K Means ++ for Initialization || Lesson 106 || Machine Learning || Learning Monkey ||

Thanks for the lecture
Please correct me if I am wrong for following things-
1)Incase of K Means Clustering, Centroids(Initial or any further) may or may not be from our original data points.
2)Incase of K Means ++, apart from 1st Centroid other k-1 Centroids(initial) are always selected from our original data points.

Thank you for your wonderful explanation of K-Means++

Really loved your explanation !

Very Good ! Explanation . Perfect Teaching!!!

Quite clear explanation of condition ....

nicely explained bro :)
thanks

Wonderfully explained thank you

Good one sir

wish have subtitle

Complete and simple explanation, thank you very much!

Hi..I have a doubt regarding the selection of the third centroid.whether we take max distance from centroid 1 or centroid 2..Thanks in Advance

He nails it perfectly! So easily explained. Thanks.

Good man

The Explanation is perfect, Thank you :)

Good morning sir,
what about outliers, becoz outliers have more probability of getting selected due to distance parameter which we have taken for finding probability

Thanks for sharing your knowledge I was triying to get this concept on various blogs but was getting congused.. You made it simple for me... Thanks to you

For step 3, instead of leaving it to probability, can we just select the data point that is furthest away from all the previously chosen centroids as the next centroid?

Really great explanation, right to the point of the algorithm. Thanks!

Thanks for the clear explanation.

Thank you vry much. You explained it very well. :)

thank you very much :)

thanku very much , crisp and clear explanantion

great explanation thank you

You made it very simple to understand, thanks !

made it crystel clear.

Amazing as always. Much appreciated man!

Great explanation. Keep up the great work.

Very clearr explanation!! Love your work man. Keep it up and appreciate it :)

thanks for the clear content...

Thanks alot!!🤩

Best explanation I've seen so far; so helpful!

helpful

Simple and great explanation!

Whenever I get confused about Kmeans++...I always watch this video