Twitter Explains New Timeline Algorithm & Use Of Machine Learning


In a new blog post, Twitter offers an in-depth look at its timeline algorithm changes and explains more about its use of machine learning.

Over a year ago, Twitter updated its timeline algorithm to rank tweets based on how relevant the platform assumed them to be for individual users.

Before, tweets were shown in time order (from newest to oldest), but now the platform scores each tweet based on things like recency, media, interactions, who posted the tweet and the kind of content the user has found interesting in the past.

In the blog post, which could serve as inspiration for dating companies looking to switch up their content/feeds, Twitter explains that it re-evaluates tweets every time a user visits the site, to make sure their timeline is always up-to-date and relevant to them.

The platform also outlines how it uses machine learning within this algorithm and how it can measure and up-keep the model’s accuracy and efficiency.

Twitter said: “Using deep learning as the central modeling component in timeline ranking already gives great results in a production setting.

“However, the other main reason why Twitter made this change is to open the door to further improvements. In the field of machine learning, deep learning and the development of AI-related work these last few years has led to an unprecedented (and ongoing) burgeoning of new ideas and algorithms.

“We believe it is critical to let our ML-powered products potentially benefit from the full range of what is out there. We can do so by using an extensible platform that natively supports deep learning.”

To read the full post, please click here.