Mv

This is an artificial data set with dependencies between the attribute values. The cases are generated using the following method:
X₁ : uniformly distributed over [-5,5]
X₂ : uniformly distributed over [-15,-10]
X₃ : IF (X₁ > 0) THEN X₃ = green
ELSE X₃ = red with probability 0.4 and X₄=brown with prob. 0.6
X₄ : IF (X₃=green) THEN X₄=X₁+2X₂
ELSE X₄=X₁/2 with prob. 0.3, and X₄=X₂/2 with prob. 0.7
X₅ : uniformly distributed over [-1,1]
X₆ : X₆=X₄×[epsilon], where [epsilon] is uniformly distribute over [0,5]
X₇ : X₇=yes with prob. 0.3 and X₇=no with prob. 0.7
X₈ : IF (X₅ < 0.5) THEN X₈ = normal ELSE X₈ = large
X₉ : uniformly distributed over [100,500]
X₁₀ : uniformly distributed integer over the interval [1000,1200]

Obtain the value of the target variable Y using the rules:
IF (X₂ > 2 ) THEN Y = 35 - 0.5 X₄
ELSE IF (-2 <= X₄ <= 2) THEN Y = 10 - 2 X₁
ELSE IF (X₇ = yes) THEN Y = 3 -X₁/X₄
ELSE IF (X₈ = normal) THEN Y = X₆ + X₁
ELSE Y = X₁/2

• Source: Contact me (ltorgo@liacc.up.pt).
• Characteristics: 40768 cases, 10 attributes (3 nominal, 7 continuous).
• Download : mv.tar.gz(132146 bytes)