The Douyin label of strong thinking is cognitive thinking.

Douyin tags are mainly divided into three categories:

1. Crowd tags: divided into basic attribute tags and behavioral interest tags.

Basic tags are based on the user’s registration information, mainly including gender, age, region, etc. In addition, the more important thing is the behavioral interest tag, which is based on the user’s actions on Douyin. Track, determine user preferences through big data analysis, and then add specific labels.

2. Short video tags: Short videos have a special tagging mechanism. For short videos released, the system will tag the video based on the effective viewing audience, commenting audience and the video content itself, and then accurately recommend it. Give more matching interest users.

3. Live broadcast room tags: divided into content tags and e-commerce tags

It is the same as the tagging based on behavioral interests mentioned above, except here it is for creators. In the same way, you can also label your live broadcast room accordingly based on the effective viewers, comments, likes, fans, etc. of the live broadcast room. This is a shallow interest tag.

The e-commerce label is based on the user’s click on the shopping cart, products, order rate, successful transaction rate, etc. in the live broadcast room. The system will give your live broadcast room an accurate e-commerce label, and then it will be based on the existing Tag recommendations or expand more people with similar tags for recommendation.

The relationship between tags and weights: tags are also divided into tags under basic weights and real-time tags

1. Tags determine the "quality" of the push, and the basic weight determines the system's push The basic weight of quantity is formed by interest tags and e-commerce tags.

To tag interest in a live broadcast room, you only need to tag the live broadcast room through the design of the live broadcast room's people and goods, script planning, target users' viewing, staying, interaction, and conversion of fans. E-commerce labels require the accumulation of historical e-commerce orders, so as to label accounts with accurate e-commerce labels. Once high-density transactions are completed within a period of time at the beginning, the accounts can be labeled with basic e-commerce crowds.

2. Real-time tags, real-time traffic forms real-time tags. In each live broadcast, accurate product planning and paid traffic must be used to continuously deepen account tags. The platform will monitor interactions and transaction crowds in real time, and promote Flow models will become more and more accurate.