The Facebook Algorithm Mom Problem • Hacker News Board

Re: The Facebook Algorithm Mom Problem

#121

Post by chickenfries » Wed, Jul 12, 2017, 6:17 PM UTC

delecti wrote: post #33

chickenfries wrote: post #31

Earlier quoted context omitted.

Pruning my friends improved my experience, but they grow back over time as people rediscover you or you make new acquaintances. I wonder if you could make a game or website that forced you to prune your facebook friends down, perhaps by looking at your social graph and telling you that you haven't messaged with a person in x years or something, or that you've never even "liked" something they posted. Some heuristics…

Here's a trick to constantly pruning down your friends list. Don't worry about pruning down your whole list, just check the people who have birthdays every day you log in. I just checked; there was a single friend who had a birthday today, and I haven't had any contact with her in years. That makes for a nice easy prune. Of course the downsides are that you risk disappointing someone who obsessively checks their frie…

Good idea! Thanks for the tip.

Re: The Facebook Algorithm Mom Problem

#122

Post by erikb » Wed, Jul 12, 2017, 6:20 PM UTC

I also have a problem with social media in general, especially with following people instead of institutions/groups. Usually what 99.9% of people like is totally not what I like. So if you base the content I should consume on the assumptions that I like what my connections liked you are nearly going the opposite direction of what I want.

PS: Maybe some of you have the experience of having an active following. I notice that many social networks like Twitter, FB, Youtube, allow comments. But almost never does the content creator/sharer actually react to comments. Some may use comments in future content, but some don't react at all. Are these people not even reading the comments? Why are people commenting when it's so obvious that it's just going down a black hole? For instance, on Twitter a share with additional text is nearly the same amount of work than a comment. And it's obvious that you reach more people by share-commenting rather than simply writing your comment underneath the content. So why do people do that? And why does Twitter has the option to comment even?

Re: The Facebook Algorithm Mom Problem

#123

Post by Dylan16807 » Wed, Jul 12, 2017, 6:21 PM UTC

ryanbrunner wrote: post #95

I find a lot of sites feel like they're overtuning their recommendation engines, to the detriment of using the site. YouTube is particularly bad for this - given the years of history and somewhat regular viewing of the site, I feel like it should have a relatively good idea of what I'm interested in. Instead, the YouTube homepage seems myopically focused on the last 5-10 videos I watched.

Trying youtube music recently made that abundantly clear, with an awful awful playlist extrapolated from something like 4 recent songs.

Re: The Facebook Algorithm Mom Problem

#124

Post by avar » Wed, Jul 12, 2017, 6:23 PM UTC

ryanbrunner wrote: post #95

I find a lot of sites feel like they're overtuning their recommendation engines, to the detriment of using the site. YouTube is particularly bad for this - given the years of history and somewhat regular viewing of the site, I feel like it should have a relatively good idea of what I'm interested in. Instead, the YouTube homepage seems myopically focused on the last 5-10 videos I watched.

This is because they're not trying to give you stuff you'd generally like given your entire history. The recommendation feature powers the part of YouTube that auto-plays the upcoming video.

So e.g. if I go and view Russian dashcam videos they're going to automatically play more of them, even though I've shown no prior interest in that topic.

Having two systems for recommendations would introduce a lot of UI complexity, so I can see why they didn't go for that, and why the recommendations are consequently tuned for people who are actively watching videos on some topic right now.

Re: The Facebook Algorithm Mom Problem

#125

Post by Florin_Andrei » Wed, Jul 12, 2017, 6:24 PM UTC

ryanbrunner wrote: post #95

I find a lot of sites feel like they're overtuning their recommendation engines, to the detriment of using the site. YouTube is particularly bad for this - given the years of history and somewhat regular viewing of the site, I feel like it should have a relatively good idea of what I'm interested in. Instead, the YouTube homepage seems myopically focused on the last 5-10 videos I watched.

It's a hard problem. If they kept recommending old topics, you'd be like "hey, I quit watching pink zebra videos 5 years ago, stop recommending that stuff to me".

Re: The Facebook Algorithm Mom Problem

#126

Post by aeturnum » Wed, Jul 12, 2017, 6:29 PM UTC

emodendroket wrote: post #19

aeturnum wrote: post #12

I'm pretty sure this description is wrong. My impression is facebook shows your content to a subset of friends and then classifies it based on likes received. If your mom 'like's 9/10 posts and your other friends like 3/10 posts, then 60% of your posts /are/ family content. Even if they're about mathematical theories.

It seems like what he is claiming is that his mom instantly clicks like on everything and so the post gets gated before his other friends ever have the opportunity to indicate whether they like it.

A lot of this seems like it comes down to how that initial pool of users is treated. If they show it to 5 users and his mom is the only one (ever) that likes it, then it makes sense to show that content to people common to him and his mom. If the other users like it, they should distribute it more widely and maybe include users from other non-overlapping networks. That all seems good. If mom likes 90% of the posts, then show that 90% to the family and maybe they care and maybe they don't. If it gets engagement from other people, share it more widely. That seems like a good approach.

