In this repo I use decision trees to solve several well-known machine learning problems.
Rules:
Rule 1: (981/146, lift 1.3) AGE = Adulto SEX = Hombre -> class Muere [0.850]
Rule 2: (681/153, lift 1.1) CLASS = 3aa -> class Muere [0.775]
Rule 3: (30, lift 3.0) CLASS in {1a, 2a} AGE = Ninio -> class Sobrevive [0.969]
Rule 4: (251/17, lift 2.9) CLASS in {1a, 2a} SEX = Mujer -> class Sobrevive [0.929]
Rules:
Rule 1: (1150/244, lift 1.2) SEX = Hombre -> class Muere [0.787]
Rule 2: (457/113, lift 1.1) CLASS = 3aa -> class Muere [0.752]
Rule 3: (188/17, lift 2.8) CLASS in {crew, 1a, 2a} SEX = Mujer -> class Sobrevive [0.905]
Rules:
Rule 1: (1163/244, lift 1.2) SEX = Hombre -> class Muere [0.790]
Rule 2: (483/127, lift 1.1) CLASS = 3aa -> class Muere [0.736]
Rule 3: (178/13, lift 2.9) CLASS in {2a, crew, 1a} SEX = Mujer -> class Sobrevive [0.922]
Note: Model 2 and 3 are effectively the same.
Rules:
Rule 1: (33, lift 2.9) PetalLengthCm <= 1.9 -> class Iris-setosa [0.971]
Rule 2: (66/33, lift 1.5) PetalLengthCm > 1.9 -> class Iris-versicolor [0.500]
Rule 3: (26, lift 2.9) PetalLengthCm > 5 -> class Iris-virginica [0.964]
Rule 4: (16, lift 2.8) SepalWidthCm <= 2.9 PetalWidthCm > 1.5 -> class Iris-virginica [0.944]
Rules:
Rule 1: (33, lift 2.9) PetalLengthCm <= 1.9 -> class Iris-setosa [0.971]
Rule 2: (32/1, lift 2.8) PetalLengthCm > 1.9 PetalLengthCm <= 4.9 PetalWidthCm <= 1.7 -> class Iris-versicolor [0.941]
Rule 3: (28, lift 2.9) PetalWidthCm > 1.7 -> class Iris-virginica [0.967]
Rule 4: (31/2, lift 2.7) PetalLengthCm > 4.9 -> class Iris-virginica [0.909]