This repository contains the datasets and code related to the work reported in the paper referenced below.
M. R. H. Maia, A. Plastino and A. A. Freitas, "An Ensemble of Naive Bayes Classifiers for Uncertain Categorical Data", 2021 IEEE International Conference on Data Mining (ICDM), 2021, pp. 1222-1227, DOI: 10.1109/ICDM51629.2021.00148.
The code was implemented and tested using Python 3.9.
Dependencies are specified in file requirements.txt.
Assuming a Python environment with all required dependencies is available, follow the steps below to run the experiments:
- Change the working directory to
code(e.g., withcd code) - Run
eval.py(python eval.py)
The results produced should be the following:
dataset = AG-C.elegans.csv | model = NB-NV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.765957 0.631579 0.692308 57.0 0.300000 0.450000 0.360000 20.0 0.584416 0.540789 0.546053 0.533114
1 0.700000 0.686275 0.693069 51.0 0.407407 0.423077 0.415094 26.0 0.597403 0.554676 0.537707 0.538838
2 0.680000 0.693878 0.686869 49.0 0.444444 0.428571 0.436364 28.0 0.597403 0.561224 0.567420 0.545322
3 0.860465 0.685185 0.762887 54.0 0.484848 0.727273 0.581818 22.0 0.697368 0.706229 0.707912 0.705915
4 0.808511 0.730769 0.767677 52.0 0.517241 0.625000 0.566038 24.0 0.697368 0.677885 0.670673 0.675819
5 0.711538 0.787234 0.747475 47.0 0.583333 0.482759 0.528302 29.0 0.671053 0.634996 0.623991 0.616477
6 0.734694 0.765957 0.750000 47.0 0.592593 0.551724 0.571429 29.0 0.684211 0.658841 0.676449 0.650075
7 0.772727 0.680000 0.723404 50.0 0.500000 0.615385 0.551724 26.0 0.657895 0.647692 0.640000 0.646886
8 0.804348 0.740000 0.770833 50.0 0.566667 0.653846 0.607143 26.0 0.710526 0.696923 0.696923 0.695591
9 0.800000 0.653061 0.719101 49.0 0.527778 0.703704 0.603175 27.0 0.671053 0.678382 0.685185 0.677910
mean 0.763824 0.705394 0.733447 50.6 0.492431 0.566134 0.526717 25.7 0.656869 0.635764 0.635231 0.631939
dataset = AG-C.elegans.csv | model = ENB-NV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.920000 0.403509 0.560976 57.0 0.346154 0.900000 0.500000 20.0 0.532468 0.651754 0.665789 0.602626
1 0.947368 0.352941 0.514286 51.0 0.431034 0.961538 0.595238 26.0 0.558442 0.657240 0.651584 0.582552
2 0.882353 0.306122 0.454545 49.0 0.433333 0.928571 0.590909 28.0 0.532468 0.617347 0.596210 0.533157
3 0.947368 0.333333 0.493151 54.0 0.368421 0.954545 0.531646 22.0 0.513158 0.643939 0.750000 0.564076
4 0.947368 0.346154 0.507042 52.0 0.403509 0.958333 0.567901 24.0 0.539474 0.652244 0.736378 0.575961
5 0.842105 0.340426 0.484848 47.0 0.456140 0.896552 0.604651 29.0 0.552632 0.618489 0.684519 0.552457
6 1.000000 0.404255 0.575758 47.0 0.508772 1.000000 0.674419 29.0 0.631579 0.702128 0.812913 0.635811
7 1.000000 0.320000 0.484848 50.0 0.433333 1.000000 0.604651 26.0 0.552632 0.660000 0.718462 0.565685
8 0.863636 0.380000 0.527778 50.0 0.425926 0.884615 0.575000 26.0 0.552632 0.632308 0.770000 0.579788
9 0.947368 0.367347 0.529412 49.0 0.456140 0.962963 0.619048 27.0 0.578947 0.665155 0.760393 0.594762
mean 0.929757 0.355409 0.514243 50.6 0.426276 0.944712 0.587472 25.7 0.554443 0.650060 0.714625 0.579447
dataset = AG-C.elegans.csv | model = ENB-NV+BRS
