Votes Data Set
Total number of instances = 435
Number of classes: 2
probability of class 0 in the whole set: 0.386207
probability of class 1 in the whole set: 0.613793
feature 0, with number of values = 2
feature 1, with number of values = 2
feature 2, with number of values = 2
feature 3, with number of values = 2
feature 4, with number of values = 2
feature 5, with number of values = 2
feature 6, with number of values = 2
feature 7, with number of values = 2
feature 8, with number of values = 2
feature 9, with number of values = 2
feature 10, with number of values = 2
feature 11, with number of values = 2
feature 12, with number of values = 2
feature 13, with number of values = 2
feature 14, with number of values = 2
feature 15, with number of values = 2
Naive Bayes trained on the whole data set:
0/1 loss if no feature was used (error of random):0.386207
feature: 0, with gini loss: 0.396761, with 0/1 loss: 0.31716 (when used alone for prediction)
class: 0
value: 0, prob: 0.808383
value: 1, prob: 0.191617
class: 1
value: 0, prob: 0.396154
value: 1, prob: 0.603846
feature: 1, with gini loss: 0.474093, with 0/1 loss: 0.386207 (when used alone for prediction)
class: 0
value: 0, prob: 0.506667
value: 1, prob: 0.493333
class: 1
value: 0, prob: 0.502075
value: 1, prob: 0.497925
feature: 2, with gini loss: 0.213477, with 0/1 loss: 0.123792 (when used alone for prediction)
class: 0
value: 0, prob: 0.861446
value: 1, prob: 0.138554
class: 1
value: 0, prob: 0.114504
value: 1, prob: 0.885496
feature: 3, with gini loss: 0.0782587, with 0/1 loss: 0.0422133 (when used alone for prediction)
class: 0
value: 0, prob: 0.982036
value: 1, prob: 0.0179641
class: 1
value: 0, prob: 0.0574713
value: 1, prob: 0.942529
feature: 4, with gini loss: 0.235713, with 0/1 loss: 0.154558 (when used alone for prediction)
class: 0
value: 0, prob: 0.946108
value: 1, prob: 0.0538922
class: 1
value: 0, prob: 0.217899
value: 1, prob: 0.782101
feature: 5, with gini loss: 0.389962, with 0/1 loss: 0.334111 (when used alone for prediction)
class: 0
value: 0, prob: 0.892857
value: 1, prob: 0.107143
class: 1
value: 0, prob: 0.476923
value: 1, prob: 0.523077
feature: 6, with gini loss: 0.347355, with 0/1 loss: 0.235299 (when used alone for prediction)
class: 0
value: 0, prob: 0.756098
value: 1, prob: 0.243902
class: 1
value: 0, prob: 0.229885
value: 1, prob: 0.770115
feature: 7, with gini loss: 0.269003, with 0/1 loss: 0.16727 (when used alone for prediction)
class: 0
value: 0, prob: 0.842767
value: 1, prob: 0.157233
class: 1
value: 0, prob: 0.173585
value: 1, prob: 0.826415
feature: 8, with gini loss: 0.292049, with 0/1 loss: 0.196018 (when used alone for prediction)
class: 0
value: 0, prob: 0.88024
value: 1, prob: 0.11976
class: 1
value: 0, prob: 0.244
value: 1, prob: 0.756
feature: 9, with gini loss: 0.47084, with 0/1 loss: 0.386207 (when used alone for prediction)
class: 0
value: 0, prob: 0.556886
value: 1, prob: 0.443114
class: 1
value: 0, prob: 0.471698
value: 1, prob: 0.528302
feature: 10, with gini loss: 0.407875, with 0/1 loss: 0.356088 (when used alone for prediction)
class: 0
value: 0, prob: 0.863354
value: 1, prob: 0.136646
class: 1
value: 0, prob: 0.494163
value: 1, prob: 0.505837
feature: 11, with gini loss: 0.236471, with 0/1 loss: 0.142138 (when used alone for prediction)
class: 0
value: 0, prob: 0.866242
value: 1, prob: 0.133758
class: 1
value: 0, prob: 0.14741
value: 1, prob: 0.85259
feature: 12, with gini loss: 0.330589, with 0/1 loss: 0.234339 (when used alone for prediction)
class: 0
value: 0, prob: 0.85625
value: 1, prob: 0.14375
class: 1
value: 0, prob: 0.291339
value: 1, prob: 0.708661
feature: 13, with gini loss: 0.292809, with 0/1 loss: 0.225135 (when used alone for prediction)
class: 0
value: 0, prob: 0.97546
value: 1, prob: 0.0245399
class: 1
value: 0, prob: 0.351351
value: 1, prob: 0.648649
feature: 14, with gini loss: 0.339472, with 0/1 loss: 0.259863 (when used alone for prediction)
class: 0
value: 0, prob: 0.905063
value: 1, prob: 0.0949367
class: 1
value: 0, prob: 0.363636
value: 1, prob: 0.636364
feature: 15, with gini loss: 0.4154, with 0/1 loss: 0.295792 (when used alone for prediction)
class: 0
value: 0, prob: 0.655405
value: 1, prob: 0.344595
class: 1
value: 0, prob: 0.930481
value: 1, prob: 0.0695187