Video není dostupné.
Omlouváme se.
Feature Selection using 1 Way ANOVA or F-Test in 10 Minutes
Vložit
- čas přidán 15. 06. 2020
- If you watch the entire video, you should be able to explain
1) What is feature selection, why it is important ?
2) How this can be fitted in the framework of 1 way ANOVA and F Test
3) A demonstration of F Score have been Shown
www.kaggle.com...
Thank you sir
really helpful
Thank you for sharing Saptarsi. Very simple and easy to understand.
Thanks a lot
Thanbk you very much sir
Thanks for your comments
It is really amazing in understanding one way annova
Thanks a lot Santanu
Thank you Sir, it is well explained.
Thank you
Great video, thank you
Thanks a lot
Good job! Thanks
Thanks for your comments
If we apply some transformations on the columns which have skewed data to normalize it then maybe we will get much higher accuracy with feature selection.
Thanks for your suggestion, later on will try to come up with video on standard transformations and skew ness of data
This also provides framework by virtue you can diffirentiate between one way and two way annova
Thanks
Sir for applying ANOVA for feature selection we need to apply normality test to demonstrate whether our data follows a normal distribution? Or we can apply for any type of data set without checking the normal distribution.... could you please clarify it?
Thank you so much sir!...Can we use Anova for bigger data set? Or any advanced algorithm like PCA or LDA can be used for large data set?...and first step we should do before ANOVA is we should see if the population is normally distributed!
That are way too many questions. But generally large data set is subjective, anova should not be an issue, for faster calculation of PCA, LDA there are mini batch, incremental kind of methods. Ideally a normal distribution check is good, but not used so much.
@@SaptarsiGoswami Thank you so much for your reply sir!...Please keep up the good work!
great video. I can't understand how scores gets calculated can you help me??
Thank you so much
You're most welcome