how data scientists at tech companies make predictions about you

[bri_xu]

somewhere there’s a data scientist at YouTube calculating how many girls you talk to

[bri_xu]

Just got married so I will only talk to one girl for the next 50 years, that’s 0.02 girls a year

[Mileikowsky]

You can tell it’s real data science because he unnecessarily makes everything into a three letter acronym

[yuukichan]

Not L's theme to verify a lack of rizz 💀

[Faxtyper]

Hate that when I lie for goof and there an Asian dude summoned to correct me, happens all the time 😔

[_SinisterRose_]

Homie really took "no bitches" to a new level

[AidenDaly-vv]

There is not enough data to come to a conclusion because we only see 30-180 second long videos of his interactions. Let's assume most of these videos are 90 seconds. When you multiply 90 by 96 you get 8640. 8640 divided by 60 equals 144 and 144 divided by 60 equals 2.4. that gives us 2.4 hours of his interactions, which is 0.006 percent of his year.

[aunixattire]

hell nah, i aint even meeting 1000 people in 80 years

[samp_]

Bro politely said" you have no rizz"

[vishawashkumar]

Data scientist here, we never deal with exact zero. So whenever some data class comes out zero, assuming the data source is huge, we perform "smoothing". One of the simplest form of smoothing is Laplace smoothing. Where you add 1 to the class with zero value as well as to the total population. And calculate off of that.

[ralexcraft]

Homie forgot he was using a convenience sample.

[robindehood]

Great insights and visualisation. You're good at this.

[WAVE]

Einstein et al. concludes within their yearly data analysis that you have approximately zero maidens

[janus]

an THEN we start getting confidence intervals and stuff starts getting CRAZY

[vincent_hall]

Wow!
This is some hardcore, well researched, studied in depth Data Science!

[Sabert]

WRONG! Us data scientists don’t talk to anybody, we use AI to mimic human interaction.

(Please help me it’s become sentient and is holding me hostage)

We love AI in the data science community!

(I’m begging you he’s threatening to put me in a while loop without a break)

[oanling]

Most informative "No bitches ?"

[personpersonson]

This is a good display of why sample size matters

[jams_toast]

This is actually a very good example of problems with using small amounts of data especially when it’s stretched for big numbers

[zaynalee]

Bro talks while he sleeps rare talent 😂