods’ decision whether to restrict or ban a certain user.

Twitch has enabled this tool by default for all channels. But has availed the option to adjust or deactivate it. One can also manually add users whom they find to be suspicious.