Data Mining Question

A dataset with 9 examples contains 3 features a1, a2, and a3, and a class variable. Two of the features are binary and one is continuous, while the class variable is binary. The dataset is described below:

a1	a2	a3	Class
T	T	1.0	+
T	T	6.0	+
T	F	5.0	-
F	F	4.0	+
F	T	7.0	-
F	T	3.0	-
F	F	8.0	-
T	F	7.0	+
F	T	5.0	-

Answer each of the following questions with respect to the dataset. You must show your work and intermediary calculations. You must do this yourself and not use a data mining tool. Please circle your final answers for each part.

1. What is the misclassification error rate associated with splitting on a1 and with splitting on a2, and which feature yields the best split?
2. What is the Gini value for a1 and a2 and which yields the best split (based on Gini)?