Está en la página 1de 14

Anexo 1

=== Run information ===

Evaluator: weka.attributeSelection.InfoGainAttributeEval
Search: weka.attributeSelection.Ranker -T -1.7976931348623157E308 -N -1
Relation: contraceptives-weka.filters.unsupervised.attribute.Remove-R3,8-
weka.filters.unsupervised.attribute.Remove-R6
Instances: 1474
Attributes: 6
edad
edu
n_hijos
islam
trabajo
metodo
Evaluation mode: evaluate on all training data

=== Attribute Selection on all input data ===

Search Method:
Attribute ranking.

Attribute Evaluator (supervised, Class (nominal): 6 metodo):


Information Gain Ranking Filter

Ranked attributes:
0.10188 3 n_hijos
0.06583 2 edu
0.06166 1 edad
0.00983 4 islam
0.00263 5 trabajo

Selected attributes: 3,2,1,4,5 : 5


=== Run information ===

Evaluator: weka.attributeSelection.GainRatioAttributeEval
Search: weka.attributeSelection.Ranker -T -1.7976931348623157E308 -N -1
Relation: contraceptives-weka.filters.unsupervised.attribute.Remove-R3,8-
weka.filters.unsupervised.attribute.Remove-R6
Instances: 1474
Attributes: 6
edad
edu
n_hijos
islam
trabajo
metodo
Evaluation mode: evaluate on all training data

=== Attribute Selection on all input data ===

Search Method:
Attribute ranking.

Attribute Evaluator (supervised, Class (nominal): 6 metodo):


Gain Ratio feature evaluator

Ranked attributes:
0.06614 3 n_hijos
0.0419 2 edu
0.03565 1 edad
0.01617 4 islam
0.00323 5 trabajo

Selected attributes: 3,2,1,4,5 : 5



Anexo 2
=== Run information ===

Scheme: weka.classifiers.trees.J48 -C 0.25 -M 15


Relation: contraceptives-weka.filters.unsupervised.attribute.Remove-R3,8-
weka.filters.unsupervised.attribute.Remove-R6
Instances: 1474
Attributes: 6
edad
edu
n_hijos
islam
trabajo
metodo
Test mode: split 66.0% train, remainder test

=== Classifier model for training split (973 instances) ===

J48 pruned tree


------------------

n_hijos <= 0: A (65.0/1.0)


n_hijos > 0
| edad <= 37
| | edu <= 3
| | | n_hijos <= 2
| | | | trabajo = no
| | | | | edad <= 25
| | | | | | n_hijos <= 1: C (49.0/24.0)
| | | | | | n_hijos > 1: A (37.0/18.0)
| | | | | edad > 25: A (47.0/25.0)
| | | | trabajo = si: A (34.0/12.0)
| | | n_hijos > 2: C (227.0/97.0)
| | edu > 3
| | | n_hijos <= 2
| | | | n_hijos <= 1
| | | | | edad <= 24: C (30.0/16.0)
| | | | | edad > 24: A (40.0/22.0)
| | | | n_hijos > 1: C (74.0/41.0)
| | | n_hijos > 2
| | | | edad <= 32: C (44.0/22.0)
| | | | edad > 32: B (59.0/27.0)
| edad > 37
| | edu <= 3: A (153.0/45.0)
| | edu > 3
| | | n_hijos <= 2: A (20.0/6.0)
| | | n_hijos > 2
| | | | edad <= 46: B (76.0/28.0)
| | | | edad > 46: A (18.0/9.0)
Number of Leaves : 15

Size of the tree : 29

=== Evaluation on test split ===

Time taken to test model on test split: 0 seconds

=== Summary ===

Correctly Classified Instances 281 56.0878 %


Incorrectly Classified Instances 220 43.9122 %
Total Number of Instances 501

=== Confusion Matrix ===

a b c <-- classified as
145 10 67 | a = A
31 34 44 | b = B
44 24 102 | c = C
=== Run information ===

Scheme: weka.classifiers.trees.J48 -C 0.25 -M 15


Relation: contraceptives-weka.filters.unsupervised.attribute.Remove-R3,8-
weka.filters.unsupervised.attribute.Remove-R6-weka.filters.unsupervised.attribute.Remove-R5
Instances: 1474
Attributes: 5
edad
edu
n_hijos
islam
metodo
Test mode: split 66.0% train, remainder test