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.862069 0.438596 0.581395 57.0 0.333333 0.800000 0.470588 20.0 0.532468 0.619298 0.661404 0.592349
1 0.923077 0.470588 0.623377 51.0 0.470588 0.923077 0.623377 26.0 0.623377 0.696833 0.705882 0.659082
2 0.791667 0.387755 0.520548 49.0 0.433962 0.821429 0.567901 28.0 0.545455 0.604592 0.631195 0.564370
3 0.920000 0.425926 0.582278 54.0 0.392157 0.909091 0.547945 22.0 0.565789 0.667508 0.805556 0.622258
4 0.826087 0.365385 0.506667 52.0 0.377358 0.833333 0.519481 24.0 0.513158 0.599359 0.724359 0.551804
5 0.740741 0.425532 0.540541 47.0 0.448980 0.758621 0.564103 29.0 0.552632 0.592076 0.680117 0.568170
6 0.920000 0.489362 0.638889 47.0 0.529412 0.931034 0.675000 29.0 0.657895 0.710198 0.822450 0.674991
7 0.909091 0.400000 0.555556 50.0 0.444444 0.923077 0.600000 26.0 0.578947 0.661538 0.730000 0.607644
8 0.892857 0.500000 0.641026 50.0 0.479167 0.884615 0.621622 26.0 0.631579 0.692308 0.746154 0.665062
9 0.941176 0.326531 0.484848 49.0 0.440678 0.962963 0.604651 27.0 0.552632 0.644747 0.725624 0.560747
mean 0.872676 0.422967 0.569777 50.6 0.435008 0.874724 0.581053 25.7 0.575393 0.648846 0.723274 0.608260
dataset = AG-C.elegans.csv | model = NB-EV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.867925 0.807018 0.836364 57.0 0.541667 0.650000 0.590909 20.0 0.766234 0.728509 0.724561 0.724266
1 0.803922 0.803922 0.803922 51.0 0.615385 0.615385 0.615385 26.0 0.740260 0.709653 0.845023 0.703364
2 0.825000 0.673469 0.741573 49.0 0.567568 0.750000 0.646154 28.0 0.701299 0.711735 0.810496 0.710705
3 0.895833 0.796296 0.843137 54.0 0.607143 0.772727 0.680000 22.0 0.789474 0.784512 0.853535 0.784423
4 0.844444 0.730769 0.783505 52.0 0.548387 0.708333 0.618182 24.0 0.723684 0.719551 0.816907 0.719464
5 0.680000 0.723404 0.701031 47.0 0.500000 0.448276 0.472727 29.0 0.618421 0.585840 0.689288 0.569460
6 0.683333 0.872340 0.766355 47.0 0.625000 0.344828 0.444444 29.0 0.671053 0.608584 0.785767 0.548459
7 0.760000 0.760000 0.760000 50.0 0.538462 0.538462 0.538462 26.0 0.684211 0.649231 0.756154 0.639711
8 0.692308 0.720000 0.705882 50.0 0.416667 0.384615 0.400000 26.0 0.605263 0.552308 0.675769 0.526235
9 0.750000 0.612245 0.674157 49.0 0.472222 0.629630 0.539683 27.0 0.618421 0.620937 0.731293 0.620876
mean 0.780276 0.749946 0.764811 50.6 0.543250 0.584226 0.562993 25.7 0.691832 0.667086 0.768879 0.661920
dataset = AG-C.elegans.csv | model = ENB-EV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.739726 0.947368 0.830769 57.0 0.250000 0.050000 0.083333 20.0 0.714286 0.498684 0.722807 0.217643
1 0.694444 0.980392 0.813008 51.0 0.800000 0.153846 0.258065 26.0 0.701299 0.567119 0.849170 0.388368
2 0.666667 0.938776 0.779661 49.0 0.625000 0.178571 0.277778 28.0 0.662338 0.558673 0.809767 0.409437
3 0.742857 0.962963 0.838710 54.0 0.666667 0.181818 0.285714 22.0 0.736842 0.572391 0.842593 0.418431
4 0.742857 1.000000 0.852459 52.0 1.000000 0.250000 0.400000 24.0 0.763158 0.625000 0.840946 0.500000
5 0.637681 0.936170 0.758621 47.0 0.571429 0.137931 0.222222 29.0 0.631579 0.537051 0.702494 0.359342
6 0.626667 1.000000 0.770492 47.0 1.000000 0.034483 0.066667 29.0 0.631579 0.517241 0.784666 0.185695
7 0.685714 0.960000 0.800000 50.0 0.666667 0.153846 0.250000 26.0 0.684211 0.556923 0.753846 0.384308
