“We want to ensure everyone on Twitter sees the best content on the platform, so we’re expanding recommendations to all users, including those who may not have seen them in the past,” Twitter support tweeted on Thursday.
Twitter’s recommended tweets appear in the platform’s Home feed where tweets are organised according to an algorithm.
“Think of them as personalised suggestions that are shown to you based on actions you take on Twitter,” the company noted in its blog post in September.
“The content we recommend to you is informed by actions you take on Twitter, also known as signals,” it said, adding that the suggestions are based on topics users follow, tweets they engage with and the content other people in their network like.
Twitter noted that its recommendations team partners closely with its “Health, Trust & Safety, and Machine Learning Ethics teams to make sure we’re recommending high-quality content”.
While the change may help Twitter grow its user base – which the social media company’s new owner Elon Musk claims has hit all-time highs since his takeover – it remains to be seen how well the recommendations will be moderated.
We want to ensure everyone on Twitter sees the best content on the platform, so we’re expanding recommendations to all users, including those who may not have seen them in the past.
You can learn more about them, and how to best control your experience: https://t.co/ekYWf57JSc
— Twitter Support (@TwitterSupport) November 30, 2022
Currently, Twitter provides two types of feeds to users. One is “latest tweets”, that displays content from people one follows in chronological order and Home, a curated collection of popular tweets from the accounts one follows.
It remains unclear how efficient recommendations on the platform will be, considering Twitter more than halved its entire workforce last month from 7,500 to roughly 2,000 last month,
The company’s entire human rights and machine learning ethics teams as well as outsourced contract workers working on the platform’s safety concerns have all been fired.
The platform also enables users to manage recommendations using tools such as the “sparkle” icon at the top-right corner of the Home timeline to provide feedback on the feature.
“You can also give us feedback on recommendations you see on your Home timeline. For example, liking or Retweeting recommended content will send us a signal that you find it interesting. Selecting ‘Not interested in this Tweet/Topic’ from the Tweet menu, on the other hand, will tell us you’d like to see less of that type of content,” the company noted in its blog post.
“With millions of people signing up for Twitter every day, we’re also testing new ways for people who are new to our service to tell us more about what they’re interested in. This way we can easily recommend content, accounts, and Topics you truly want to see,” it said.