Question 2 Supervised learning models

1. The following dataset contains some data points with 3 features x1, x2, x3 and a labely.Apply kNN to predict the label of the query point(10,4,4)usingk =3.Show your work including the calculated distances(in 2 decimal places)of the data points to the query point, their respective ranks, the k nearest neighbours, and the prediction result.

Data point	x1	x2	x3	y
#1	15	2	2	6.3
#2	4	8	11	4.0
#3	12	6	8	9.3
#4	4	2	7	3.5
#5	8	8	2	4.5
#6	11	4	7	6.9
#7	10	11	-1	8.3

1. The following dataset contains the same features x1, x2, x3 as in part (a)above, but the label y is categorical (classes P and Q) instead of numerical. Apply kNN to predict the label of the query point (10, 4, 4) using k = 3, and show the predicted result.

Data point	x1	x2	x3	y
#1	15	2	2	P
#2	4	8	11	Q
#3	12	6	8	P
#4	4	2	7	P
#5	8	8	2	Q
#6	11	4	7	Q
#7	10	11	-1	P

1. Thefollowing dataset containssome example dataitems with 3categorical features

x1 (classesAandB),x2(classes C andD),x3(classes E, FandG)andalabely

(classes P, Q, andR). Apply the ID3 algorithm to builda decision tree. Show your work including the calculated entropy values, the information gain values, and the resulting decision tree. In case of a tie, select the first of the candidate features for splitting. Use 4 decimal places for floating-point numbers. [15]

Example #	x1	x2	x3	y
#1	A	C	E	P
#2	A	C	F	P
#3	A	C	G	Q
#4	A	D	E	R
#5	A	D	F	P
#6	A	D	G	Q
#7	B	C	E	P
#8	B	C	F	R
#9	B	C	G	Q
#10	B	D	E	R
#11	B	D	F	R