Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Twitter's Recommendation Algorithm

blog.twitter.com

Post reply on HN
Print view View original

391–400 of 1001 posts

Re: Twitter's Recommendation Algorithm

#391

Post by suddenclarity »

ryzvonusef wrote: post #304
ryzvonusef wrote: post #281

https://twitter.com/jarokrolewski/status/1641892148084629504 > the main neural network part of @Twitter recsys algo is based on 2021 work of #SinaWeibo - Chinese clone of Twitter interesting claim

Some more strange quirks: https://twitter.com/Ben_Cary_/status/1641893540614623258 > Twitter use to rank posts higher for those who had more followers/less people they follow > They are removing that as of today but kinda interesting that someone with 10k/10k followers would get less reach than if they had 10k followers and only followed 6k

Might be a language barrier from my side but it doesn't really sound like a "strange quirk" but rather a wise decision. Following people to get followers is one of the oldest spam methods on social media. It's not surprising that they would reduce the reach of accounts using those methods.

Re: Twitter's Recommendation Algorithm

#392

Post by pnt12 »

robopsychology wrote: post #256
jonathanmayer wrote: post #45

Context: I teach at Princeton and study social media and recommendation systems. From a very quick skim of the repositories, this appears to be quite limited transparency. The documentation gives a decent high-level overview of how Tweet recommendation works—no surprises—and the code tracks that roadmap. Those are meaningful positive steps. But the underlying policies and models are almost entirely missing (there are…

[flagged]

It's fast to read stuff when you have the domain knowledge. The weights won't be a 5kb Scala file: they'd probably be a big binary file, which is easy to search it github/locally after cloning.

Otherwise, if they are provided, someone in the thread will surely point to them.

Re: Twitter's Recommendation Algorithm

#393

Post by 5e92cb50239222b »

cubefox wrote: post #155
daveguy wrote: post #112

Earlier quoted context omitted.

So they could pretend to be open. It's the "Open"AI model. Open-washing?

This is a very cynical take. They should be commended for publishing recommendation code at all, which no other major social network does.

I still hear reverse-FUD about nvidia supposedly fully open-sourcing their Linux driver, when in reality they opened a tiny kernel portion of it that allows the main proprietary blob to connect to necessary kernel interfaces. You have to call out this bullshit when you see it.

Re: Twitter's Recommendation Algorithm

#396

Post by nonethewiser »

corbulo wrote: post #201

It's disappointing the comments are so obsessed with the political angle to this that there's a total lack of appreciation (or discussion) of opening up the most influential social media platform in the world.

The funny thing is that angle owes itself to Elon coming through on his promise to open source this.

This is a great thing.

Re: Twitter's Recommendation Algorithm

#397

Post by JustSomeNobody »

bboygravity wrote: post #29
nebula8804 wrote: post #10

Earlier quoted context omitted.

[flagged]

Or an alternative view: no matter what he does there will always be haters and there are massive (political) incentives to destroy his public persona. Those incentives also align nicely with those of hedge funds that are short TSLA as well as family offices that are short Tesla (hello Billy boy Gates).

> ... no matter what he does there will always be haters.

Yes. Why? Because he's an jerk that deserves it. He's done and said enough dumb shit that people should take him with an ocean's worth of salt.

Re: Twitter's Recommendation Algorithm

#398

Post by burkaman »

darth_avocado wrote: post #366
tech234a wrote: post #85

I wonder what the "author_is_elon", "author_is_power_user", "author_is_democrat", and "author_is_republican" labels are for [1]. [1]: https://github.com/twitter/the-algorithm/blob/main/home-mixe...

I would not be surprised if “author_is_elon” was added after he bought the company and worked the engineers too hard to figure out why his tweets don’t have a lot of engagement.

https://www.platformer.news/p/yes-elon-musk-created-a-specia...

Re: Twitter's Recommendation Algorithm

#399

Post by localplume »

bobbygoodlatte wrote: post #266
roddylindsay wrote: post #226

For ranking the candidates these predictions are combined into a score by weighting them: "recap.engagement.is_favorited": 0.5 "recap.engagement.is_good_clicked_convo_desc_favorited_or_replied": 11* (the maximum prediction from these two "good click" features is used and weighted by 11, the other prediction is ignored). "recap.engagement.is_good_clicked_convo_desc_v2": 11* "recap.engagement.is_negative_feedback_v2":…

"recap.engagement.is_replied": 27 "recap.engagement.is_replied_reply_engaged_by_author": 75 I wonder if this is why threads rank so obnoxiously high. They get artificially boosted by the author replying to their own tweet

isn't that the author replying to a reply on their tweet? so its promoting positive discussion, hence pushing the engagement higher?

Re: Twitter's Recommendation Algorithm

#400

Post by jstummbillig »

daveguy wrote: post #112
meghan_rain wrote: post #106

Earlier quoted context omitted.

So why did they opensource it?

So they could pretend to be open. It's the "Open"AI model. Open-washing?

If we are willing to not assume some borderline "it's what they want you to think" conspiracy play, obviously there was always going to be a lot of highly interested and qualified people taking a very close look at this and, at some point, there was always going to be very definitive conclusion of what's the deal with what they released.

If your play was "it's some source code, hence people will think we are open, and that should be really good for us", that would make you a very special kind of idiot in this space.

Post reply on HN