=== Classifier model for training split (973 instances) ===

J48 pruned tree


------------------

n_hijos <= 0: A (65.0/1.0)


n_hijos > 0
| edad <= 37
| | edu <= 3
| | | n_hijos <= 2: A (167.0/85.0)
| | | n_hijos > 2: C (227.0/97.0)
| | edu > 3
| | | n_hijos <= 2
| | | | n_hijos <= 1
| | | | | edad <= 24: C (30.0/16.0)
| | | | | edad > 24: A (40.0/22.0)
| | | | n_hijos > 1: C (74.0/41.0)
| | | n_hijos > 2
| | | | edad <= 32: C (44.0/22.0)
| | | | edad > 32: B (59.0/27.0)
| edad > 37
| | edu <= 3: A (153.0/45.0)
| | edu > 3
| | | n_hijos <= 2: A (20.0/6.0)
| | | n_hijos > 2
| | | | edad <= 46: B (76.0/28.0)
| | | | edad > 46: A (18.0/9.0)

Number of Leaves : 12

Size of the tree : 23

=== Evaluation on test split ===

Time taken to test model on test split: 0 seconds


=== Summary ===

Correctly Classified Instances 283 56.487 %


Incorrectly Classified Instances 218 43.513 %
Total Number of Instances 501

=== Confusion Matrix ===

a b c <-- classified as
158 10 54 | a = A
32 34 43 | b = B
55 24 91 | c = C
=== Run information ===

Scheme: weka.classifiers.trees.J48 -C 0.25 -M 15


Relation: contraceptives-weka.filters.unsupervised.attribute.Remove-R3,8-
weka.filters.unsupervised.attribute.Remove-R6-weka.filters.unsupervised.attribute.Remove-R4
Instances: 1474
Attributes: 5
edad
edu
n_hijos
trabajo
metodo
Test mode: split 66.0% train, remainder test

=== Classifier model for training split (973 instances) ===

J48 pruned tree


------------------

n_hijos <= 0: A (65.0/1.0)


n_hijos > 0
| edad <= 37
| | edu <= 3
| | | n_hijos <= 2
| | | | trabajo = no
| | | | | edad <= 25
| | | | | | n_hijos <= 1: C (49.0/24.0)
| | | | | | n_hijos > 1: A (37.0/18.0)
| | | | | edad > 25: A (47.0/25.0)
| | | | trabajo = si: A (34.0/12.0)
| | | n_hijos > 2: C (227.0/97.0)
| | edu > 3
| | | n_hijos <= 2
| | | | n_hijos <= 1
| | | | | edad <= 24: C (30.0/16.0)
| | | | | edad > 24: A (40.0/22.0)
| | | | n_hijos > 1: C (74.0/41.0)
| | | n_hijos > 2
| | | | edad <= 32: C (44.0/22.0)
| | | | edad > 32: B (59.0/27.0)
| edad > 37
| | edu <= 3: A (153.0/45.0)
| | edu > 3
| | | n_hijos <= 2: A (20.0/6.0)
| | | n_hijos > 2
| | | | edad <= 46: B (76.0/28.0)
| | | | edad > 46: A (18.0/9.0)

Number of Leaves : 15
Size of the tree : 29

=== Evaluation on test split ===

Time taken to test model on test split: 0 seconds

=== Summary ===

Correctly Classified Instances 281 56.0878 %


Incorrectly Classified Instances 220 43.9122 %
Total Number of Instances 501

=== Confusion Matrix ===

a b c <-- classified as
145 10 67 | a = A
31 34 44 | b = B
44 24 102 | c = C
=== Run information ===

Scheme: weka.classifiers.trees.J48 -C 0.25 -M 15


Relation: contraceptives-weka.filters.unsupervised.attribute.Remove-R3,8-
weka.filters.unsupervised.attribute.Remove-R6-weka.filters.unsupervised.attribute.Remove-R3
Instances: 1474
Attributes: 5
edad
edu
islam
trabajo
metodo
Test mode: split 66.0% train, remainder test

=== Classifier model for training split (973 instances) ===

J48 pruned tree


------------------

edu <= 3
| edad <= 37
| | trabajo = no
| | | edu <= 2
| | | | edad <= 23: A (44.0/17.0)
| | | | edad > 23: C (133.0/62.0)
| | | edu > 2: C (163.0/84.0)
| | trabajo = si: A (86.0/38.0)
| edad > 37: A (159.0/45.0)
edu > 3
| edad <= 30
| | trabajo = no
| | | islam = si: C (88.0/43.0)
| | | islam = no: A (17.0/8.0)
| | trabajo = si: A (49.0/23.0)
| edad > 30
| | edad <= 44
| | | trabajo = no
| | | | edad <= 35
| | | | | islam = si: B (36.0/21.0)
| | | | | islam = no: C (15.0/7.0)
| | | | edad > 35: B (86.0/37.0)
| | | trabajo = si
| | | | edad <= 40
| | | | | islam = si: C (26.0/14.0)
| | | | | islam = no: B (19.0/9.0)
| | | | edad > 40: B (15.0/6.0)
| | edad > 44: A (37.0/18.0)

