Today's headline SMS content

In the process of recommending information, today's headlines will add a second recommendation to the content with low click-through rate, which is wrong.

Today's headline, which is recognized as a super-large traffic platform, is characterized by its intelligent recommendation system. However, some people read hundreds of thousands, millions or even tens of millions of headlines today, while others only read dozens or hundreds of traffic.

In addition to the quality of the content itself and the difference between accounts, the biggest key lies in its algorithm recommendation rules. Understanding the recommendation rules of today's headlines is the core key to the refined operation here.

Expand knowledge:

We all know that the content created by the media is for readers, so today's headlines will recommend more content that readers like and reduce content that readers don't like, so articles with poor quality will not occupy more recommended resources. Today's headline official explanation: When recommending headline articles, it will be recommended to users who are interested in it in batches.

That is, after the article is published, it is recommended to the users who are most likely to be interested in it. The criterion to judge whether users are interested is whether the labels of articles and readers are consistent, and fans with headline numbers will be the first users to see articles. After the first recommendation, the data left by readers after reading the article will play a "decisive role" in the second recommendation. Please pay attention to the words used here.

These data include click rate, collection number, comment number, forwarding number, reading rate, page stay time and so on. Today's headlines pay the most attention to the click-through rate and have the highest weight. We can understand the logic of today's headline machine in this way, that is, the more articles a reader clicks, the more likely it is to be a good article.

If the click rate of the first recommendation after the publication of the article is low, the machine will judge that the article is not worthy of further recommendation, and the second recommendation will reduce the recommendation volume; If the click rate is high and the machine determines that the article is more popular with readers, then the number of second recommendations will increase, and so on.