Build final classification model

Comparing the recall performance between the logistic regression model (0.4) and the best performing random forest model (0.2), you've learned that the model with the best performance is the logistic regression model. In this exercise, you will build the logistic regression model using all of the train data and you will prepare the necessary vectors for evaluating this model's test performance.

This exercise is part of the course

Machine Learning in the Tidyverse

View Course

Exercise instructions

Build a logistic regression model predicting Attrition using all available features in the training_data.
Prepare the binary vector of actual test values, test_actual.
Prepare the binary vector of predicted values where a probability greater than 0.5 indicates TRUE and store this as test_predicted.

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# Build the logistic regression model using all training data
best_model <- glm(formula = ___, 
                  data = ___, family = "binomial")


# Prepare binary vector of actual Attrition values for testing_data
test_actual <- testing_data$___ == "___"

# Prepare binary vector of predicted Attrition values for testing_data
test_predicted <- predict(___, ___, type = "response") > ___

Edit and Run Code