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

331–340 of 1001 posts

Re: Twitter's Recommendation Algorithm

#331

Post by sva_ »

CathalMullan wrote: post #55
tric wrote: post #11

From https://github.com/twitter/the-algorithm/blob/7f90d0ca342b92... ( "author_is_elon", candidate => candidate .getOrElse(AuthorIdFeature, None).contains(candidate.getOrElse(DDGStatsElonFeature, 0L))), ( "author_is_power_user", candidate => candidate .getOrElse(AuthorIdFeature, None) .exists(candidate.getOrElse(DDGStatsVitsFeature, Set.empty[Long]).contains)), ( "author_is_democrat", candidate => candidate .getOrEls…

Only used for metrics, apparently. [0] /** * These author ID lists are used purely for metrics collection. We track how often we are * serving Tweets from these authors and how often their tweets are being impressed by users. * This helps us validate in our A/B experimentation platform that we do not ship changes * that negatively impacts one group over others. */ [0]: https://github.com/twitter/the-algorithm/blob/7f…

Ahh, the group of Elons.

I was wondering why I see so many tweets by him, and what his "Group's" impression quote is.

This is actually pretty hilarious.

Re: Twitter's Recommendation Algorithm

#332

Post by jonathankoren »

afavour wrote: post #184
jaywalk wrote: post #115

Earlier quoted context omitted.

\* \* These author ID lists are used purely for metrics collection. We track how often we are \* serving Tweets from these authors and how often their tweets are being impressed by users. \* This helps us validate in our A/B experimentation platform that we do not ship changes \* that negatively impacts one group over others. \* From: https://github.com/twitter/the-algorithm/blob/7f90d0ca342b92...

Still smells to high heaven to me. Not the Elon part, I don't really care about that. But collecting metrics about "republican" vs "democrat" sounds like a particularly bad set of priorities at work.

Anyone that has worked in social network recommendations (raises hand) knows that they'll be accused of being politically biased, particularly if the recommendations aren't explicitly promoting biased news sites on the Right. (e.g. The Associated Press[0] is leftist propaganda! Where's unbiased news like Gateway Pundit[1] or InfoWars[5]?!) So data scientists and engineers will get pulled in to investigate the latest ref working[2][4], and this will let them easily determine that no, there is no bias.

None of this will matter though, because the complaints are made in bad faith.[2][3]

You may say this is biased comment, but I’m not going to engage in false equivalences, when the outrage and results of the outrage aren’t symmetrical. Cite one story where a major social network (Twitter, Facebook, Google News, YouTube, etc) publicly came out and said that they were adjusting their algorithms to make it more lefty. I’ll wait. This bad faith of the complaints are particularly obvious when the most popular and influential right wing television channel, Fox News, has been caught red handed knowingly spreading conspiracy theories for ratings.[6]

[0] "Associated Press is the least biased according to both Democrats and Republicans." https://www.businessinsider.com/most-biased-news-outlets-in-...

[1] "The Gateway Pundit (TGP) is an American far-right fake news website. The website is known for publishing falsehoods, hoaxes, and conspiracy theories." https://en.wikipedia.org/wiki/The_Gateway_Pundit

[2] https://www.latimes.com/archives/la-xpm-1996-07-22-mn-26779-...

[3] "Internal report finds ‘virtually identical’ rates of conservative and liberal topics, but guidelines updated to ‘exclude possibility of improper actions’" https://www.theguardian.com/technology/2016/may/24/facebook-...

[4] "There is some strategy to it [bashing the ‘liberal’ media]. If you watch any great coach, what they try to do is ‘work the refs.’ Maybe the ref will cut you a little slack on the next one." -- Rich Bond, 1992 Republican Party Chairman https://www.americanprogress.org/article/think-again-working...

[5] “InfoWars is an American far-right conspiracy theory and fake news website owned by Alex Jones. https://en.wikipedia.org/wiki/InfoWars

[6] https://nymag.com/intelligencer/2023/03/all-the-texts-fox-ne...

Re: Twitter's Recommendation Algorithm

#334

Post by undersuit »

sebzim4500 wrote: post #196
hutzlibu wrote: post #148

Earlier quoted context omitted.

Well yes, but I am pretty sure, elon envisoned warm welcome by the OS community and help for free and now that is off to a bad start.

I doubt he thought he'd get any help for free, given I don't think you can run this code there isn't a meaningful way to contribute. It's just about transparency or PR, take your pick.

He literally tweeted:

"No doubt, many embarrassing issues will be discovered, but we will fix them fast!"

Re: Twitter's Recommendation Algorithm

#335

Post by bcrosby95 »

lhnz wrote: post #156
tech234a wrote: post #141

Earlier quoted context omitted.

That makes sense; I guess that means Elon is considered a "group" now.

I know there's a joke about this regarding his ego and there's certainly some truth in that, however it's also quite believable that after a deployment he might have noticed the popularity of his tweets going down (since he no doubt checks his reach), so I can kind of understand how he might see "republicans", "democrats" and "celebrities_it_makes_sense_to_check_this_with_my_account_as_i_am_a_very_active_user" as cor…

The proper way to do that is create a pool of celebrities and monitor them. Not just the CEO's account.

For very active accounts, I assume that's what the "vits" or "power user" one is for. Or, heck, "vits" might actually be what you said.

Re: Twitter's Recommendation Algorithm

#336

Post by WhereIsTheTruth »

sho_hn wrote: post #157
WhereIsTheTruth wrote: post #121

Is it even what they use in production? There is code that favor Elon's tweets so I'd yes that's probably what they use

Humorous conspiracy theory: Imagine if it is not, but sanitized, and then someone added in Elon Boost to make it look credible. :-)

Or perhaps it does nothing at all, and it was there so we talk about it, the "is_democrat"/"is_republican" is also ridiculous, as if the goal was to demonstrate a point about social media in general, hmm

Re: Twitter's Recommendation Algorithm

#337

Post by PuppyTailWags »

sabellito wrote: post #213
cvhashim04 wrote: post #183

Earlier quoted context omitted.

When your visa is on the line, you’ll do anything

Engineers who work at twitter can easily find another job in the US.

It's harder with visas, because you're not gambling if you can find another job, you're gambling if your employer can acquire a visa in time for you to not be deported.

Re: Twitter's Recommendation Algorithm

#338

Post by nonethewiser »

thefreeman wrote: post #249
alfor wrote: post #122

Earlier quoted context omitted.

I don’t see a way out of this with the GPT/AI able to create fake persona in an instant.

What does that matter? If people find the content engaging then it will be amplified. If not, it shouldn't be there in the first place. This whole "AI / Bot swarm" excuse is just smoke and mirrors for "I want more people to pay twitter".

If _accounts_ find the tweet engaging

Re: Twitter's Recommendation Algorithm

#340

Post by ryzvonusef »

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

https://twitter.com/_johnforte/status/1641900138305134594

    > Twitter is also using the page rank algo that google created. Basically, if a lot of people interact with the user they create more authority in the system.
Post reply on HN