Question 2. – Classification Trees (20 points)
The admissions office of the X Business School wants to base its future initial screening for
admissions decisions on the data below, based on both grades as well as recommendations from
faculty. The target variable is “admit,” which is characterized by either Yes or No. Grades A B A A B B B B A B Recommendations High Low High Low High High Low High Low Low Admit? Yes No No Yes Yes No No Yes No No Using this dataset, the following model is developed: a. (5 points) What is the probability that an applicant with a HIGH recommendation will
be admitted? 1 (Question 2 continues) Apply this model to the test data provided in the following table. The
model will recommend admitting an applicant, if the predicted probability that this applicant will
be admitted is greater than 50%.
Grades B B B B A A A B B B Recommendations Low High Low High High Low High Low Low High Admit? No Yes No Yes No Yes Yes No No No Predicted Admit?
be admitted is greater than 50%.
Grades B B B B A A A B B B Recommendations Low High Low High High Low High Low Low High Admit? No Yes No Yes No Yes Yes No No No Predicted Admit?
b. (5 points) Complete the third row of this table (Predicted_Admit?). What is the accuracy rate
of the model on this test dataset?
c. (5 points) Create a confusion matrix. Be sure to properly label the columns and rows.
d. (5 points) What does a false positive mean in this context using plain English? What about a
false negative? Your description should be specific to this particular context. 2
false negative? Your description should be specific to this particular context. 2
Categories: