AV Practice Problem Loan Prediction

ensemble_methods
r

#1

I am trying this practice problem. Someone suggested me to use ensemble techniques. How do I use ensemble techniques on a categorical variable? Loan_Status is a categorical variable with only two outcomes ‘Y’ and ‘N’.
Is it possible by giving hypothetical number to ‘Y’ and ‘N’?


#2

@gau2112 - You can make different models like a random forest, boosting and decision tree and create a data frame of each model and use mode function to select a maximum occurring value for each observation from a different model.

For example
    # adding output of different model to make a single data frame  name as submit_match3
    submit_match3$cuisine2 <- submit_match2$cuisine 
    submit_match3$cuisine1 <- submit_match1$cuisine

     MODE function to extract the predicted value with the highest frequency per id.

    #function to find the maximum value row wise
    Mode <- function(x) {
    u <- unique(x)
    u[which.max(tabulate(match(x, u)))]
    }
    x <- Mode(submit_match3[,c("cuisine","cuisine2","cuisine1")])
    y <- apply(submit_match3,1,Mode)

Hope this helps!

Regards,
Hinduja


#3

Thanks a lot.


#4

For Python, use following EnsembleClassifier that combines different models.
Check below URL who developed this module and also explained with examples.


#5

hey.can someone please share the solution in R. If possible please mail.It will be really helpful. TIA!

ankitpaul7@gmail.com