Find optimal independent variable values in a classification problem

I have a data set with 4 independent variables (X) and one dependent variable(Y). All the X variables here are on a rating scale of 1-10. How do i approach this problem.

  1. Should i treat the X variables as ordinal or interval (I googled a lot but still could not decide)
  2. Since Y is binary here, i cant find the correlation between the X variables and Y i assume. How to handle such a situation.
  3. What should i do to improve the success rate here. Success rate is the number of positive outcomes(YES) out of the total (see attached pic)
  4. Is there a threshold value that i can find for each of the X variable that will result in a positive outcome. For eg, If the minimum values of X variables are 4,7,4,6 then the outcome will be positive
  5. In general are there any suggestions on how i can approach such a business problem. My ultimate goal here is to increase the success rate to 60% which is currently at 36%survey%20analysis

Since the variables can take only particular discrete values we can consider these as categorical data as for us wifi speed 9.98 won’t make a sense as such and see how are our results .
Also the input data has approximately equal no. of sold="yes’ and sold=“no” right ?
beacuse otherwise we have to implement techniques that handle skewed datasets

© Copyright 2013-2019 Analytics Vidhya