The "Tool" Behind TikTok's Addiction Allows This Trendy App To Read Users' Mind Only After Hours
TikTok is one of the most popular social networks today and it is no coincidence that many people consider it a new "drug".
Recently, information about these four goals has explained why TikTok has become an "addictive drug" for many people, especially young ones, according to the New York Times.
The document titled "TikTok Algo 101" was produced by TikTok's engineering team in Beijing. A spokesperson for the company, Hilary McQuaide, confirmed the authenticity and said the kit was written to explain to employees how the TikTok algorithm works.
The document provides an overview of core values and insights into the company's understanding of human emotions, such as sadness, happiness, anger, etc. This explains why TikTok users find it difficult to take their eyes off the screen.
It also helps to unravel the relationship between TikTok and its parent company, ByteDance, at a time when the US Department of Commerce is preparing a report on whether TikTok poses a security risk to the country.
This app has enjoyed quite a success while rival apps struggled, and this is part of the reason why TikTok monopolizes the market, easily gaining a strong foothold.
For the average user (not to mention content creators), TikTok can be a good place to track personal interests, from popular to sensitive topics. In particular, the TikTok algorithm helps users better understand their own desires. It is likened to an X-ray machine that clearly projects the inside of the user.
Analysts are also trying to figure out the key to TikTok's success. A recent report by The Wall Street Journal demonstrated how TikTok relies heavily on the amount of time you spend watching each video to direct you to multiple videos with similar content.
However, the process can sometimes be dangerous, especially for people with depression, who often view content related to suicide or self-harm, issues TikTok says it is working on to prevent.
The set of documents indicates that TikTok's ultimate goal is to increase daily active users, so they chose to optimize the two metrics "Retention " and "Time spent". "TikTok wants to keep you there as long as possible. The experience is sometimes described as an addiction," commented playwright David Mamet.
“The algorithm tries to get people addicted rather than giving them what they really want. I think it’s a crazy idea to let TikTok’s algorithm steer the life of our kids.
Each video a kid watches, TikTok gains a piece of information on him. In a few hours, the algorithm can detect his musical tastes, his physical attraction, if he’s depressed, if he might be into drugs, and many other sensitive information. There’s a high risk that some of this information will be used against him,” said Guillaume Chaslot, founder of Algo Transparency, a social networking research group based in Paris.
The document also reveals that watch time is not the only factor that concerns TikTok. The document lays out a basic equation for how videos are scored, where a prediction is driven by machine learning and actual user behavior is aggregated.
“The recommender system gives scores to all the videos based on this equation, and returns to users videos with the highest scores. For brevity, the equation shown in this doc is highly simplified. The actual equation in use is much more complicated, but the logic behind is the same,” according to the document content.
The document detailed how the company tweaked its systems to identify and block videos designed to fool the algorithm. In fact, the total value that users watch all videos from a content creator is higher than the value of viewing individual videos combined.
For example: If a user likes a certain type of video, but the app keeps promoting that type of video to him, he will quickly get bored and close the app. There are two solutions to this problem: "Make some assumptions, and break down the value into the value equation," according to the document.
Julian McAuley, a professor at the University of California San Diego, said algorithms are “totally reasonable, but traditional stuffs”. TikTok's advantage comes from combining machine learning technology with “fantastic volumes of data,” he said. “There seems to be some perception (by the media? or the public?) that they’ve cracked some magic code for recommendation, but most of what I’ve seen seems pretty normal.”
In fact, this document helps shed light on the kind of recommendation systems tech companies typically present. For example, leaked Facebook documents illustrate how Facebook takes comments seriously, and this has helped divisive content go viral. While the models can be complex, there's nothing secret or confusing about TikTok's recommendation algorithm outlined in the document.
H/T: The New York Times (nytimes.com)
Source: Getty Images
Recently, information about these four goals has explained why TikTok has become an "addictive drug" for many people, especially young ones, according to the New York Times.
The document titled "TikTok Algo 101" was produced by TikTok's engineering team in Beijing. A spokesperson for the company, Hilary McQuaide, confirmed the authenticity and said the kit was written to explain to employees how the TikTok algorithm works.
The document provides an overview of core values and insights into the company's understanding of human emotions, such as sadness, happiness, anger, etc. This explains why TikTok users find it difficult to take their eyes off the screen.
It also helps to unravel the relationship between TikTok and its parent company, ByteDance, at a time when the US Department of Commerce is preparing a report on whether TikTok poses a security risk to the country.
TikTok - "the new drugs"
Source: Shelly Palmer
This app has enjoyed quite a success while rival apps struggled, and this is part of the reason why TikTok monopolizes the market, easily gaining a strong foothold.
For the average user (not to mention content creators), TikTok can be a good place to track personal interests, from popular to sensitive topics. In particular, the TikTok algorithm helps users better understand their own desires. It is likened to an X-ray machine that clearly projects the inside of the user.
Analysts are also trying to figure out the key to TikTok's success. A recent report by The Wall Street Journal demonstrated how TikTok relies heavily on the amount of time you spend watching each video to direct you to multiple videos with similar content.
However, the process can sometimes be dangerous, especially for people with depression, who often view content related to suicide or self-harm, issues TikTok says it is working on to prevent.
The set of documents indicates that TikTok's ultimate goal is to increase daily active users, so they chose to optimize the two metrics "Retention " and "Time spent". "TikTok wants to keep you there as long as possible. The experience is sometimes described as an addiction," commented playwright David Mamet.
The secret algorithm that helps TikTok read users' minds
Source: Ad Scholars
“The algorithm tries to get people addicted rather than giving them what they really want. I think it’s a crazy idea to let TikTok’s algorithm steer the life of our kids.
Each video a kid watches, TikTok gains a piece of information on him. In a few hours, the algorithm can detect his musical tastes, his physical attraction, if he’s depressed, if he might be into drugs, and many other sensitive information. There’s a high risk that some of this information will be used against him,” said Guillaume Chaslot, founder of Algo Transparency, a social networking research group based in Paris.
The document also reveals that watch time is not the only factor that concerns TikTok. The document lays out a basic equation for how videos are scored, where a prediction is driven by machine learning and actual user behavior is aggregated.
“The recommender system gives scores to all the videos based on this equation, and returns to users videos with the highest scores. For brevity, the equation shown in this doc is highly simplified. The actual equation in use is much more complicated, but the logic behind is the same,” according to the document content.
The document detailed how the company tweaked its systems to identify and block videos designed to fool the algorithm. In fact, the total value that users watch all videos from a content creator is higher than the value of viewing individual videos combined.
For example: If a user likes a certain type of video, but the app keeps promoting that type of video to him, he will quickly get bored and close the app. There are two solutions to this problem: "Make some assumptions, and break down the value into the value equation," according to the document.
Julian McAuley, a professor at the University of California San Diego, said algorithms are “totally reasonable, but traditional stuffs”. TikTok's advantage comes from combining machine learning technology with “fantastic volumes of data,” he said. “There seems to be some perception (by the media? or the public?) that they’ve cracked some magic code for recommendation, but most of what I’ve seen seems pretty normal.”
In fact, this document helps shed light on the kind of recommendation systems tech companies typically present. For example, leaked Facebook documents illustrate how Facebook takes comments seriously, and this has helped divisive content go viral. While the models can be complex, there's nothing secret or confusing about TikTok's recommendation algorithm outlined in the document.
H/T: The New York Times (nytimes.com)
Share this article
Advertisement