8 0.662162 0.980000 0.790323 50.0 0.500000 0.038462 0.071429 26.0 0.657895 0.509231 0.649231 0.194145
9 0.676471 0.938776 0.786325 49.0 0.625000 0.185185 0.285714 27.0 0.671053 0.561980 0.735450 0.416950
mean 0.687525 0.964444 0.802774 50.6 0.670476 0.136414 0.226704 25.7 0.685424 0.550429 0.769097 0.362717
dataset = AG-C.elegans.csv | model = ENB-EV+BRS
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.742857 0.912281 0.818898 57.0 0.285714 0.100000 0.148148 20.0 0.701299 0.506140 0.708772 0.302040
1 0.691176 0.921569 0.789916 51.0 0.555556 0.192308 0.285714 26.0 0.675325 0.556938 0.801659 0.420981
2 0.698413 0.897959 0.785714 49.0 0.642857 0.321429 0.428571 28.0 0.688312 0.609694 0.800292 0.537243
3 0.827586 0.888889 0.857143 54.0 0.666667 0.545455 0.600000 22.0 0.789474 0.717172 0.828283 0.696311
4 0.793651 0.961538 0.869565 52.0 0.846154 0.458333 0.594595 24.0 0.802632 0.709936 0.802885 0.663856
5 0.606061 0.851064 0.707965 47.0 0.300000 0.103448 0.153846 29.0 0.565789 0.477256 0.629494 0.296717
6 0.642857 0.957447 0.769231 47.0 0.666667 0.137931 0.228571 29.0 0.644737 0.547689 0.753852 0.363403
7 0.712121 0.940000 0.810345 50.0 0.700000 0.269231 0.388889 26.0 0.710526 0.604615 0.761538 0.503068
8 0.666667 0.880000 0.758621 50.0 0.400000 0.153846 0.222222 26.0 0.631579 0.516923 0.691538 0.367946
9 0.714286 0.816327 0.761905 49.0 0.550000 0.407407 0.468085 27.0 0.671053 0.611867 0.702192 0.576695
mean 0.709567 0.902707 0.794569 50.6 0.561361 0.268939 0.363656 25.7 0.688072 0.585823 0.748050 0.492720
dataset = AG-D.melanogaster.csv | model = NB-NV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.000000 0.000000 0.000000 3.0 0.823529 0.875000 0.848485 16.0 0.736842 0.437500 0.437500 0.000000
1 0.600000 0.500000 0.545455 6.0 0.785714 0.846154 0.814815 13.0 0.736842 0.673077 0.660256 0.650444
2 0.400000 0.181818 0.250000 11.0 0.357143 0.625000 0.454545 8.0 0.368421 0.403409 0.409091 0.337100
3 0.600000 0.300000 0.400000 10.0 0.500000 0.777778 0.608696 9.0 0.526316 0.538889 0.538889 0.483046
4 0.666667 0.250000 0.363636 8.0 0.625000 0.909091 0.740741 11.0 0.631579 0.579545 0.573864 0.476731
5 1.000000 0.428571 0.600000 7.0 0.733333 1.000000 0.846154 11.0 0.777778 0.714286 0.714286 0.654654
6 0.800000 0.666667 0.727273 6.0 0.846154 0.916667 0.880000 12.0 0.833333 0.791667 0.819444 0.781736
7 0.333333 0.166667 0.222222 6.0 0.666667 0.833333 0.740741 12.0 0.611111 0.500000 0.506944 0.372678
8 0.750000 0.750000 0.750000 4.0 0.928571 0.928571 0.928571 14.0 0.888889 0.839286 0.848214 0.834523
9 0.600000 0.375000 0.461538 8.0 0.615385 0.800000 0.695652 10.0 0.611111 0.587500 0.575000 0.547723
mean 0.575000 0.361872 0.444194 6.9 0.688150 0.851159 0.761023 11.6 0.672222 0.606516 0.608349 0.554987
dataset = AG-D.melanogaster.csv | model = ENB-NV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.222222 0.666667 0.333333 3.0 0.900000 0.562500 0.692308 16.0 0.578947 0.614583 0.541667 0.612372
1 0.357143 0.833333 0.500000 6.0 0.800000 0.307692 0.444444 13.0 0.473684 0.570513 0.807692 0.506370
2 0.615385 0.727273 0.666667 11.0 0.500000 0.375000 0.428571 8.0 0.578947 0.551136 0.556818 0.522233
3 0.538462 0.700000 0.608696 10.0 0.500000 0.333333 0.400000 9.0 0.526316 0.516667 0.544444 0.483046
