ICT Tac plans to make some changes to the For You page, i.e. the main screen of the application: in particular, the intention is to refine the algorithm that fills the page with an infinite stream of videos so that avoid showing too much content of the same type.
“We continue to develop new strategies to break up repetitive patterns and examine how our system can better vary the types of content suggested in a sequence. That’s why we’re testing ways to avoid recommending a similar batch of content, like crash diets, fitness, moods of sadness, or breakups, to protect us from viewing a category of content that might work well as a single video, but become problematic when viewed as a series” reads a post on the company’s official blog.
The vlogging platform is also working on a system that can detect when the algorithm only shows the user limited video types. The underlying concept that while videos discussing loneliness or weight loss may not necessarily violate platform rules, they could still negatively impact the user if they constitute the majority content displayed on the For You page.
“Our goal is for each person’s For You feed to showcase a variety of content, creators and topics,” TikTok points out, also noting that too many similar videos in the feed don’t reflect the kind of experience the company is trying to achieve. wish to offer. the public. To better develop the new algorithm mechanisms, TikTok is also talking with knowledgeable figures from the medical field, clinical psychology and artificial intelligence ethics.
TikTok is also continuing to develop an option that will allow users to filter the appearance, on the For You page, of videos related to certain words or hashtags giving the user more control over what they want to see on the home screen.
The algorithm that populates the For You page is central to how TikTok works and what has made it such a popular and successful app. The algorithm monitors user engagement patterns, such as keeping track of how long the screen (and user) dwells on particular content, as well as likes, to understand which types of videos offered successively are more likely to keep the user hooked on the service.
It’s unclear when the changes will actually be implemented, and TikTok itself warns that it will take some time to ensure the changes are made in the best way.