Confusion matrix, Precision, Recall| Data Science Interview questions

Here we will discuss about the confusion matrix. We will also discuss about the Precision and Recall. These are the most important question For Data science, Machine Learning and Artificial Intelligence.

[@MsRAJDIP]

you are too good at explanation, your contents are better than all paid courses in coursera,udemy etc

[@duongtrungtran6300]

I'm preparing for the interview. This very helpful! Thanks a lot

[@louerleseigneur4532]

Thanks Krish

[@suneelgrover1]

Thanks Krish, This really helped me to understand concepts of confusion matrix which actually confused me a lot earlier :)

[@DS_AIML]

Thankyou Krish.Its really helpful.

[@kill_me412]

Nice Explanation. It would be very helpful if the delivery speech were slow.

[@jagadeesh6074]

Thanks sir .............u r videos very helpful to us.good work sir

[@madanmaram276]

Great Explanation

[@akashsharma3607]

Really helpful ... Thanks

[@pranjalgupta9427]

Thanks😄🙏❤

[@supriyomukherjee9256]

Excellent Explanation...

[@8149272052]

thankyou so much sir very crstal clear explanation

[@bloomingtonvolleyball639]

Very good explanation. Can you please post a video demonstrating how to deploy a machine learning model and create and use an API out of it. Thanks a lot.

[@rahuljaiswal9379]

sir could u please correct ur type 1 error...... u taught wrong . type 1 error is always FP and type 2 is FN.... or please let me know if i did wrong

[@krishnaik06]

No u are right..that was a mistake

[@ravishlover3945]

Hi krish, everything fine but i think type 1 error is the "false positive "
n type 2 is the false negative.

[@lorenzoleongutierrez7927]

bravo!

[@Tjos4155]

very very good video krish ..

[@sukumarroychowdhury4122]

Just Recall & Precision are enough. No need to be further confused with FPR (& TPR as well)

[@kkckk4360]

Hi sir ,u r explination awesome.
Could you please explain the roles and responsibilities for Data Scientist position.

[@brindabansikdar9726]

FP: Type 1 error; not dangerous
FN: Type 2 error: dangerous in medical analysis

[@brindabansikdar9726]

In this video >> wrongly described

[@himanshusahoo143]

Excellent explanation sir

[@madhugarg7499]

Nice explanation Krish. At 22.59, for the 3-categories, how can we calculate the '0' & '1' values ? Because we are having 3 rows & 3 columns.
Kindly advise.

[@harshayalemane5313]

Thank you for the video , can you please explain how to build the confusion matrix for actual = [-1,1,5,1,-10]
predicted = [0,1,1,1,0], without programming ie how to build it manually , awaiting your response , thanks

[@genesonsubra438]

Hi Krish, can you do an unsupervised model where you have a target column, and verify the if fitting clustering methods matches the prediction? You can try aggressive driving from kaggle.
thanks

[@shyamchaurasiya1069]

Thank you so very much ♥️♥️♥️!

[@pallabidas3064]

thanx a lot sir

[@raneshmitra8156]

kindly upload videos on Skewness, kurtosis.........

[@roblee3667]

Can you explain further evaluation concepts like ROC and AOC curves

[@SHRINANGLIDARBAR]

Good job sir 👌💞

[@gandharvsikri1378]

Hello
Could you upload a video on how generalized linear regression model handles categorical variables ?

[@supriyomukherjee9256]

Hello Sir, Could you please upload some videos on techniques pertaining to handling imbalance dataset. Thanks a lot..

[@krishnaik06]

Hi, I will be uploading till tomorrow

[@dineshs4153]

please ADD subtitiles

[@srinivask3120]

Hi Krish, In real time projects do we get target variable. if not how can we get target variable.

[@arkaprabhabasu267]

so recall is TNR and TPR?

[@bishwajeetsingh8834]

FN / TP + FN = FPR

[@arjyabasu1311]

Sir does all these terminologies work for regression problems too??

[@krishnaik06]

This works for classification problem

[@arjyabasu1311]

@@krishnaik06 Sir then what is the standard process for evaluating regression problems?

[@rajeshs2840]

@@arjyabasu1311 Root mean squared error. for continuous value outputs.... Confusion matrix is only for discrete value outputs.