[S1E3] Understanding K Nearest Neighbors Algorithms | 5 Minutes With Ingo

Показать описание

This week, Ingo Mierswa, RapidMiner's CEO, Co-Founder & Data Scientist in Residence, explains how the K-nearest-neighbors algorithm is used to formulate ideas by comparing points in a data-space and using the most similar data points as guidelines for predictions.

Plus, Ingo asks Data Scientist Number 7 to clean up Glen's mess, Doc Brown makes another split-second appearance and we hear the brief chuckle of a Python wearing a tuxedo.

DISCLAIMER: No data scientists were harmed during the filming of this episode. In fact, they were too busy doing amazing things with RapidMiner.

MUSIC CREDITS: Evil Thoughts, The Oscillator, James Taylor Quartet, Real Self Records, 2000

Рекомендации по теме

Комментарии

I've read elsewhere to learn more about kNN and this was by far the quickest and most simple explanation I have ever seen. Perfect guys, thank you!

domgilberto

Thanks, I'm studying k-NN at the moment. I like how the internet is giving us, the public, unprecedented access to the people who run the companies that we would never have had before. :)

RichardEricCollins

Hey Ingo, these series of videos are amazing. Seriously.
Best wishes from Brazil.

AB-ootl

great! i can easily imagine how the algorithm works, thanks!

andreanmaulana

that is an awsome explanation of KNN, thanks guys keep it up

MohamedEmad-hpxu

Great and interesting way to convey information

marciacamilleri

this really helped, i got a test in a two hours, i should do fine cause of you

yahbabay

interesting way to learn ML, good! Subscribed!

haikunHuanghk

Great video! However, I would like to see it from a mathematical perspective.

uendertandrade

Actually though a harder glass is more likely to break than a soft one.

lplplplplp

Great video..
I have a problem...
Actually I already asked in rapidminer forum, but no one has given an answer yet..

I can't find a satisfying answer for KNN-algorithm with same euclidean distance in rapidminer..

say k=5. Now I try to classify an unknown object by getting its 5 nearest neighbours. What to do, if distance is a lot of the same distance.. if after determining the 4 nearest neighbors, the next 2 (or more) nearest objects have the same distance and diferent label? Which object of these 2 or more rapidminer chosen as the 5th nearest neighbor?

I confused.. I try in excel, and the result is diferent with rapidminer for some data. in excel the result label is "LU":

but the result in rapidminer is "LT" :

result rapidminer weighted vote is checked is "LU" :

How rapidminer work with case like that...

how rapidminer sorting the distance ?...

something wrong with my data ?, or rapidminer sorting random if distance is same ?

thanks in advance for your help

virtualprogrammer-nasa