Induction Engine Demo: Train File Report

Return to Conclusion.
 
Sample Size = 250
Decision Tree (Entropy):

    -0.770161 0.811035


Sample Size = 250
Decision Tree (C4.5):

    0.504032 0.71758


Sample Size = 250
Naive Bayesian Algorithm:

    -0.177419 0.853991


Sample Size = 250
Decision Tree (Gini):

    -0.552419 0.806071


Sample Size = 250
Naive Models Algorithm:

Majority Model (Class NOredeem)   -1.4879 0.287719
Class redeemYES   0.653226 0.319687


Sample Size = 250
Logistic Regression Algorithm:

0.3   -0.0846774 0.619522
0.5   -0.758065 0.439941
0.7   -1.21774 0.489446


Sample Size = 250
k-Nearest Neighbor Algorithm:

1   -0.326613 0.724897
3   -0.78629 0.809476
5   -1.06855 0.540176
7   -1.06855 0.545107
10   -1.06452 0.443978
15   -1.4879 0.287719
20   -1.4879 0.287719
25   -1.4879 0.287719



Sample Size = 500
Decision Tree (C4.5):

    0.897177 0.48792


Sample Size = 500
Decision Tree (Entropy):

    -0.381048 0.645503


Sample Size = 500
Decision Tree (Gini):

    -0.211694 0.723084


Sample Size = 500
Naive Bayesian Algorithm:

    -0.330645 1.16528


Sample Size = 500
Naive Models Algorithm:

Class redeemYES   0.754032 0.370191
Majority Model (Class NOredeem)   -1.57863 0.333171


Sample Size = 500
Logistic Regression Algorithm:

0.3   0.483871 0.545738
0.5   -0.302419 0.577356
0.7   -1.21169 0.571838


Sample Size = 500
k-Nearest Neighbor Algorithm:

1   -0.120968 0.430549
3   -0.292339 0.609586
5   -0.461694 0.640649
7   -0.715726 0.536738
10   -0.71371 0.492455
15   -1.25403 0.515895
20   -1.25403 0.348791
25   -1.46976 0.342087



Sample Size = 1000
Decision Tree (Entropy):

    0.146 0.773146


Sample Size = 1000
Decision Tree (C4.5):

    1.188 0.491091


Sample Size = 1000
Decision Tree (Gini):

    0.125 0.687348


Sample Size = 1000
Naive Bayesian Algorithm:

    -1.241 0.835776


Sample Size = 1000
Naive Models Algorithm:

Class redeemYES   0.89 0.454438
Majority Model (Class NOredeem)   -1.701 0.408995


Sample Size = 1000
Logistic Regression Algorithm:

0.3   0.93 0.5295
0.5   0.381 0.336814
0.7   -0.918 0.325151


Sample Size = 1000
k-Nearest Neighbor Algorithm:

1   -0.045 0.463605
3   -0.465 0.320498
5   -0.522 0.411931
7   -0.744 0.481654
10   -0.652 0.408728
15   -1.004 0.349131
20   -1.078 0.301698
25   -1.306 0.247382



Sample Size = 2000
Decision Tree (Entropy):

    0.4175 0.897437


Sample Size = 2000
Decision Tree (Gini):

    0.4695 0.778223


Sample Size = 2000
Naive Bayesian Algorithm:

    -1.4155 0.709754


Sample Size = 2000
Decision Tree (C4.5):

    1.3505 0.210186


Sample Size = 2000
Naive Models Algorithm:

Class redeemYES   0.92 0.309839
Majority Model (Class NOredeem)   -1.728 0.278855


Sample Size = 2000
Logistic Regression Algorithm:

0.3   1.003 0.294465
0.5   0.5835 0.24192
0.7   -0.7305 0.187981


Sample Size = 2000
k-Nearest Neighbor Algorithm:

1   0.249 0.315987
3   -0.1415 0.341539
5   -0.295 0.301643
7   -0.4565 0.256512
10   -0.387 0.169864
15   -0.763 0.170106
20   -0.864 0.204744
25   -1.1925 0.249932



Sample Size = 4000
Decision Tree (Entropy):

    0.58775 0.713632


Sample Size = 4000
Naive Bayesian Algorithm:

    0.695 0.132738


Sample Size = 4000
Decision Tree (Gini):

    0.62725 0.675614


Sample Size = 4000
Decision Tree (C4.5):

    1.42975 0.225927


Sample Size = 4000
Naive Models Algorithm:

Majority Model (Class NOredeem)   -1.75275 0.093747
Class redeemYES   0.9475 0.104163


Sample Size = 4000
Logistic Regression Algorithm:

0.3   1.03 0.159614
0.5   0.61925 0.163902
0.7   -0.62075 0.083843


Sample Size = 4000
k-Nearest Neighbor Algorithm:

1   0.78575 0.062778
3   0.289 0.114073
5   0.07975 0.143537
7   -0.1985 0.13879
10   -0.154 0.0877855
15   -0.481 0.072301
20   -0.54075 0.0998796
25   -0.80875 0.0969297



Sample Size = 7999
Decision Tree (C4.5):

    1.61236 0.112012


Sample Size = 7999
Decision Tree (Gini):

    0.678053 0.530006


Sample Size = 7999
Decision Tree (Entropy):

    0.71484 0.554014


Sample Size = 7999
Naive Bayesian Algorithm:

    0.648398 0.205127


Sample Size = 7999
Naive Models Algorithm:

Majority Model (Class NOredeem)   -1.78378 0.0750665
Class redeemYES   0.981982 0.0834072


Sample Size = 7999
Logistic Regression Algorithm:

0.3   1.03971 0.0476537
0.5   0.62696 0.033325
0.7   -0.613614 0.148847


Sample Size = 7999
k-Nearest Neighbor Algorithm:

1   1.52052 nan
3   0.236236 nan
5   0.475475 nan
7   0 nan
10   0.205205 nan
15   -0.184184 nan
20   -0.223223 nan
25   -0.36036 nan



Return to Conclusion.


©2002 PredictionWorks, Inc. All rights reserved.