Mahalanobis Distance - intuitive understanding through graphs and tables

[@kidou123456]

MD is the standardization for multivariate sample. So insightful! Thank you!

[@saravanabalagi]

One of the best explanations for Mahalanobis distance...! Good one.. :)

[@ananthakrishnank3208]

Thank you, a lot. Mahala Nobis idea is cleared.

[@ArunKumar-yb2jn]

But multi-collinearity is not a problem in distance-based models like KNN. So why exactly is it an advantage with Mahalonobis?

[@maxinelyu7875]

This is very beginner-friendly!

[@wahabfiles6260]

I am also very friendly

[@PavanKumar-ef1yy]

Best Explanation Ever

[@SaraMohandesSamani-r6g]

Thank you. I learned only from your video.

[@wilsonneto9991]

An illustrative video, with awesome material! Thanks

[@shahannazim]

thanks alot for this very nice video

[@sadlenn]

Great explanations! Thank you from Canada

[@nachiketpethe]

very nice explanation

[@adanewoldeyesus7389]

interesting explanation.

[@saeedahmed9860]

It is excellently explained. Thumbs up

[@coopersing]

Thanks - I understand this better now! Some good illustrations in this.

[@souravde6116]

Lots of doubt.
Q1) If x is a vector defined by x = [x1;x2;x3...;xn], what will be size of covariance matrix C?
Q2) If x is a matrix of M-by-N dimension, where M is no. of the state vectors and N is the total no. of respective observations of each vector in a different time instant, then how to calculate Mahalanobis norm and what is its final size of D and what is the inference we can get from this metric?
Q3) If x is a matrix of N-by-N dimension, then also how to calculate Mahalanobis norm and what is its final size of D and what is the inference we can get from this metric?

[@jaybhanushali6448]

For all the questions you have asked you first need to observe the dimension of the Cov matrix
Dimension of the cov matrix = No of features/State vectors.
Q1 > [N dimenional for N features]
So for example lets say you are observing 3 features , then the cov matrix and it's inverse will be 3x3 matrix
According to your Q2 > M is the dimension of the Cov matrix
The nature of the Mahalonobis distance highly depends on the properties of Cov matrix.
To better understand the intuition you can look at it's use case in Bayesian Decision boundary:
www.byclb.com/TR/Tutorials/neural_networks/ch4_1.htm

[@snehashishpaul2740]

Nicely explained with a very good example.

[@queenofswordwthrone]

love intuitive understandings

[@tonyrandall3146]

Good visual representation especially..

[@skroutsit]

Very neat and insightful, thanks sir

[@svarki2]

Very good explanation.

[@richardgruss8171]

Very good explanation. Thank you!

[@tirac8016]

Really good video, it helps a lot! Thank you

[@ms_1918]

excellent video

[@thrisharamkumar9566]

Thanks for your explanation, helped a lot

[@dorsaghahri5085]

Very insightful! Thank you for posting.

[@jcthomas5593]

phenomenal video!

[@sayandey1478]

Wow, this is great, so intuituve

[@KayYesYouTuber]

great explanation. thank you very much

[@adipurnomo5683]

Clear explained

[@HumbertoFioravanteFerro]

A neat explanation. Thank you, Malakar!

[@ugururesin8405]

Simply best! Thank you for the video.

[@lekhaemerald]

Great video and visualization!

[@JefferyChiang-l3x]

nice explanation! Thank you

[@nv3796]

superb

[@shyamjain8379]

Thank you so much

[@asimkarel]

Amazing

[@muratcan__22]

perfect explanation on euclidean distance

[@godnkr236]

thanks for this amazing video!

[@thesturag]

Could you please calculate Mahalanobis distance in Microsoft excel?

[@Xuneme]

A very nice way of explaination!

[@johnnypark8622]

it is great and easy to understand! thanks a lot.

[@roysamson13]

Such a great explanation Sir, Thank you.

[@BredaMattia]

Excellent explanation. Thanks

[@bokai5829]

Great, research project complete

[@buugedy]

I did not understand, what MD is. Too complex

[@julianrojas7929]

Thank you!

[@sylvansky]

AWESOME JOB!

[@riCoger]

Hard to listen to but good job anyway.

[@BenBen-ke8nn]

suggest you add a subtitle as hard to listen

[@priyadarshangupta]

Thanks.

[@robinreus275]

Snap dr geen bal van

[@mikesuri4210]

Thank you!