Part 1-Decision Tree Classifier Indepth Intuition In Hindi| Krish Naik

[@krishnaikhindi]

We are happy to announce iNeuron is coming up with the 6 months Live Full Stack Data Analytics batch with job assistance and internship starting from 18th June 2022.The instructor of the course will be me and Sudhanshu. The course price is really affordable 4000rs Inr including GST. The course content will be available for lifetime along with prerecorded videos. You can check the course syllabus below Course link: courses.ineuron.ai/Full-Stack-Data-Analytics From my side you can avail addition 10% off by using Krish10 coupon code. Don't miss this opportunity and grab it before it's too late. Happy Learning!!

[@AmeliaMelia-tj3kc]

best teacher

[@harshdeepjaggi9715]

wonderful explanation sir.... I'm already enrolled in Data Science with one of the edtech of India... no doubt waha ke teachers bhi accha padhate par jo english mei content hai wo mind mei ek baar mei acche se nhi jaata... ye content hindi wala raise ghus gya mind mei ki bus ab hamesha yaad rhega... Thankyou for your efforts..

[@netviz8673]

decision tree for both classification regression here classification. 2 techniqiues. ID3 and CART. In Cart the decision tress spilts into binary trees. a)Entropy and Gini Index(Purity spirit) b) Information Gain(feature decision tree split). To check for Pure split two techniques called Entropy and Gini impurity are used and second technique called INformation gain (how the features are selected) is used. When H(S) is zero then that is pure split. And when H(s) is 1 then that is impure split ie equal distribution (eg 3yes and 3nos). The range of entropy remains between 0 to 1. In impure split the Gini impurity comes out to be 0.5 and in pure split it is 0. So the gini impurity ranges between 0 to 0.5. So in impure split the max value of gini impurity is 0.5 and in pure split it is 0. gini impurity is preferrable over entropy because of involvement of log it may slow down Now if you have multiple features, you use information gain to know how to make the tree using the given features whether which feature will start and which one will follow later. The feature starting with which the information gain calculation comes out to be the most should be the one with which the decision tree should be started.

[@techinicalocean8733]

Thank you so much buddy God bless you

[@aravindkumar6631]

God bless you frnd

[@prathameshekal7308]

मंडळ आभारी आहे

[@adityapradhan8474]

Thank you so much sir. All these maths concepts are helping a lot! I am able to visualize clearly that what's happening in the background of all these ml algorithms.

[@jasonbourn29]

Thanx sir in Hindi explanations you tend to cover topics better(English vedios are also of far better quality than anyone else)

[@pritamrajbhar9504]

It is one of the best and simplest explanations till far

[@GopalKrishna-p4m]

after one an era No one will beat you sir !! incredible explanation thankyou so much sir

[@patelanjali2909]

your teaching skill awesomwe.

[@RiteshBhalerao-b2v]

Great explaination...hard to find anywhere else👌👌

[@akashlinganwar4810]

hello krish sir... your explanation is easy to understand and anyone can learn easily..thank you sir...😊

[@utkarsh5165]

Thank you Krish for Crystal Clear Explanation.❤

[@adityatiwari7287]

Awesome Explanation....Thanks A Lot....Keep It Up !!!

[@maukaladka4100]

Worth watching😍😍😍

[@sandeepbehera3605]

wonderful explanations sir no one can explain like you...🙏🙏🙏 thankyou..sir😇

[@palakgrover9925]

Amazing Explanation 😃

[@Keep_Laughfing]

Sir aise hi video bnate rhiye apko shayd pta bhi nhi hogaa ki ye aapki kitni bdi help h DATA SCIENCE lovers ke liye. Dil se dhanyvaad 🙏🙏🙏

[@Er_IT_prashantjha]

Bohot achha explain krte ho Sir aap 👌🏻💯

[@aamiransarii]

Many, Many Thanks .....so lovely of you

[@mainakseal5027]

what an amazing tutorial...hats off sirji!!!...

[@DrZubairulIslam]

Thanks Krish, Best Ever Video, Wao.

