How TikTok recommends videos for #ForYou feed
You're going through TikTok and find the same type of videos of dance challenges, while your friend's feed is bombarded with cat videos and cute animals. And then you find yourself thinking - Why is my feed so vastly different and how does it work.
TikTok's main goals to inspire creativity as well as build a global community where users can create and share authentically and connect with others.
The For You feed is part of what enables that connection and discovery.
When you open TikTok on your For You feed, you're presented with a stream of videos curated to your interests.
This feed is powered by a recommendation system that delivers content to each user that the system think will be of interest to that particular user.
TikTok has broken down the recommendation system behind the For You feed. There are also tips on how to counter some of the issues that all recommendation services can grapple with, and share tips for how you can personalize your discovery experience on TikTok.
On TikTok, the For You feed reflects preferences unique to each user. The system recommends content by ranking videos based on a combination of factors such as.
- User interactions - These are the videos you like or share, accounts you follow, comments you post, and content you create.
- Video information - These include details like captions, sounds, and hashtags.
- Device and account settings - Examples of these are your language preference, country setting, and device type.
All these factors are weighted based on their value to a user.
A strong indicator of interest would be whether or not a user finishes watching a longer video from beginning to end. Videos such as these would receive greater weight than a weak indicator, such as whether the video's viewer and creator are both in the same country.
Videos are then ranked to determine the likelihood of a user's interest in a piece of content, and delivered to each unique For You feed.
So how do I curate a personalized For You feed?
New users are encouraged to select categories of interest, like pets or travel, to help tailor recommendations to their preferences.
This also allows the app to develop an initial feed and from there, it will polish recommendations based on your interactions. However, if you don't select categories, TikTok offers a generalised feed of popular videos.
Your first set of likes, comments, and replays from this generalised feed will start an early round of recommendations.
The best way to curate your For You feed is to simply use and enjoy the app. Over time, your For You feed should increasingly be able to surface recommendations that are relevant to your interests and will be more 'you'.
However, it is worth noting that the For You feed isn't only shaped by your engagement through the feed itself.
When you decide to follow new accounts, for example, that action will effect your recommendations too. Discovering new hashtags, sounds, effects, and trending topics on the Discover tab will also invite new categories of content into your feed.
How do I see less of what I don't like?
Sometimes you may come across a video that does not do it for you. Just like you can long-press to add a video to your favorites, you can simply long-press on a video and tap "Not Interested" to indicate that you don't care for a particular video.
Users can also choose to hide videos from a given creator or made with a certain sound, or report a video that seems out of line with our guidelines. .
How to tackle challenges of recommendations
One of the challenges with recommendation engines is that they limit your experience.
To keep your For You feed interesting and varied, our recommendation system works to intersperse diverse types of content along with those you already know you love. TikTok also does not recommend duplicated content, content you've already seen before, or any content that's considered spam.
Safeguarding the viewing experience
TikTok recommendation system is designed with safety as a consideration. Reviewed content found to depict things like graphic medical procedures or legal consumption of regulated goods may not be eligible for recommendation.
Videos that have just been uploaded or are under review, and spam content such as videos seeking to artificially increase traffic, also may be ineligible for recommendation.