The wrong way to do it would be to show it to 5 people and as soon as his mom likes the content, decide the content won't be interesting to the other 4 people they showed it to. I haven't seen any evidence of that happening, and I don't think the article points to any (anecdotal or otherwise).

Re: The Facebook Algorithm Mom Problem

#127

Post by idyllei » Wed, Jul 12, 2017, 6:31 PM UTC

avar wrote: post #124

ryanbrunner wrote: post #95

I find a lot of sites feel like they're overtuning their recommendation engines, to the detriment of using the site. YouTube is particularly bad for this - given the years of history and somewhat regular viewing of the site, I feel like it should have a relatively good idea of what I'm interested in. Instead, the YouTube homepage seems myopically focused on the last 5-10 videos I watched.

This is because they're not trying to give you stuff you'd generally like given your entire history. The recommendation feature powers the part of YouTube that auto-plays the upcoming video. So e.g. if I go and view Russian dashcam videos they're going to automatically play more of them, even though I've shown no prior interest in that topic. Having two systems for recommendations would introduce a lot of UI complexi…

I imagine it would be possible (though difficult) to autoplay videos which are related to each other and gradually converge to something that would interest the viewer.

For example, Russian dash cams to Russia at night to the Russian sleep experiment creepypasta, to horror games to video games in general, if that is what the user tends to watch.

I know this has graph theory written all over it and the shortest-distance problem has wreaked havoc for centuries, but I think with enough resources Google/YouTube could find a good compromise in this situation.

Re: The Facebook Algorithm Mom Problem

#128

Post by crowbahr » Wed, Jul 12, 2017, 6:31 PM UTC

Semaphor wrote: post #101

ryanbrunner wrote: post #95

I find a lot of sites feel like they're overtuning their recommendation engines, to the detriment of using the site. YouTube is particularly bad for this - given the years of history and somewhat regular viewing of the site, I feel like it should have a relatively good idea of what I'm interested in. Instead, the YouTube homepage seems myopically focused on the last 5-10 videos I watched.

I'm not sure if it is because of weird tastes or something else. But for sites like Steam, YT, Netflix I always wish for more tuning parameters because their recommendations are all horrible. On Netflix, for example, I get recommended stuff that's similar to other things I didn't like. Amazon is often the best of them for books because their engine seems to only really value the last few things I looked at / bought a…

Netflix, I think, has killed their own recommendation algorithm when they removed stars and made it boolean. I don't know if those buttons even do anything anymore because I think they're just matching based off demographic and who they're trying to market to now. It's recommending shows that I never would watch in a million years and giving them high matches despite me disliking most similar shows just because I'm a 20-something male.

Re: The Facebook Algorithm Mom Problem

#129

Post by jakub_g » Wed, Jul 12, 2017, 6:34 PM UTC

mrleinad wrote: post #14

I'd like Facebook to have an option to see all posts, without filtering, just as they're posted. It's not hard, it's a simple UX, but it's just not there.

I kind of understand the point on not showing everything: too much people producing too much content (and it's hard to decide whom to unfollow when everyone produces more or less the same percentage of quality and non-quality content), which led me few years ago to immediately abandon Twitter after just a few days. I followed a handful of people and have been just catching up with yesterday's tweets whole morning each day. This seems to be less of a problem those days but it might be subjective.

What makes me go nuts though is stuff disappearing from my FB wall (using mbasic web). Lately I see that a lot: see a few interesting entries on timeline, I click one, come back to the wall and they've disappeared and some random stuff from 3 days ago took their place and I can't find them again.

Same happens in YouTube Android app. See a few interesting videos, click one (I can't "open in new tab" in the app...), go back, they're gone and replaced with something totally unrelated. Ugh.

Re: The Facebook Algorithm Mom Problem

#130

Post by duskwuff » Wed, Jul 12, 2017, 6:35 PM UTC

anigbrowl wrote: post #112

ryanbrunner wrote: post #95

I find a lot of sites feel like they're overtuning their recommendation engines, to the detriment of using the site. YouTube is particularly bad for this - given the years of history and somewhat regular viewing of the site, I feel like it should have a relatively good idea of what I'm interested in. Instead, the YouTube homepage seems myopically focused on the last 5-10 videos I watched.

YouTube is weird. I made the mistake of watching a flat earth video last year in order to get a handle on that growing weirdo fad. Well it turns out that flat earthers are really avid content consumers (presumably because of the enormous cognitive reinforcement requirements to maintain such a belief) and YouTube really wanted to help me with that. It took a couple of months for it to stop offering me a portal to a be…

For future reference, you can remove videos from your YouTube history if you're logged in:

https://www.youtube.com/feed/history

Once a video is removed from your history, you'll no longer get recommendations based on it.

If you aren't logged in, you can't view or edit your viewing history, but you can at least clear it. (Which will return you to YouTube's terrible default viewer profile... sigh)