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

TikTok reveals details of how its algorithm works

axios.com

Post reply on HN
Print view View original

151–160 of 190 posts

Re: TikTok reveals details of how its algorithm works

#151

Post by thekyle »

novok wrote: post #108
jakear wrote: post #21

Earlier quoted context omitted.

They could also be digging only into audio, doing speech recognition on it, then clustering the text. Augment that with the text users have put into the video directly using the in-app editor and you have some pretty solid data.

I would also start doing image recognition in the video frames, to extract things like gender, objects, etc.

Would this have any advantage over just using video embeddings (or a sequence of frame embeddings?) which in theory should capture those things in vectorized form.

Re: TikTok reveals details of how its algorithm works

#152

Post by ZephyrBlu »

johnchristopher wrote: post #88
undefined1 wrote: post #73

TikTok is like a highly compressed version of YouTube. YouTube incentivizes creators to make artificially long videos, resulting in a huge amount of filler. So you get a lot of videos where you can skip the first 20%+ and not miss anything. The content is buried and spread out. 1 minute of content surrounded by 9 minutes of filler. TikTok removed those 9 minutes, so it feels very refreshing in that respect. There's n…

> TikTok removed those 9 minutes, so it feels very refreshing in that respect. There's no incentive to create filler - just the opposite. It's to a point that I skip any TikTok video that starts like “I was in the forest and found this mushroom. Did you know that this kind of mushroom can be used to... ”. I skip it as soon as I hear “I was in the“ (that and the usual tone that comes with it). I don't care. Show me di…

Is reinforcing that behaviour not worrying to you?

Being able to get consistent, instant stimulation from doing effectively nothing seems dangerous to me.

We're training ourselves to not be able to focus on anything.

Re: TikTok reveals details of how its algorithm works

#153

Post by thekyle »

Drew_ wrote: post #76
ramraj07 wrote: post #55

Earlier quoted context omitted.

I don't buy this argument. YouTube ostensibly has a metric ton of information like this and even if tiktok has more training data now, I'm fairly sure their training data in principle was smaller than what YouTube historically had given their decades long presence and their ubiquity in the internet. This is on top of the documented effort by YouTube to perfect their recommendation algorithm using the best ML minds th…

YouTube's recommendations are arguably just as good as TikTok's in my opinion. The only difference is that YouTube places the burden of choice on you while TikTok makes every choice for you. If you don't quickly find something you'd like to watch on YouTube you're very likely to leave and find something else to entertain you. Meanwhile TikTok is automatically feeding you an infinite source of quick and easy to digest…

I also agree that YouTube's recommendations are pretty good. For example, I like to watch Linux distro reviews which are fairly niche and usually only get a few hundred views but my YouTube homepage is filled with them. In contrast, when I was testing out TikTok it mostly recommended me things with broad appeal like comedy skits and generic science videos that I think 90%+ of the population would find at least somewhat enjoyable.

Re: TikTok reveals details of how its algorithm works

#154

Post by YetAnotherNick »

SomaticPirate wrote: post #81

Does anyone know of an open source recommendation engine? I would love to just explore how one works bet it seems like this solely in the realm of proprietary software at the moment. I have a dream of naively making a open source recommendation engine using data from good reads and imdb. There is a real utility here that I worry will be captured entirely by companies without an open source alternative.

Facebook's dlrm is good one. There's libffm, xlearn etc. I found dlrm to be good choice between extensible and fast.

Re: TikTok reveals details of how its algorithm works

#155

Post by johnchristopher »

ZephyrBlu wrote: post #152
johnchristopher wrote: post #88

Earlier quoted context omitted.

> TikTok removed those 9 minutes, so it feels very refreshing in that respect. There's no incentive to create filler - just the opposite. It's to a point that I skip any TikTok video that starts like “I was in the forest and found this mushroom. Did you know that this kind of mushroom can be used to... ”. I skip it as soon as I hear “I was in the“ (that and the usual tone that comes with it). I don't care. Show me di…

Is reinforcing that behaviour not worrying to you? Being able to get consistent, instant stimulation from doing effectively nothing seems dangerous to me. We're training ourselves to not be able to focus on anything.