4 0.416667 0.625000 0.500000 8.0 0.571429 0.363636 0.444444 11.0 0.473684 0.494318 0.443182 0.476731
5 0.400000 0.571429 0.470588 7.0 0.625000 0.454545 0.526316 11.0 0.500000 0.512987 0.623377 0.509647
6 0.428571 1.000000 0.600000 6.0 1.000000 0.333333 0.500000 12.0 0.555556 0.666667 0.875000 0.577350
7 0.444444 0.666667 0.533333 6.0 0.777778 0.583333 0.666667 12.0 0.611111 0.625000 0.611111 0.623610
8 0.400000 1.000000 0.571429 4.0 1.000000 0.571429 0.727273 14.0 0.666667 0.785714 0.946429 0.755929
9 0.538462 0.875000 0.666667 8.0 0.800000 0.400000 0.533333 10.0 0.611111 0.637500 0.612500 0.591608
mean 0.436136 0.766537 0.555952 6.9 0.747421 0.428480 0.544697 11.6 0.557602 0.597509 0.656222 0.573102
dataset = AG-D.melanogaster.csv | model = ENB-NV+BRS
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.200000 0.666667 0.307692 3.0 0.888889 0.500000 0.640000 16.0 0.526316 0.583333 0.541667 0.577350
1 0.428571 1.000000 0.600000 6.0 1.000000 0.384615 0.555556 13.0 0.578947 0.692308 0.769231 0.620174
2 0.692308 0.818182 0.750000 11.0 0.666667 0.500000 0.571429 8.0 0.684211 0.659091 0.625000 0.639602
3 0.615385 0.800000 0.695652 10.0 0.666667 0.444444 0.533333 9.0 0.631579 0.622222 0.500000 0.596285
4 0.384615 0.625000 0.476190 8.0 0.500000 0.272727 0.352941 11.0 0.421053 0.448864 0.431818 0.412861
5 0.400000 0.571429 0.470588 7.0 0.625000 0.454545 0.526316 11.0 0.500000 0.512987 0.636364 0.509647
6 0.428571 1.000000 0.600000 6.0 1.000000 0.333333 0.500000 12.0 0.555556 0.666667 0.833333 0.577350
7 0.500000 0.666667 0.571429 6.0 0.800000 0.666667 0.727273 12.0 0.666667 0.666667 0.625000 0.666667
8 0.363636 1.000000 0.533333 4.0 1.000000 0.500000 0.666667 14.0 0.611111 0.750000 0.928571 0.707107
9 0.538462 0.875000 0.666667 8.0 0.800000 0.400000 0.533333 10.0 0.611111 0.637500 0.612500 0.591608
mean 0.455155 0.802294 0.580808 6.9 0.794722 0.445633 0.571053 11.6 0.578655 0.623964 0.650348 0.597937
dataset = AG-D.melanogaster.csv | model = NB-EV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.000000 0.000000 0.000000 3.0 0.823529 0.875000 0.848485 16.0 0.736842 0.437500 0.583333 0.000000
1 0.166667 0.166667 0.166667 6.0 0.615385 0.615385 0.615385 13.0 0.473684 0.391026 0.512821 0.320256
2 0.600000 0.272727 0.375000 11.0 0.428571 0.750000 0.545455 8.0 0.473684 0.511364 0.750000 0.452267
3 0.800000 0.400000 0.533333 10.0 0.571429 0.888889 0.695652 9.0 0.631579 0.644444 0.661111 0.596285
4 0.666667 0.500000 0.571429 8.0 0.692308 0.818182 0.750000 11.0 0.684211 0.659091 0.659091 0.639602
5 1.000000 0.142857 0.250000 7.0 0.647059 1.000000 0.785714 11.0 0.666667 0.571429 0.714286 0.377964
6 0.666667 0.333333 0.444444 6.0 0.733333 0.916667 0.814815 12.0 0.722222 0.625000 0.833333 0.552771
7 0.600000 0.500000 0.545455 6.0 0.769231 0.833333 0.800000 12.0 0.722222 0.666667 0.729167 0.645497
8 0.000000 0.000000 0.000000 4.0 0.764706 0.928571 0.838710 14.0 0.722222 0.464286 0.446429 0.000000
9 1.000000 0.375000 0.545455 8.0 0.666667 1.000000 0.800000 10.0 0.722222 0.687500 0.775000 0.612372
mean 0.550000 0.269058 0.361347 6.9 0.671222 0.862603 0.754973 11.6 0.655556 0.565831 0.666457 0.481758
dataset = AG-D.melanogaster.csv | model = ENB-EV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.000000 0.000000 0.000000 3.0 0.842105 1.000000 0.914286 16.0 0.842105 0.500000 0.562500 0.000000
