How TikTok Regulates For You Recommendation System
TikTok describes the For You feed system (photo: dock. unsplash)

JAKARTA TikTok realized thatfeedFor You is one of the important features that they must optimize. The reason is, this feature will provide video recommendations for users.

This feature is the main door of TikTok. Whenever users open the platform, they will go straight to the For You page. This shows that TikTok relies heavily on the video recommendation system they have.

Basically, thefeedFor You or familiarly known as For You Page (FYP) relies on user interactions such as videos that are liked or shared. The FYP system will also analyze the details of videos that are often viewed, such as text, voice, and hashtags.

All user interactions along with settings from the device and account will be considered by TikTok. The system will analyze the video, then provide recommendations based on the video rankings they have set.

All of these factors are processed by our recommendation system and weighted based on their value for users. Then the video is ranked to determine the user's possible interest in a content," TikTok said in an official statement.

If the video displayed by the FYP system doesn't match the user's preferences, TikTok asks its users to press the Not Interested sign on the videos they see. This sign really helps them in modifying content.

When users ignore content that they don't like at the FYP, the FYP system will be affected and there may be a Bubble Filter. This condition can interfere with experience users and this situation is very TikTok avoiding.

By optimizing personalization and relevance, there is a risk of presenting increasingly homogeneous video shows. This is a problem we take seriously because we maintain our recommendation system," explained TikTok.

To optimize the FYP display, TikTok will also limit repeat interruptions. They will not display duplicate content, content that has been viewed before, and content that is considered spam by the system.


The English, Chinese, Japanese, Arabic, and French versions are automatically generated by the AI. So there may still be inaccuracies in translating, please always see Indonesian as our main language. (system supported by DigitalSiber.id)