Instagram Boss Gives An Explanation On How Applications Recommend Content Viewed By Users
JAKARTA - Instagram realizes that it has presented a lot of misunderstandings about how its app algorithms work, and whether they are involved in banning the shadow of certain creator content or not. The company now provides the answer.
In a recent blog post from Instagram boss Adam Mosseri, he explained about how the app ranked content on each section on Instagram.
"Instagrams do not have a single algorithm that oversees what is being done and is not seen by people in the app. We use various algorithms, classifiers, and processes, each with its own goals," Mosseri said, quoted Thursday, June 1.
"We want to make the best use of people's time, and we believe that using technology to personalize everyone's experience is the best way to do it."
Mosseri says features like Feed, Stories, Explore, Reels, Search, and others use its own algorithm that is tailored to the account owner. So, it's no surprise that most of the data used to encrypt content comes from the user himself.
For example, Stories, which are partially ranked based on how often users see someone's account Stories and engage with account stories, such as sending likes or DMs.
So Instagram can prioritize Stories from accounts they don't think they want to miss. Mosseri stated that its app will also see people's overall relationships and how likely they are connected as friends or family.
"People tend to look for their closest friends in Stories, use Exploration to find new content and creators, and be entertained at Reels. We look up various things differently in various parts of this app, and have added features and controls such as Close Friends, Favorites, and Following so that you can adapt to your experience further," Mosseri said.
VOIR éGALEMENT:
Likewise with Reels recommendations, Mosseri said what he liked, stored, shared, or interacted with users informed what to display next.
Instagram is a little more explicit in this regard, the company specifically points out the possibility of users resharing videos, watching them to completion, liking them, or clicking on audio pages as predictions.
Visuals and audio used in videos as well as information about people sharing them, such as follower level or engagement, also play a role in determining Reel's ranking.
Lastly, Mosseri also responded to the shadow ban. Some people believe, creators need to pay for ads to reach a better reach, but this information turns out to be wrong.
He emphasized that the company is not pressing content to encourage people to buy ads. With this incident, Mosseri actually increased transparency if something makes content owned by creators less visible.
"You have to know and be able to appeal. That's why we created a feature like Account Status to help you understand why your account content may not be eligible for recommendation, allow you to delete any content that affects your account, and appeal the decision if you think we made a mistake," said Mosseri.