Twitter has made public the parts of the code that recommends content to users, allowing them to test and suggest modifications that can change the way tweets show up on their timelines.
Twitter CEO Elon Musk, who has championed code transparency, believes the move will lead to rapid improvements on the platform and increases users' trust.
The code has been uploaded on code-sharing platform GitHub, Twitter said on March 30.
Also Read | United States District Court orders Github to identify who leaked Twitter's source code
The algorithm first rounds up the top trending tweets on the platform through various sources then these tweets are ranked by machine-learning model according to relevance, Twitter said in a blog post.
Then, it checks to see if there are any tweets from people you have blocked, and removes them. It also checks for NSFW tweets and filters threads seen by users.
Also Read | Only verified accounts can vote in Twitter polls from April 15, says Elon Musk
The top-ranked tweets are then presented on a user's timeline. Twitter said the algorithm filters nearly 500 million tweets that are posted on the platform.
For now, the code released on GitHub only deals with how tweets are chosen and shown on timeline and not the other parts of the algorithm—Search, Explore and Ads.
Discover the latest Business News, Sensex, and Nifty updates. Obtain Personal Finance insights, tax queries, and expert opinions on Moneycontrol or download the Moneycontrol App to stay updated!
