K-Means Clustering Explanation and Visualization

[@johnmooney658]

Really nicely illustrated, concise explanation. Well done!

[@TheDataPost]

Thank you!

[@charleslueker2597]

Wow, best explanation Ive seen

[@mueez.mp4]

Amazing what visualizations can do for explanations

[@sonydominates]

Wow, short and to the point. Thank you!

[@viralgamegamer6159]

Best explanation I have seen soooo farr 👏

[@rgss-eb2xc]

Concise and Clear!

[@hulaalol]

best showcase that explaining this algorithm can be done in 3:28 minutes. Other videos here on youtube are 20+ minutes lol

[@TheDataPost]

Thanks so much! Happy it helps.

[@hussamcheema]

Simplest and to the point explanation
Thank You

[@partyonpython5663]

Initial centroids are based on points already in the dataset, not selecting them randomly like he did in the beginning.

[@mnicp01]

Wow! Fantastic explanation! thank you!

[@chillyspoon]

Excellent explanation

[@erythsea]

Amazing video needed. this for my data mining course

[@lucianoinso]

Beautiful and to the point, thanks!

[@lautaa33]

Best explanation ever

[@MohamedHany-oq7xr]

That's a clear explanation, thanks alot

[@siyabongahlophe8316]

Best explanation!

[@JonJones-pz9kb]

good explanation

[@mohamedshatarah7264]

Amazing video. Thx a lot

[@juststudying1019]

straight to the point, thanks.

[@yousufazad6914]

freaking awesome video man! Thumbs up

[@kosstath6761]

@TheDataPost Which tool/software do you use for the visualization?

[@iKicchi]

great explanation..thanks

[@amal3227]

short and clear.Thank you

[@Jawi43]

thanks really short and to the point!

[@flowey7466]

Thank you, really good visualisation

[@AmNotLegend]

does it mean it has to visulise the points, and then may select initial centroids?

[@gauss8134]

Very good video

[@hamedabdollahpour5413]

Very clear. Thank you. But to me it's not clear how we identify number of clusters needed in a real case

[@bignigeful]

Now i understand. Thanks

[@afifkhaja]

Thanks for this! How to use the number of clusters if you don't know beforehand?

[@anisjayaram]

A gaussian mixed model can be used to estimate the number of clusters

[@anisjayaram]

Or you can use agglomerative clustering where the number of clusters will be equal to the number of observations

[@luisalfonsoprudenciobarrie1741]

You can use methods such as "The Elbow Method" to estimate the correct number of clusters for each dataset. What it does is it gets the WSS for each cluster and it selects the number of cluster where the WSS presents diminishing returns. But can you always find more info online. Good luck!

[@reunguju7501]

thanks! very good illiustration

[@leihejun844]

And this was how AI took the world.

[@thecalculatedcreativecoder1428]

This is a very basic algorithm. I don't this is the world ending AI we need to be concerned about. Now, reinforcement learning networks are potentially dangerous.

[@fanfoire]

Vey clear, thank you

[@tomererez6980]

very well done

[@oraclesql]

great vid. Thank you

[@lucasmoratoaraujo8433]

very nice! thank you!

[@sandyalmansour5255]

Great!

[@ulqsch7112]

great vid

[@Ali-qm8ix]

amazing

[@ittohere]

beaufitul

[@TheBlueFluidBreathe]

Thanks

[@TheMrCalindo]

it's perfect

[@zes3813]

wrr

[@nullprophet10]

great

[@SaulArcans]

I love yoy, thanks

[@henriquebigodinho9600]

you could use more of the actual terminology, like fit and predict phases, but overall congrats very weel and concise video

[@hannav7125]

I love you

[@SansGetsumei]

short and clear.Thank you

[@Lucky-rf5db]

thanks