Classification problems are one where output variables are a categorical variables.

Example:

  • For sanctioning Loan checking whether customer would be default or not.

  • Checking incoming email is SPAM or not SPAM

Binary Classification : Two possible output.


Example :

Blood Sugar Level Diabetes

190 No

240 Yes

300 Yes

160 No

200 Yes

269 Yes

129 No

141 No

220 No

337 Yes


Sigmoid curve : This helps to deal with the deciding boundary problem using probability [ extremely low values in the start, extremely high values in the end, and intermediate values in the middle.]

Chesking best fit curve for sigmoud

by changing the beta0 and beta1 values. we can find.

P=1/(1+e^−(β0+β1x))


https://colab.research.google.com/drive/1S8OhFfYYP1MkG1mT0l9bznzqDekqEAko

Odds =p/(1-p)