1 1.000000 0.166667 0.285714 6.0 0.722222 1.000000 0.838710 13.0 0.736842 0.583333 0.474359 0.408248
2 0.500000 0.090909 0.153846 11.0 0.411765 0.875000 0.560000 8.0 0.421053 0.482955 0.670455 0.282038
3 0.666667 0.200000 0.307692 10.0 0.500000 0.888889 0.640000 9.0 0.526316 0.544444 0.633333 0.421637
4 1.000000 0.250000 0.400000 8.0 0.647059 1.000000 0.785714 11.0 0.684211 0.625000 0.647727 0.500000
5 0.000000 0.000000 0.000000 7.0 0.611111 1.000000 0.758621 11.0 0.611111 0.500000 0.649351 0.000000
6 0.000000 0.000000 0.000000 6.0 0.666667 1.000000 0.800000 12.0 0.666667 0.500000 0.791667 0.000000
7 0.000000 0.000000 0.000000 6.0 0.647059 0.916667 0.758621 12.0 0.611111 0.458333 0.722222 0.000000
8 0.000000 0.000000 0.000000 4.0 0.764706 0.928571 0.838710 14.0 0.722222 0.464286 0.428571 0.000000
9 1.000000 0.125000 0.222222 8.0 0.588235 1.000000 0.740741 10.0 0.611111 0.562500 0.825000 0.353553
mean 0.416667 0.083258 0.138784 6.9 0.640093 0.960913 0.768359 11.6 0.643275 0.522085 0.640519 0.282848
dataset = AG-D.melanogaster.csv | model = ENB-EV+BRS
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.000000 0.000000 0.000000 3.0 0.842105 1.000000 0.914286 16.0 0.842105 0.500000 0.666667 0.000000
1 0.333333 0.166667 0.222222 6.0 0.687500 0.846154 0.758621 13.0 0.631579 0.506410 0.551282 0.375534
2 0.500000 0.090909 0.153846 11.0 0.411765 0.875000 0.560000 8.0 0.421053 0.482955 0.704545 0.282038
3 0.666667 0.200000 0.307692 10.0 0.500000 0.888889 0.640000 9.0 0.526316 0.544444 0.677778 0.421637
4 1.000000 0.250000 0.400000 8.0 0.647059 1.000000 0.785714 11.0 0.684211 0.625000 0.647727 0.500000
5 0.000000 0.000000 0.000000 7.0 0.611111 1.000000 0.758621 11.0 0.611111 0.500000 0.727273 0.000000
6 0.000000 0.000000 0.000000 6.0 0.666667 1.000000 0.800000 12.0 0.666667 0.500000 0.861111 0.000000
7 0.000000 0.000000 0.000000 6.0 0.647059 0.916667 0.758621 12.0 0.611111 0.458333 0.708333 0.000000
8 0.000000 0.000000 0.000000 4.0 0.764706 0.928571 0.838710 14.0 0.722222 0.464286 0.642857 0.000000
9 1.000000 0.125000 0.222222 8.0 0.588235 1.000000 0.740741 10.0 0.611111 0.562500 0.712500 0.353553
mean 0.350000 0.083258 0.134517 6.9 0.636621 0.945528 0.760918 11.6 0.632749 0.514393 0.690007 0.280575
dataset = AG-M.musculus.csv | model = NB-NV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.500000 0.666667 0.571429 3.0 0.800000 0.666667 0.727273 6.0 0.666667 0.666667 0.694444 0.666667
1 0.666667 1.000000 0.800000 2.0 1.000000 0.857143 0.923077 7.0 0.888889 0.928571 1.000000 0.925820
2 1.000000 0.666667 0.800000 3.0 0.833333 1.000000 0.909091 5.0 0.875000 0.833333 0.833333 0.816497
3 1.000000 0.250000 0.400000 4.0 0.571429 1.000000 0.727273 4.0 0.625000 0.625000 0.625000 0.500000
4 0.666667 0.500000 0.571429 4.0 0.600000 0.750000 0.666667 4.0 0.625000 0.625000 0.687500 0.612372
5 1.000000 0.600000 0.750000 5.0 0.600000 1.000000 0.750000 3.0 0.750000 0.800000 0.800000 0.774597
6 1.000000 0.333333 0.500000 3.0 0.714286 1.000000 0.833333 5.0 0.750000 0.666667 0.666667 0.577350
7 0.000000 0.000000 0.000000 3.0 0.571429 0.800000 0.666667 5.0 0.500000 0.400000 0.400000 0.000000
