Recommendations are a part of almost any social media app you regularly use, behind them are algorithms that use tools and methods that give each user different results based on different factors. In a recent blogpost, TikTok has shared some secrets about its recommendation feed and how it works. The post also gives users different tips to customize their feeds in a way that they don't receive unwanted recommendations.
TikTok's recommendation algorithm works a lot like YouTube's and the type of content you engage in will affect your recommendations on the app. For example: if you mostly like and comment on content about space, you'll be suggested videos about space as well. Through this, TikTok's algorithm also gets to know about videos you might not be interested in and not recommend them to you.
That's not all, TikTok has also said that video captions, sounds, and hashtags also have an effect on your feed. Your device or account settings like language preference and country selection can also optimize what you see, however, TikTok says this data point has a lower weight as compared to other data points because they are not necessarily "preferences."
Engagement is the real deal here, if someone skips a video or does not finish it completely, this will indicate a lack of interest in a topic of that sort. The post also says that it is more about content than the content creator, this means that if a content creator suddenly starts posting content you don't like, they won't appear in your feed.
Content curation on TikTok is applaudable, one reason users are addicted to the app is that they see what they want to see on their feeds. But the recommendation algorithm still falls shorts in some areas which TikTok has stated.
“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 stated. “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.”
Let's say if you mostly engage in makeup related videos you might miss out on the important posts like those about Black Lives Matter protests and those regarding LBTQ if you don't do something to drive the algorithm in that direction. Sometimes, users might see videos with the same music based on the same trend thrice or more in a row. It is unsure how TikTok exactly decides which video to present on user feeds but at least the platform acknowledges that it needs improvement.
TikTok is also considerate about not showing dangerous content, it says that content involving gore or use of drugs and alcohol may be inappropriate to recommend to many audiences especially young kids. This is the reason why many content creators talk about being banned for some of their content or upload a certain video multiple times.
The platform is also criticized for not recommending content from the LGBTQ community. The same problem is faced by YouTube which is undergoing a lawsuit after many LGBTQ content creators have said that the site suppressed their videos. TikTok is working to improve that too.
TikTok's recommendation algorithm is by far one of the best out there and a good thing about the company is that it is open about its shortcomings and is working on them.
In the blog, TikTok has shared more on how users can optimize their feeds to get better recommendations that are aligned with their interests.