- </> - Follow us on #Twitter: @Teamadmp -</>-
STARTING FROM THE BOTTOM NOW WE’RE HERE ?? ..
With 850 millions of downloads in 2020, Tik-Tok, will be THE APP, of our post-pandemic era and will change the game for ever, when it comes to attraction, attention and engagment with its users globally!
Tik-Tok will also change the game by being one of the first app, whose USP (Unique Selling Point) is its awesome algorithm
Tik-Tok also changes the Chinese game, where Tencent Wechat, the other social media Chinese giants (Part of the BATS group made of Baidu, AliBaBa, Tencent) failed famously to ‘conquer’ the globe
—> Zhang Yiming, the founder and CEO of ByteDance/Douyin (Chinese Tik-Tok) /Tik-Tok, was clever enough to buy #Musical.ly for $1 Billion in 2017, before attacking global markets with Tik-Tok.
What enables TikTok to stand out from other social media platforms using recommendation algorithms such as #Twitter, Netflix, and YouTube are that Tik-Tok (TT) will entirely anticipate and decide what the user will watch next, instead of presenting a list of recommendations, based on their historic of viewings on the platform, or ‘Top Trending content’, to users, as #Twitter, #Netflix and others, would do.
I. What does the #TIK-TOK A.I., actually ‘KUNG-FU’… so well ?
CONSUMERS (audience) side: Tik-Tok (TT) uses A.I. to calculate what type of content will best fit the consumer needs and wants. —> When TT consumers use the app, TT consumers don’t really need to enter their preference, beyond their GEOLOCATION, tastes, age group and Email address.
The TT’s A.I. will quickly benchmark it against its sizeable database of #Bytedance (TT first iteration, launched globally in 2015) consumers’ behaviors.
PRODUCERS (content creators) side: #Tik-Tok (TT) will help content creators build and edit videos, with music, filters and other special FX (Special Effects).
CHECK THE VIDEO BELOW :
.Tik Tok takes advantage of AI not only for the production of videos but also for their delivery.
As opposed to simply delivering recommendations, the A.I. Algorithms are immediately prompted into interpreting their user’s behavior and delivering new content, solely based on these interpretations.
READ Harvard University (USA) survey below, for more insights, please:
As a result, In a short span of time, TT can make spot-on predictions regarding which videos will catch a user’s interest. This, in turn, drives UP, user engagement.
II. LET’s SEE HOW?
Basically, TikTok’s full-screen design allows every video to understand both positive and negative signals from its users.
The positive signal is understood from:
The user liking, following or watching the video till the end.
Eye-tracking device
Wrist-action from the user, and smartphone overall ‘position’ when watching
Swipe action, from user, to move from one video to the next
The Negative Signal is understood from:
Eye-tracking and wrist-recording action, from the user
The speed at which viewers swipe away or press down on video.
As quoted by Connie Chan, a renowned Chinese investor :
“TikTok is the first mainstream consumer app where artificial intelligence IS the product. It’s representative of a broader shift.”
Here’s how complex Artificial Intelligence systems and Machine Learning systems can be explained, with the TREE-GRAPH below :
Aggregating global users statistics, enable the production of this tree, which will be constantly optimized throughout the users life cycle. —> The suggestions from the Tik-Tok algorithm will get better and better, over a very short period of time !!
III. HOW DOES IT WORK WITH ANALYTICS?
Using our knowledge of BIG DATA management and A.I. processing systems, 4 types of analytics will help shape the Tik-Tok model, in order to work so efficiently :
TT will then use ‘Unsupervised learning’ and ‘Reinforcement learning’, as seen below, to optimize its suggestion’s model for its highly ‘predictive Algorithm’ :
End.
#TEAMADMP
#ADMP Digital #METAVERSE TEST BELOW :
END. END.
*^* THE AWESOMENESS OF TIK-TOK'S A.I. EXPLAINED HERE !! *^*