8 0.000000 0.000000 0.000000 1.0 0.800000 0.571429 0.666667 7.0 0.500000 0.285714 0.285714 0.000000
9 1.000000 0.666667 0.800000 3.0 0.833333 1.000000 0.909091 5.0 0.875000 0.833333 0.800000 0.816497
mean 0.683333 0.468333 0.555765 3.1 0.732381 0.864524 0.792985 5.1 0.705556 0.666429 0.679266 0.636306
dataset = AG-M.musculus.csv | model = ENB-NV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.50 1.000000 0.666667 3.0 1.000000 0.500000 0.666667 6.0 0.666667 0.750000 0.888889 0.707107
1 0.40 1.000000 0.571429 2.0 1.000000 0.571429 0.727273 7.0 0.666667 0.785714 1.000000 0.755929
2 0.60 1.000000 0.750000 3.0 1.000000 0.600000 0.750000 5.0 0.750000 0.800000 0.800000 0.774597
3 0.50 0.750000 0.600000 4.0 0.500000 0.250000 0.333333 4.0 0.500000 0.500000 0.687500 0.433013
4 0.50 0.750000 0.600000 4.0 0.500000 0.250000 0.333333 4.0 0.500000 0.500000 0.687500 0.433013
5 0.75 0.600000 0.666667 5.0 0.500000 0.666667 0.571429 3.0 0.625000 0.633333 0.733333 0.632456
6 0.50 0.666667 0.571429 3.0 0.750000 0.600000 0.666667 5.0 0.625000 0.633333 0.666667 0.632456
7 0.25 0.333333 0.285714 3.0 0.500000 0.400000 0.444444 5.0 0.375000 0.366667 0.400000 0.365148
8 0.00 0.000000 0.000000 1.0 0.750000 0.428571 0.545455 7.0 0.375000 0.214286 0.142857 0.000000
9 0.50 0.333333 0.400000 3.0 0.666667 0.800000 0.727273 5.0 0.625000 0.566667 0.666667 0.516398
mean 0.45 0.643333 0.529573 3.1 0.716667 0.506667 0.593642 5.1 0.570833 0.575000 0.667341 0.570925
dataset = AG-M.musculus.csv | model = ENB-NV+BRS
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.50 1.000000 0.666667 3.0 1.000000 0.500000 0.666667 6.0 0.666667 0.750000 1.000000 0.707107
1 0.40 1.000000 0.571429 2.0 1.000000 0.571429 0.727273 7.0 0.666667 0.785714 1.000000 0.755929
2 0.60 1.000000 0.750000 3.0 1.000000 0.600000 0.750000 5.0 0.750000 0.800000 0.800000 0.774597
3 0.50 0.750000 0.600000 4.0 0.500000 0.250000 0.333333 4.0 0.500000 0.500000 0.750000 0.433013
4 0.50 0.750000 0.600000 4.0 0.500000 0.250000 0.333333 4.0 0.500000 0.500000 0.625000 0.433013
5 0.75 0.600000 0.666667 5.0 0.500000 0.666667 0.571429 3.0 0.625000 0.633333 0.733333 0.632456
6 0.50 0.666667 0.571429 3.0 0.750000 0.600000 0.666667 5.0 0.625000 0.633333 0.666667 0.632456
7 0.25 0.333333 0.285714 3.0 0.500000 0.400000 0.444444 5.0 0.375000 0.366667 0.400000 0.365148
8 0.00 0.000000 0.000000 1.0 0.666667 0.285714 0.400000 7.0 0.250000 0.142857 0.142857 0.000000
9 0.50 0.333333 0.400000 3.0 0.666667 0.800000 0.727273 5.0 0.625000 0.566667 0.733333 0.516398
mean 0.45 0.643333 0.529573 3.1 0.708333 0.492381 0.580937 5.1 0.558333 0.567857 0.685119 0.562819
dataset = AG-M.musculus.csv | model = NB-EV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.500000 0.333333 0.400000 3.0 0.714286 0.833333 0.769231 6.0 0.666667 0.583333 0.722222 0.527046
1 0.666667 1.000000 0.800000 2.0 1.000000 0.857143 0.923077 7.0 0.888889 0.928571 0.857143 0.925820
2 1.000000 0.333333 0.500000 3.0 0.714286 1.000000 0.833333 5.0 0.750000 0.666667 0.733333 0.577350
3 0.333333 0.250000 0.285714 4.0 0.400000 0.500000 0.444444 4.0 0.375000 0.375000 0.562500 0.353553
4 1.000000 0.500000 0.666667 4.0 0.666667 1.000000 0.800000 4.0 0.750000 0.750000 0.687500 0.707107
5 0.666667 0.400000 0.500000 5.0 0.400000 0.666667 0.500000 3.0 0.500000 0.533333 0.666667 0.516398
