SMOTE - Synthetic Minority Oversampling Technique

[@alkodjdjd]

Finally someone who speaks in an understandable english

[@chevasflorian4055]

Thanks for speaking slow. That's great for people who are not English speaking natives. In addition, your explanation is very clear. Thanks!

[@viswatejaforu]

I would strongly recommend this course in udemy for Machine learning, the way he explained all the topics is Cristal clear, one should understand it.

[@chd9841]

this is really nice technique...its so simple but no one thought about it for so many years!!

[@datadriven597]

Lucky to find this hidden gem. Keep up the good work!!

[@karkinissan]

That was absolutely amazing. Understood it completely. Thank you!

[@ayush612]

So well explained!! Nice methodology of teaching! Thank you

[@javieralfonso8702]

Very well explained Jitesh, thank you, I was having a hard time understanding SMOTE and in only 8 minutes I got it thans to you. :)

[@daesoolee1083]

This is an amazing technique for actual applications. Blowing my mind.

[@yassoaytac5747]

For a non-data scientist, those techniques are really hard to understand. Especially when you read i.e. the article SMOTE from Chawla,Bowyer et al. 2002... Thanks for this video. Now i got at least the idea of smote :)

[@NekoJimin]

Thank you, bless your soul!

[@janithinduwara9394]

Very clear sir !

[@sathyanarayanapoojari7664]

Nice explanation

[@brunoomenaribeiro3187]

That was beatiful, compliments from Brazil, Jitesh!

[@namanshah9216]

Explained in a very easy manner. thanks!!

[@ashishsangwan5925]

Nice theoretical explanation

[@nikitadetkov3900]

Actually, that was very cool, thanks for explanation! Good luck with future videos!

[@neelanjanmitra3379]

this is gold

[@coreygates9646]

Thank you so much! excellent explanation and visuals!

[@meysamasgarichenaghlu6530]

Thank you, you made my day, great video.

[@hardiksharma9157]

where do we use value of "the difference multiplied by random number"?

[@hansikamannapperuma8001]

Thank you so much

[@RaviTeja-zk4lb]

Can we apply smote on imbalance dataset which contains only categorical features which are mostly either 0 or 1

[@usamazahid1]

Perfect 🥰

[@AhmadHassan-on6zq]

Great

[@arashalizade9583]

Thank you

[@andrewyang6806]

Awesome intuition! Thank you!

[@juancarlosmendozasaavedra7613]

Excelent, thanks for talking slowly

[@learnds2210]

Thanks for great tutorial

[@aniamendoz4062]

@ Jitesh Khurkhuriya This was awesome. Even a noob like myself could understand :). However you mentioned there will be another video with a practical example that I am unable to find. Could you please help me find it. Do you also have other lessons on other techniques like ARIMA?

[@kkjitesh]

You can find it here kzbin.info/www/bejne/kJXJZZd-nbh_aZo

[@elsmith1237]

Thank you so much! Simple explanation

[@arindamnaskar2075]

Hi, thanks for the video. Just have one query: Is there any particular way to select the feature vector?

[@dantebutcher6315]

I misspelled "smite" and now I'm here

[@UrontoShahid]

Does SMOTE is applicable in Image Segmentation dataset?

[@javierlinkolklopezgonzales4274]

Thank you very much. Absolutely amazing. Do you have an email to write to?

[@bobthetalkwalker]

Thank you, Sir!

[@legttv97]

Step 1. Playback speed: 1.5

[@christopherkelman1563]

Perfect, Thank you

[@cooky123]

waiting for the implementation! thx

[@kkjitesh]

here it is kzbin.info/www/bejne/kJXJZZd-nbh_aZo

[@lindseyxiao5366]

Where is the implementation video ? I really need it ...

[@kkjitesh]

kzbin.info/www/bejne/kJXJZZd-nbh_aZo

[@vaseemmk]

Exellent!!!!

[@okby8600]

Good visuals

[@sarveswararaovangala4391]

as he is speaking slow, watch this video at 1.5X to get normal pace.

[@pixel7038]

Not everyone has the same learning pace dawg.

[@mohamedmahyoub5877]

1.75x even better lol

[@rishabhrohit]

Yes, for us 1.5 times works the best. But for those who are not from India, sometimes it gets harder for them to understand the accent so the normal speed is perfect for them.

[@ogsconnect1312]

Thanks

[@sinan_islam]

looks like new data is generated by interpolation

[@chd9841]

what are some other techniques for such imbalanced cases?

[@ashishsangwan5925]

www.analyticsvidhya.com/blog/2017/03/imbalanced-classification-problem/

[@hellochii1675]

4:54 `~~

[@yuhaooo8143]

may i ask why imbalanced data affect the result?? thx

[@Chato13]

What do you mean?

[@SpecificSpecificity]

lit af

[@adityajillellamudi4634]

play it at 1.5 speed

[@ogsconnect1312]

Thanks