Twitter's Recommendation Algorithm • Hacker News Board

Re: Twitter's Recommendation Algorithm

#141

Post by tech234a » Fri, Mar 31, 2023, 7:32 PM UTC

jaywalk wrote: post #115

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...

\* \* 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...

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

Re: Twitter's Recommendation Algorithm

#142

Post by bluetidepro » Fri, Mar 31, 2023, 7:32 PM UTC

Sebguer wrote: post #135

bluetidepro wrote: post #87

Earlier quoted context omitted.

> "Today, the For You timeline consists of 50% In-Network Tweets and 50% Out-of-Network Tweets on average, though this may vary from user to user." I have spent significant effort creating a network and there you go choosing to ignore my efforts by putting in 50% of crap-I-don't-want-to-see. That is why I despise your algorithm. This is just one feed (the "For You" recommendations feed), they also have the "following…

If you try to use the Following tab on Android, every refresh brings you back to the For You tab.

Is your app up to date? I have it on my iPhone, iPad, and an Android device which all have no problem always remembering the "Following" tab selection. As well as desktop/web.

Re: Twitter's Recommendation Algorithm

#143

Post by justapassenger » Fri, Mar 31, 2023, 7:32 PM UTC

meghan_rain wrote: post #106

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…

So why did they opensource it?

You must be new to Musk's business practices.

Re: Twitter's Recommendation Algorithm

#144

Post by thfuran » Fri, Mar 31, 2023, 7:32 PM UTC

krapp wrote: post #109

jeron wrote: post #96

Earlier quoted context omitted.

You say it’s not nefarious but isn’t that how echo chambers are created?

I don't believe echo chambers are nefarious - there's no hidden agenda involved with them. That's just how recommendation algorithms work, and it's what most people want. But if someone finds some code that suppresses recommendations from a specific political ideology across the board, that would be nefarious, IMO.

Echo chambers may not be nefarious, but they are insidious.

Re: Twitter's Recommendation Algorithm

#145

Post by gregw134 » Fri, Mar 31, 2023, 7:33 PM UTC

jaywalk wrote: post #115

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...

\* \* 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...

So now engineers working on the algo can ensure their launches won't lower Elon's tweet visibility. Looks like those remaining at Twitter have a knack for corporate survival.

Re: Twitter's Recommendation Algorithm

#146

Post by systemvoltage » Fri, Mar 31, 2023, 7:33 PM UTC

TMWNN wrote: post #126

systemvoltage wrote: post #72

Astounding amount of cynicism here, so I'll say something positive: Transparency is undoubtly important, I'm glad we can see how all of this works and what sort of effort goes into building a social media system. It's licensed under GPL which is a bummer (would have preferred BSD) but it's better than nothing.

>Astounding amount of cynicism here You can tell that those who rushed in to find something to criticize can't, when they are reduced to making jokes about coding stylistic conventions.

Yea I mean, all discussions about Twitter have double standards. If this was literally any other company, there would be resounding praise.

Re: Twitter's Recommendation Algorithm

#147

Post by bradly » Fri, Mar 31, 2023, 7:33 PM UTC

meghan_rain wrote: post #106

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…

So why did they opensource it?

Because they let go many of the engineers working on it?

Re: Twitter's Recommendation Algorithm

#148

Post by hutzlibu » Fri, Mar 31, 2023, 7:34 PM UTC

dawnerd wrote: post #139

hutzlibu wrote: post #128

Earlier quoted context omitted.

Judging by the many "issues" already, it might have been a bad idea to release on a friday, though.

This is 100% not their working copy.

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.

Re: Twitter's Recommendation Algorithm

#149

Post by krapp » Fri, Mar 31, 2023, 7:35 PM UTC

tric wrote: post #124

krapp wrote: post #66

Earlier quoted context omitted.

The repo suggests it's about tracking engagement metrics[0], so Team Red people see more Team Red content and vice versa. Nothing nefarious. [0] https://github.com/twitter/the-algorithm/blob/7f90d0ca342b92...

Why specifically track political parties? Where is author_is_american? Or author_is_mayonnaise_enjoyer? Maybe it was a choice made many years ago that they thought was appropriate, but we can't yet know it's not used for other purposes. We can at least be reasonably sure they've added the author_is_elon within the past year. I would have thought there would be many more descriptors, or non-controversial descriptors.…

Twitter became a very politically charged platform after 2016, mainly among Americans. They'd be idiots not to take advantage of that.

Re: Twitter's Recommendation Algorithm

#150

Post by capableweb » Fri, Mar 31, 2023, 7:35 PM UTC

gregw134 wrote: post #145

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...

So now engineers working on the algo can ensure their launches won't lower Elon's tweet visibility. Looks like those remaining at Twitter have a knack for corporate survival.

Yeah, surely the fan-boys who remain at Twitter are interested in lowering the visibility of Elon, not the opposite.