6 0.666667 0.666667 0.666667 3.0 0.800000 0.800000 0.800000 5.0 0.750000 0.733333 0.633333 0.730297
7 0.000000 0.000000 0.000000 3.0 0.625000 1.000000 0.769231 5.0 0.625000 0.500000 0.800000 0.000000
8 0.000000 0.000000 0.000000 1.0 0.857143 0.857143 0.857143 7.0 0.750000 0.428571 0.357143 0.000000
9 0.000000 0.000000 0.000000 3.0 0.625000 1.000000 0.769231 5.0 0.625000 0.500000 0.666667 0.000000
mean 0.483333 0.348333 0.404876 3.1 0.680238 0.851429 0.756267 5.1 0.668056 0.599881 0.668651 0.544592
dataset = AG-M.musculus.csv | model = ENB-EV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.00 0.000000 0.000000 3.0 0.666667 1.000000 0.800000 6.0 0.666667 0.500000 0.833333 0.000000
1 0.50 0.500000 0.500000 2.0 0.857143 0.857143 0.857143 7.0 0.777778 0.678571 0.857143 0.654654
2 1.00 0.333333 0.500000 3.0 0.714286 1.000000 0.833333 5.0 0.750000 0.666667 0.666667 0.577350
3 0.50 0.250000 0.333333 4.0 0.500000 0.750000 0.600000 4.0 0.500000 0.500000 0.500000 0.433013
4 1.00 0.250000 0.400000 4.0 0.571429 1.000000 0.727273 4.0 0.625000 0.625000 0.687500 0.500000
5 1.00 0.200000 0.333333 5.0 0.428571 1.000000 0.600000 3.0 0.500000 0.600000 0.800000 0.447214
6 0.50 0.333333 0.400000 3.0 0.666667 0.800000 0.727273 5.0 0.625000 0.566667 0.533333 0.516398
7 0.00 0.000000 0.000000 3.0 0.625000 1.000000 0.769231 5.0 0.625000 0.500000 0.666667 0.000000
8 0.00 0.000000 0.000000 1.0 0.857143 0.857143 0.857143 7.0 0.750000 0.428571 0.714286 0.000000
9 0.00 0.000000 0.000000 3.0 0.625000 1.000000 0.769231 5.0 0.625000 0.500000 0.666667 0.000000
mean 0.45 0.186667 0.263874 3.1 0.651190 0.926429 0.764800 5.1 0.644444 0.556548 0.692560 0.415853
dataset = AG-M.musculus.csv | model = ENB-EV+BRS
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.000000 0.000000 0.000000 3.0 0.666667 1.000000 0.800000 6.0 0.666667 0.500000 0.833333 0.000000
1 0.666667 1.000000 0.800000 2.0 1.000000 0.857143 0.923077 7.0 0.888889 0.928571 0.857143 0.925820
2 1.000000 0.333333 0.500000 3.0 0.714286 1.000000 0.833333 5.0 0.750000 0.666667 0.666667 0.577350
3 0.500000 0.250000 0.333333 4.0 0.500000 0.750000 0.600000 4.0 0.500000 0.500000 0.500000 0.433013
4 1.000000 0.250000 0.400000 4.0 0.571429 1.000000 0.727273 4.0 0.625000 0.625000 0.625000 0.500000
5 1.000000 0.200000 0.333333 5.0 0.428571 1.000000 0.600000 3.0 0.500000 0.600000 0.733333 0.447214
6 0.500000 0.333333 0.400000 3.0 0.666667 0.800000 0.727273 5.0 0.625000 0.566667 0.600000 0.516398
7 0.000000 0.000000 0.000000 3.0 0.625000 1.000000 0.769231 5.0 0.625000 0.500000 0.733333 0.000000
8 0.000000 0.000000 0.000000 1.0 0.857143 0.857143 0.857143 7.0 0.750000 0.428571 0.714286 0.000000
9 0.000000 0.000000 0.000000 3.0 0.625000 1.000000 0.769231 5.0 0.625000 0.500000 0.666667 0.000000
mean 0.466667 0.236667 0.314060 3.1 0.665476 0.926429 0.774564 5.1 0.655556 0.581548 0.692976 0.468246
dataset = AG-S.cerevisiae.csv | model = NB-NV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.897436 1.000000 0.945946 35.0 0.0 0.0 0.0 4.0 0.897436 0.500000 0.500000 0.0
1 0.923077 1.000000 0.960000 36.0 0.0 0.0 0.0 3.0 0.923077 0.500000 0.500000 0.0
2 0.894737 1.000000 0.944444 34.0 0.0 0.0 0.0 4.0 0.894737 0.500000 0.500000 0.0
