TikTok reveals details of how its algorithm works • Hacker News Board

Re: TikTok reveals details of how its algorithm works

#21

Post by jakear » Thu, Sep 10, 2020, 2:22 PM UTC

xnx wrote: post #18

gcmac wrote: post #15

Earlier quoted context omitted.

I’d say it’s more likely they have super advanced/clever ways of doing the latter. The algorithm could be a simple dot product and the result could be great or terrible depending on how good the feature extraction is. Pulling useful features out of videos is no small task. The fact that everyone raves about how good the recommendations are indicates to me that this is where their innovation lies.

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.

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.

Re: TikTok reveals details of how its algorithm works

#22

Post by fqye » Thu, Sep 10, 2020, 2:24 PM UTC

Kednicma wrote: post #2

Here's the secret sauce: "Using machine learning, the algorithm serves videos to users based on their proximity to other clusters of users and content that they like." It's impressive how they arranged to be transparent for everything else, but kept this important part opaque. Rumor is that the actual secret sauce here is human curation; people hand-select videos with high appeal and label them "viral", "popular", et…

It is not possible to have a very large pool of curators that could 'hand-select' the 'right' content that could hook up people.

Tictok and Toutiao in China often push irrelevant content or new content by new creator to users to give them potentially interesting things outside their usual sweet spot and also to give new creators chances to win followers. They are exceptionally good at this.

Re: TikTok reveals details of how its algorithm works

#23

Post by danmg » Thu, Sep 10, 2020, 2:29 PM UTC

How TikTok's "algorithm" works is pretty obvious if you've spent any time on there.

It learns what you like by how much of a video you watch and how you interact with it, and it establishes some kind of weighted feature vector based on hashtags used in the description, words used in the comments, text drawn on the video, audio background, possibly some audio transcription of words said, if you commented on it, if you forwarded it to a friend, and so on. There may be some network based recommendations, based on who you follow but those seem to be weighted very weakly, and that makes sense if you're trying to keep the platform from getting botted.

It also seems to do some non-dominant sorting to keep from only showing you things from the same type of video.

Facebook's competing short video service is terrible in comparison. It only wants to show me Trevor Noah clips, and things tangentially related to things I may have "Liked" 12 years ago.

Re: TikTok reveals details of how its algorithm works

#24

Post by albertshin » Thu, Sep 10, 2020, 2:29 PM UTC

simion314 wrote: post #16

Kednicma wrote: post #2

Here's the secret sauce: "Using machine learning, the algorithm serves videos to users based on their proximity to other clusters of users and content that they like." It's impressive how they arranged to be transparent for everything else, but kept this important part opaque. Rumor is that the actual secret sauce here is human curation; people hand-select videos with high appeal and label them "viral", "popular", et…

It mentions that it also serves you from time to time a video from a different cluster and analyzes your reaction so you are not stuck in one cluster without hope to get out.

feel like dating apps do something similar...

but I wonder why I end up in YouTube holes with bottomless pits though until I hit the reset button.

Re: TikTok reveals details of how its algorithm works

#25

Post by rootsudo » Thu, Sep 10, 2020, 2:30 PM UTC

prionassembly wrote: post #5

rootsudo wrote: post #3

Heh, "We're a 2-year-old company operating with the expectations of a 10-year-old company," said Michael Beckerman, TikTok's vice president in charge of U.S. public policy. "We didn't have the opportunity to grow up in the golden years of the internet, when tech companies could do no wrong. We grew up in the techlash age ,where there's a lot of skepticism of platforms, how they moderate content and how their algorith…

I think there's a part in your comment that you thought but somehow neglected to type out. It reads to me like two completely disconnected parts.

I meant by copy/pasting that quote that TikTok's excuse of "we're just 2yrs old" is not applicable because the Tech world is entering an era of regulation, which put TikTok on the front page of whom to target first, same as any other business that enters the marketplace. Using Juul as a reference to the regulation that they have to abide by (now) and how they innovated to become known/verb.

Though, you can very much argue that Juul is both Tech & Tobacco. But, imagine the same statement from Juul.

Re: TikTok reveals details of how its algorithm works

#26

Post by judge2020 » Thu, Sep 10, 2020, 2:32 PM UTC

xnx wrote: post #9

ChefboyOG wrote: post #4

I don't see any details about its algorithm, just that they use a recommendation engine? Things like this aren't proprietary info, they're just how recommendation engines work: "Once TikTok collects enough data about the user, the app is able to map a user's preferences in relation to similar users and group them into "clusters." Simultaneously, it also groups videos into "clusters" based on similar themes, like "bas…

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…

> Did you click through to the profile? Did you view other videos from the profile?

Probably why multi-part videos are so popular (you have to click their profile then find the part 2 video to finish the story they were telling).

Re: TikTok reveals details of how its algorithm works

#27

Post by FirstLvR » Thu, Sep 10, 2020, 2:33 PM UTC

so... wheres the code?

Re: TikTok reveals details of how its algorithm works

#28

Post by Firebrand » Thu, Sep 10, 2020, 2:34 PM UTC

jakear wrote: post #8

It's a (seemingly) pretty simple matter of keeping track of how long someone looks at a video, and optimizing to show them videos they'll spend a long time looking at. It's pretty powerful too, I spend much more time looking at TikTok than any other digital content aggregator (to the extent I had to delete it from my phone). They also do some things like sprinkle in random fresh videos, potentially unrelated to your…

I've been seeing a lot of low to mid-tier creators dropping out of the payment program lately after their view counts plummeted. It's speculated that TikTok only pays for video completion count while enrolled in the Creator Fund since view counts tend to go back up after leaving.

Having more leverage with third parties outweighs the tens of dollars they'd receive from a video, I guess.

Re: TikTok reveals details of how its algorithm works

#29

Post by ChefboyOG » Thu, Sep 10, 2020, 2:34 PM UTC

xnx wrote: post #9

ChefboyOG wrote: post #4

I don't see any details about its algorithm, just that they use a recommendation engine? Things like this aren't proprietary info, they're just how recommendation engines work: "Once TikTok collects enough data about the user, the app is able to map a user's preferences in relation to similar users and group them into "clusters." Simultaneously, it also groups videos into "clusters" based on similar themes, like "bas…

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.

Re: TikTok reveals details of how its algorithm works

#30

Post by chaotic_mind » Thu, Sep 10, 2020, 2:35 PM UTC

To be fair, the algorithm they told is mostly what we would have perceived, knowing if that is Machine Learning driven.

Although, I am quite intrigued what they would be showcasing in their "transparency centers". If they show what extent their data can be utilized. and not simply a simulation of the ML stuff.