As promised by Elon Musk, Twitter has opened some of its source code including recommendation algorithm to public. Twitter has made two GitHub repositories containing the source code for its recommendation algorithm and some of the machine learning (ML) models. The company says that it is the “first step to be more transparent.”
“Our initial release of the so-called algorithm is going to be quite embarrassing and people are gonna find a lot of mistakes but we’re going to fix them very quickly,” Musk said.
The recommendation algorithm code deals with how the company recommends you tweets shown in Twitter’s “For You” feed. The company has not released code for its search algorithm, but Musk promises to open-source the search algorithm as well.
Twitter also published a detailed blog post to summerize its recommendation algorithm. The recommendation pipeline is made of three main stages:
- Fetch the best Tweets from different recommendation sources in a process called candidate sourcing.
- Rank each Tweet using a machine learning model.
- Apply heuristics and filters, such as filtering out Tweets from users you’ve blocked, NSFW content, and Tweets you’ve already seen.
The For You timeline contains 50% of relevant and recent tweets from followers and other 50% from people that company thinks the user will find interesting.
