TikTok feeds are based on a recommendation algorithm that employs several tools and factors to personalize it for each user, similar to how many other social media platforms function. Now, the developers behind the video-sharing app have published a new blog post which details the way the recommendation feed works, and it also has tips for personalizing the feed in order to avoid being suggested random videos you might not like.
TikTok’s recommendation system is created around input factors in a similar way to the one YouTube uses to measure and monitor engagement. The way people interact with the platform impacts the recommendations offered, including posting a comment or following an account.
TikTok’s Impressive Algorithm
For instance, if a user only follows food accounts, and only likes or comments on videos about food, TikTok will offer them more food content. This also helps the platform’s algorithm know about videos people might not be interested in.
User interactions are also considered in this process. TikTok says that video information, which ‘might include details like captions, sounds, and hashtags,’ and device or account settings are also affecting the feed. For example, language preference, country setting, and device type will also factor in to ‘make sure the system is optimized for performance.’
Everything amounts to engagement, again. If someone finishes a video rather than flipping to the next one halfway through, this is registered as an interest. TikTok is often acclaimed for its recommendation system, but its algorithm still has its own vulnerabilities the company has also mentioned in the post.
“One of the inherent challenges with recommendation engines is that they can inadvertently limit your experience — what is sometimes referred to as a ‘filter bubble,’’ the post reads. “By optimizing for personalization and relevance, there is a risk of presenting an increasingly homogenous stream of videos. This is a concern we take seriously as we maintain our recommendation system.”
Behind the ‘For You’ Feed
This means that the app might not suggest certain videos if a user doesn’t particularly tune the system in that direction. The company’s post has also addressed the filter bubble by explaining its aim of obstructing repetitive content.
The ‘For You’ feed usually won’t display two videos in a row ‘made with the same sound or by the same creator,’ the post reads. The idea is that more kinds of content will appear in a feed than ones that are rather similar. However, this doesn’t work all the time. The way TikTok chooses which videos to suggest for every personalized feed is still unclear, but it is a part the company is at least recognizing as needing improvement.
TikTok was also criticized for the fact that it marginalized groups for not recommending content. This is an issue YouTube also regularly faces, but at least TikTok admitted it had hidden content from some creators, trying to make it a short-form solution to bullying.
“Early on, in response to an increase in bullying on the app, we implemented a blunt and temporary policy,” a spokesperson said in December 2019. “While the intention was good, the approach was wrong, and we have long since changed the earlier policy in favor of more nuanced anti-bullying policies and in-app protections.”
The TikTok blog has more information on how to personalize your own ‘For You’ page, which can be of great use. Overall, TikTok’s algorithm is one of the best elements of its worldwide success.
Paula is an outstanding reporter for Henri Le Chat Noir, always finding new and interesting topics to bring to the portal. She mostly crafts Science and Technology news articles, covering everything one needs to know about those niches. Paula studied at Concordia University.