Question

A realtor is studying housing values in the suburbs of Minneapolis and has given you a dataset with the following attributes: crime rate in the neighborhood, proximity to Mississippi river, number of rooms per dwelling, age of unit, distance to Minneapolis and Saint Paul Downtown, distance to shopping malls. The target variable is the cost of the house (with values high and low). Given this scenario, indicate the choice of classifier for each of the following questions and give a brief explanation.

a) If the realtor wants a model that not only performs well but is also easy to interpret, which one would you choose between SVM, Decision Trees and kNN?
b) If you had to choose between RIPPER and Decision Trees, which one would you prefer for a classification problem where there are missing values in the training and test data?
c) If you had to choose between RIPPER and KNN, which one would you prefer if it is known that there are very few houses that have high cost?

Answer 1

a. decision trees

b. decision trees

c. rippers

Step-by-step explanation:

a) I will choose Decision trees because these can be better interpreted compared to these other two KNN and SVM. using Decision tress gives us a better explanation than the other 2 models in this question.

b) In a classification problem with missing values, Decision trees are better off rippers since Rippers avoid the missing values.

c) Ripper are when we know some are high cost houses.