A B C D E Gini Var A
4.8 3.4 1.9 0.2 positive Positive Negative Row total
5 3 1.6 0.2 positive >=5.2 3 7 10
5 3.4 1.6 0.4 positive <5.2 5 1 6
5.2 3.5 1.5 0.2 positive Grand total 16
5.2 3.4 1.4 0.2 positive Probability = Number of events/ total number of events
4.7 3.2 1.6 0.2 positive Gini(3,7) 0.42
4.8 3.1 1.6 0.2 positive Gini(5,1) 0.27777778
5.4 3.4 1.5 0.4 positive Gini Var A 0.36666667
7 3.2 4.7 1.4 negative
6.4 3.2 4.5 1.5 negative Gini Var C
6.9 3.1 4.9 1.5 negative Positive Negative Row total
5.5 2.3 4 1.3 negative >=4.2 0 6 6
6.5 2.8 4.6 1.5 negative <4.2 8 2 10
5.7 2.8 4.5 1.3 negative Grand total 16
6.3 3.3 4.7 1.6 negative
4.9 2.4 3.3 1 negative Gini(0,6) 0
Gini(8,2) 0.32
Gini Var C 0.2
Lowest Gini and highest information gain
calculate for all variables
Entropy & Information Gain of Var A
Positive Negative Row total
>=5.2 3 7 10
<5.2 5 1 6
Grand total 16
Entropy(3,7) 0.8812909
Entropy(5,1) 0.65002242
Entropy Var A 0.79456522
Dependent VaPositive Negative
8 8 16
Entropy(8,8) 1
Information gain of Var A 0.20543478
Entropy & Information Gain of Var C
Positive Negative Row total
>=4.2 0 6 6
<4.2 8 2 10
Grand total 16
Entropy(0,6) 0
Entropy(8,2) 0.72192809
Entropy Var C 0.45120506
Information gain of Var C 0.54879494