I don't really think so (at least regarding my tiktok usage)

Being thrown right into the mushroom thing, into the action, forces me to keep focus and attention. The pseudo-introduction is just dulling my mind while not bringing anything of consistence.

I had some tiktoks that started like "So I read something incredible the other day and I just had to try it. I am gonna show you what happened when I decided to eat an orange under the shower". Don't care. Starts with you eating the orange.

As another sibling commentator wrote it's a different narrative structure.

> Being able to get consistent, instant stimulation from doing effectively nothing seems dangerous to me.

The same could be said from books but I get what you mean. I still read books, so I guess I am okay (it could be worse !).

edit: to be fair, I have very low tolerance and patience for people who beats around the bush :/. On the other hand, when walking outside I can stop and look around me for 2 or 3 minutes every 100 meters because there's always some piece of architecture or natural landscape to look at.

Re: TikTok reveals details of how its algorithm works

#156

Post by johnchristopher »

novok wrote: post #107
dirtyid wrote: post #92

Earlier quoted context omitted.

Yeah I wish TikTok has variable speed / fast forward option. Another factor for ML to train and automatically speed up to the juicy bits. There's a lot of room to waste even in 1 minute formats. Though TikTok itself is mostly time wasting and this remark is more an indictment of my attention span.

Is it possible to scrub videos / skip around in tiktok? I've tried googling it but to no avail.

Pretty sure you can't.

Re: TikTok reveals details of how its algorithm works

#157

Post by xnx »

wombatmobile wrote: post #44
xnx wrote: post #18

Earlier quoted context omitted.

There's so much good meta-data (likes, comments, duration, sound used, views, like/view ratio, skips, loops, subscribes, etc.) that I'd be surprised if they were digging into the contents of the video at all right now.

> I'd be surprised if they were digging into the contents of the video at all right now. Why would you be surprised to learn TikTok is doing video content analysis?

Google was able to build a very useful search engine that ran for decades relying on the significance of links and keywords, without much understanding of the meaning of page content. You can get very far with the readily available data, before you need to delve into the fancy stuff to make it a few percent better.

Re: TikTok reveals details of how its algorithm works

#159

Post by xnx »

ChefboyOG wrote: post #29
xnx wrote: post #9

Earlier quoted context omitted.

More data beats smarter algorithms any day, and TikTok gets a lot of data because its videos are so short and interaction rate so high. There are tons of signals it can use as inputs: How much of a video did you let play before swiping next? How many times did you let the video loop? Did you like? Did you comment? Did you like a comment? Did you click through to the profile? Did you view other videos from the profile…

I never thought of it like this. Just had one of those mind exploding moments as I realized the frequency of interactions on TikTok vs other platforms and the virtuous cycle it creates (order of magnitude more data -> better recommendations -> even more interactions). Thanks for this.

Glad I was able to help introduce you to that idea. I was first introduced to it by Peter Norvig (Google): https://static.googleusercontent.com/media/research.google.c...

Re: TikTok reveals details of how its algorithm works

#160

Post by erklik »

jrockway wrote: post #82
undefined1 wrote: post #73

TikTok is like a highly compressed version of YouTube. YouTube incentivizes creators to make artificially long videos, resulting in a huge amount of filler. So you get a lot of videos where you can skip the first 20%+ and not miss anything. The content is buried and spread out. 1 minute of content surrounded by 9 minutes of filler. TikTok removed those 9 minutes, so it feels very refreshing in that respect. There's n…

I watch a lot of YouTube and haven't really noticed any artificially long videos in my bubble of channels. I agree that intros are annoying (I have no idea why people do them), but other than that... it's just the usual video stuff. The nature of the format is that a one page blog post becomes a 10 minute video. People talk slower than they read, so it's somewhat unavoidable. There are also YouTubers that are really…

Really? I feel like I could easily most of the videos I watch and still understand what the video is about.

The first few minutes are the intro, a message that the video is sponsored and we will find out about them later, last 3/4 minutes is always the "Like, Subscribe" and a sponsorship message usually. This means that both the front-and the back can be cut out easily without any affect on the actual content.

Post reply on HN