Number of Leaves : 15
Size of the tree : 29

=== Evaluation on test split ===

Time taken to test model on test split: 0 seconds

=== Summary ===

Correctly Classified Instances 232 46.3074 %


Incorrectly Classified Instances 269 53.6926 %
Total Number of Instances 501

=== Confusion Matrix ===

a b c <-- classified as
114 14 94 | a = A
42 27 40 | b = B
48 31 91 | c = C
=== Run information ===

Scheme: weka.classifiers.trees.J48 -C 0.25 -M 15


Relation: contraceptives-weka.filters.unsupervised.attribute.Remove-R3,8-
weka.filters.unsupervised.attribute.Remove-R6-weka.filters.unsupervised.attribute.Remove-R2
Instances: 1474
Attributes: 5
edad
n_hijos
islam
trabajo
metodo
Test mode: split 66.0% train, remainder test

=== Classifier model for training split (973 instances) ===

J48 pruned tree


------------------

n_hijos <= 0: A (65.0/1.0)


n_hijos > 0
| edad <= 37
| | n_hijos <= 2
| | | trabajo = no
| | | | edad <= 28: C (174.0/90.0)
| | | | edad > 28: A (59.0/28.0)
| | | trabajo = si: A (78.0/43.0)
| | n_hijos > 2: C (330.0/159.0)
| edad > 37
| | n_hijos <= 2: A (53.0/13.0)
| | n_hijos > 2
| | | edad <= 47
| | | | islam = si
| | | | | edad <= 45
| | | | | | trabajo = no
| | | | | | | n_hijos <= 5: B (49.0/25.0)
| | | | | | | n_hijos > 5: A (49.0/21.0)
| | | | | | trabajo = si: A (21.0/10.0)
| | | | | edad > 45: A (35.0/14.0)
| | | | islam = no: B (36.0/17.0)
| | | edad > 47: A (24.0/5.0)

Number of Leaves : 12

Size of the tree : 23

=== Evaluation on test split ===

Time taken to test model on test split: 0 seconds


=== Summary ===

Correctly Classified Instances 260 51.8962 %


Incorrectly Classified Instances 241 48.1038 %
Total Number of Instances 501

=== Confusion Matrix ===

a b c <-- classified as
124 12 86 | a = A
40 14 55 | b = B
40 8 122 | c = C
=== Run information ===

Scheme: weka.classifiers.trees.J48 -C 0.25 -M 15


Relation: contraceptives-weka.filters.unsupervised.attribute.Remove-R3,8-
weka.filters.unsupervised.attribute.Remove-R6-weka.filters.unsupervised.attribute.Remove-R1
Instances: 1474
Attributes: 5
edu
n_hijos
islam
trabajo
metodo
Test mode: split 66.0% train, remainder test

=== Classifier model for training split (973 instances) ===

J48 pruned tree


------------------

n_hijos <= 0: A (65.0/1.0)


n_hijos > 0
| edu <= 3
| | n_hijos <= 7
| | | n_hijos <= 2: A (200.0/92.0)
| | | n_hijos > 2
| | | | edu <= 1
| | | | | n_hijos <= 4: C (24.0/11.0)
| | | | | n_hijos > 4: A (23.0/9.0)
| | | | edu > 1: C (250.0/132.0)
| | n_hijos > 7: A (50.0/17.0)
| edu > 3
| | n_hijos <= 2
| | | n_hijos <= 1
| | | | trabajo = no: C (60.0/36.0)
| | | | trabajo = si: A (23.0/11.0)
| | | n_hijos > 1: C (81.0/46.0)
| | n_hijos > 2: B (197.0/99.0)

Number of Leaves : 10

Size of the tree : 19

=== Evaluation on test split ===

Time taken to test model on test split: 0 seconds

=== Summary ===

Correctly Classified Instances 261 52.0958 %


Incorrectly Classified Instances 240 47.9042 %
Total Number of Instances 501

=== Confusion Matrix ===

a b c <-- classified as
129 17 76 | a = A
22 46 41 | b = B
46 38 86 | c = C

También podría gustarte