data source: ../topic-classification-data/vectors_x2
vector size: 46551

MULTINOMIAL NAIVE BAYES

performance on training set:

accuracy: 60.206185567
f1 average: 0.540737408453
f1 by class: [ 0.66063348  0.53179191  0.59607843  0.45549738  0.88343558  0.08695652
  0.          0.96226415  0.95867769  0.51851852  0.47826087  0.10526316
  0.          0.          0.          0.          0.0952381   0.          0.
  0.          0.          0.14814815  0.13793103  0.8988764   0.125
  0.07407407  0.21052632  0.44444444  0.88888889  0.56666667  0.06451613
  0.          0.          0.94444444  0.94623656  0.87804878  0.95934959
  0.70731707  0.7761194   0.        ]

performance on test set:

accuracy: 41.4799728445
f1 average: 0.349654277152
f1 by class: [ 0.55084746  0.40909091  0.44217687  0.32926829  0.8969697   0.          0.
  0.49315068  0.63043478  0.13636364  0.15789474  0.          0.          0.
  0.          0.          0.          0.          0.          0.          0.
  0.          0.          0.21428571  0.          0.          0.
  0.05555556  0.73846154  0.08695652  0.          0.          0.
  0.60714286  0.72        0.63888889  0.95726496  0.38235294  0.50909091
  0.        ]

SUPPORT VECTOR MACHINE

session 0 training set:

accuracy: 99.9312714777
f1 average: 0.999301950945
f1 by class: [ 1.          1.          1.          0.99428571  1.          1.          1.
  1.          1.          1.          1.          1.          1.          1.
  1.          1.          1.          0.96296296  1.          1.          1.
  1.          1.          1.          1.          1.          1.          1.
  1.          1.          1.          1.          1.          1.          1.
  1.          1.          1.          1.          1.        ]

session 0 test set:

accuracy: 88.9341479973
f1 average: 0.887021553017
f1 by class: [ 0.91034483  0.81632653  0.7672956   0.88268156  0.96688742  0.93023256
  0.8         0.95495495  0.94308943  0.93975904  0.97142857  0.97297297
  0.97560976  1.          0.84210526  0.93023256  0.95238095  0.48        0.9
  1.          0.8         0.98039216  0.84745763  0.8411215   0.94736842
  0.48979592  0.82191781  0.82857143  0.96470588  0.95454545  0.93333333
  0.86486486  0.96969697  0.98734177  0.96969697  0.97674419  0.95867769
  0.7311828   0.96385542  0.48888889]

session 1 training set:

accuracy: 100.0
f1 average: 1.0
f1 by class: [ 1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.
  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.  1.
  1.  1.  1.  1.]

session 1 test set:

accuracy: 88.5947046843
f1 average: 0.882999921128
f1 by class: [ 0.89473684  0.82474227  0.7625      0.89772727  0.96052632  0.93023256
  0.8         0.95495495  0.92682927  0.92682927  0.97142857  0.97297297
  0.97560976  1.          0.87179487  0.93023256  0.97560976  0.32        0.9
  0.94117647  0.66666667  0.96        0.89285714  0.82692308  0.94736842
  0.56603774  0.81081081  0.8115942   0.95348837  0.94505495  0.93333333
  0.9         0.96969697  0.98734177  0.95833333  0.97674419  0.96666667
  0.7173913   0.96385542  0.41860465]

averages over sessions
train accuracy: 99.9656357388
train f1 score: 0.999650975472
test accuracy: 88.7644263408
test f1 score: 0.885010737072
