Great question! At the end of the day, a model is really just a function that takes in some numeric input and spits out a numeric output. For categorical variables/inputs, we need a way to convert them into numeric form. One-hot encoding accomplishes this. It's a lossless way of encoding the information contained in a categorical input.
Correct. Since the information provided about the features is a bit vague, I tested out each categorical feature as both ordinal and nominal to see what made the models perform best. The combination I have in the video was the best I could find.
Sir what if we have to take input from user and then predict about the disease... please share how to convert into that type of model
Can you do this in weka?
This was of great help, thank you
Thanks so much, Akhila!
Very helpful thanks
Really great video thanks for your effor
Thank you!
Great video Bro
great video! I've got a qstn~ Why one-hot encoding process is needed ??
Actually, I am new in machine learning and I can't manage what the exact purpose of every single processes ㅠ.ㅠ :D
Great question! At the end of the day, a model is really just a function that takes in some numeric input and spits out a numeric output.
For categorical variables/inputs, we need a way to convert them into numeric form.
One-hot encoding accomplishes this. It's a lossless way of encoding the information contained in a categorical input.
Is chest pain a type of pain or is it a level of pain? If it is a level of pain then it would not be nominal. Agree?
Correct. Since the information provided about the features is a bit vague, I tested out each categorical feature as both ordinal and nominal to see what made the models perform best. The combination I have in the video was the best I could find.
Me and friends were doing cardiac arrhythmias prediction ..
Is this the same one bro?
I don't think so. This one is specifically for heart disease. You can view the dataset I used here:
www.kaggle.com/ronitf/heart-disease-uci
How would you calculate the recall_score or f1 score?
Do you mean what's the formula or how can we do it with code?
think ST is systolic blood pressure
Ah, thanks!
Hey