Skip to main content

Table 5 Comparison between accuracies of five best trees

From: Selecting a representative decision tree from an ensemble of decision-tree models for fast big data classification

Algorithm name

UCI dataset name

Five best SySM train 80%

Five most accurate trees train 80%

Five best SySM train 90%

CART

Poker Hand

0.529

0.536

0.519

CART

SUSY

0.782

0.786

0.786

CART

RLCP

0.997

0.998

0.998

CART

KDD Cup

0.998

0.997

0.999

CART

Household Elec.

0.998

0.997

0.999

CART

HIGGS

0.998

0.997

0.999

J48

Poker Hand

0.557

0.589

0.531

J48

SUSY

0.693

0.737

0.783

J48

RLCP

0.999

0.999

0.999

J48

KDD Cup

1.000

1.000

0.998

J48

Household Elec.

0.557

0.556

0.555

J48

HIGGS

0.652

0.649

0.648

  1. In italics the most accurate per line