[@sauravsahay8803]

You make everything look so easy

[@IrfanSaleem541]

Thanks a lot. Love and Respect from Oman

[@AbhishekSharma-cj9to]

this is really amazing sir 🙏

[@SumitJindal-e3k]

That was awesome

[@mohammadshad4168]

Thanks sir for the great explanation

[@ruhiraj9551]

Hello Krish sir ..thanku so much 🙏 for a very excellent explanation .

[@maths_impact]

Wonderful explanation given by you sir in hindi.

[@ketanitaliya9493]

Thanks it is really helpful and easy to understand

[@vidvaanwithromi5213]

wonderfully explained sir!

[@arunchougale5927]

Very Good explained by you .it is lot help me Thank u very much

[@dhavalsukhadiya8654]

great explaination sir

[@nisho404]

in one word bosssss

[@jaiprakashsingh756]

sir, I really find your videos very helpful. thanks a lot.

[@osamaosama-vh6vu]

Your legend deae sir thank u be happy 😍

[@gurpreetkaur-pf1bf]

Amazing ❤

[@bryan4592]

Amazing video sir

[@SonuK7895]

Very well explained sirjiiii

[@girishgogate2733]

Thankyou krish sir ........

[@AjmalShah-s7n]

Thank you sir G

[@tuhinbarai7592]

thank you sir..its to understand...

[@BikashKonwar-w7q]

In Entropy formula summation of p(x) * log2(p(x))

[@katw434]

Thanks sir please continue this series

[@abhiWorldIN]

Awesome video

[@mrityunjayupadhyay7332]

Great explanation

[@meenalpande]

Nice explanation

[@jasanimihir4994]

As always Very well explained. I have one query sir. You told that if the dataset is very big then use gini index otherwise entropy is fine. But finding the entropy is must for the information gain as no mention of Gini index in information gain formula. So is it possible to use gini index to find information gain? Kindly throw light on that. 😊

[@shaiqmahmood]

There is a way to calculate the Information Gain using Gini index as well.

[@krishj8011]

nice tutorial

[@razashaikh8698]

Sir, can we find information gain using ginny impurity?

[@abhinayjangde]

completed 😘

[@anusuyaghosal6961]

sir you said H(s) is the entropy of root node but i think it is the entropy of target attribute

[@Pankaj_Khanal_Joshi]

Sir sklearrn and seaborn ka video banaiye . thank u

[@neelkalyani7849]

In calculating information gain, can we use gini impurity instead of entropy?

[@maths_impact]

Wonderful

[@asifmahmoud126]

Start at 21:17

[@pkumar0212]

👌

[@sadiqpurki104]

I understand

[@sugandhaarora8174]

is this required for data analyst role?

[@abhishekbhad4029]

Very nice explanation sir.i have one question how to get intership as no one is hiring for fresher

[@shreyashnarvekar6391]

Where is the practical implementation ? can anyone guide me where can I find it ?

[@priyanshupokhariya8866]

0*log0 is undefined how is it coming 0??

[@VikasSingh-nq5yx]

Sirrrr.... ❤ I have a question 🙋! If interviewer ask a question why we are using minus ( - ) sign in Entropy? Please reply........ ❤

[@saranshsehrawat8544]

its formula

[@neeraj.kumar.1]

Don't worry they don't ask these types of mathematical formulas. They can ask what is Gini impurity.

[@rushikeshwaghmare3446]

Sir acc to external sites gini impurity ranges from 0-1

[@rushikeshwaghmare3446]

Please confirm on this…

[@prerakchoksi2379]

wow

[@vatsalshingala3225]

❤❤❤❤❤❤❤❤❤❤

[@SpectreWarfare]

Video volume is very less. It is difficult to listen

[@ShivamSharma-if1oh]

My answer of Entropy is coming 0.6 not 1

[@aaqibafraaz6215]

Hello sir

[@aaqibafraaz6215]

You r doing a great job

[@importantforlifeyasharora9042]

What a definition, entropy ranges 0 to 1 and ginni impurity ranges between 0 to 0.5.😂

[@roshanYadav-y8f]

❤❤❤

[@roshanYadav-y8f]

Kya bhai tum bhi data science ki preparation ker rahe ho