3 0.842105 1.000000 0.914286 32.0 0.0 0.0 0.0 6.0 0.842105 0.500000 0.500000 0.0
4 0.842105 1.000000 0.914286 32.0 0.0 0.0 0.0 6.0 0.842105 0.500000 0.500000 0.0
5 0.837838 0.968750 0.898551 32.0 0.0 0.0 0.0 6.0 0.815789 0.484375 0.484375 0.0
6 0.894737 1.000000 0.944444 34.0 0.0 0.0 0.0 4.0 0.894737 0.500000 0.500000 0.0
7 0.921053 1.000000 0.958904 35.0 0.0 0.0 0.0 3.0 0.921053 0.500000 0.500000 0.0
8 0.842105 1.000000 0.914286 32.0 0.0 0.0 0.0 6.0 0.842105 0.500000 0.500000 0.0
9 0.894737 1.000000 0.944444 34.0 0.0 0.0 0.0 4.0 0.894737 0.500000 0.500000 0.0
mean 0.878993 0.996875 0.934230 33.6 0.0 0.0 0.0 4.6 0.876788 0.498437 0.498437 0.0
dataset = AG-S.cerevisiae.csv | model = ENB-NV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.888889 0.457143 0.603774 35.0 0.095238 0.500000 0.160000 4.0 0.461538 0.478571 0.621429 0.478091
1 0.950000 0.527778 0.678571 36.0 0.105263 0.666667 0.181818 3.0 0.538462 0.597222 0.518519 0.593171
2 0.900000 0.529412 0.666667 34.0 0.111111 0.500000 0.181818 4.0 0.526316 0.514706 0.573529 0.514496
3 0.904762 0.593750 0.716981 32.0 0.235294 0.666667 0.347826 6.0 0.605263 0.630208 0.598958 0.629153
4 0.900000 0.562500 0.692308 32.0 0.222222 0.666667 0.333333 6.0 0.578947 0.614583 0.692708 0.612372
5 0.888889 0.500000 0.640000 32.0 0.200000 0.666667 0.307692 6.0 0.526316 0.583333 0.546875 0.577350
6 0.928571 0.382353 0.541667 34.0 0.125000 0.750000 0.214286 4.0 0.421053 0.566176 0.551471 0.535504
7 0.894737 0.485714 0.629630 35.0 0.052632 0.333333 0.090909 3.0 0.473684 0.409524 0.600000 0.402374
8 0.941176 0.500000 0.653061 32.0 0.238095 0.833333 0.370370 6.0 0.552632 0.666667 0.671875 0.645497
9 1.000000 0.529412 0.692308 34.0 0.200000 1.000000 0.333333 4.0 0.578947 0.764706 0.786765 0.727607
mean 0.919702 0.506806 0.653499 33.6 0.158486 0.658333 0.255470 4.6 0.526316 0.582570 0.616213 0.577622
dataset = AG-S.cerevisiae.csv | model = ENB-NV+BRS
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 1.000000 0.285714 0.444444 35.0 0.137931 1.000000 0.242424 4.0 0.358974 0.642857 0.650000 0.534522
1 0.937500 0.416667 0.576923 36.0 0.086957 0.666667 0.153846 3.0 0.435897 0.541667 0.444444 0.527046
2 0.944444 0.500000 0.653846 34.0 0.150000 0.750000 0.250000 4.0 0.526316 0.625000 0.610294 0.612372
3 0.857143 0.375000 0.521739 32.0 0.166667 0.666667 0.266667 6.0 0.421053 0.520833 0.578125 0.500000
4 0.933333 0.437500 0.595745 32.0 0.217391 0.833333 0.344828 6.0 0.500000 0.635417 0.666667 0.603807
5 1.000000 0.406250 0.577778 32.0 0.240000 1.000000 0.387097 6.0 0.500000 0.703125 0.578125 0.637377
6 0.923077 0.352941 0.510638 34.0 0.120000 0.750000 0.206897 4.0 0.394737 0.551471 0.588235 0.514496
7 1.000000 0.428571 0.600000 35.0 0.130435 1.000000 0.230769 3.0 0.473684 0.714286 0.590476 0.654654
8 0.933333 0.437500 0.595745 32.0 0.217391 0.833333 0.344828 6.0 0.500000 0.635417 0.598958 0.603807
9 1.000000 0.441176 0.612245 34.0 0.173913 1.000000 0.296296 4.0 0.500000 0.720588 0.816176 0.664211
mean 0.952883 0.408132 0.571488 33.6 0.164068 0.850000 0.275047 4.6 0.461066 0.629066 0.612150 0.588993
dataset = AG-S.cerevisiae.csv | model = NB-EV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.965517 0.800000 0.875000 35.0 0.300000 0.750000 0.428571 4.0 0.794872 0.775000 0.900000 0.774597
1 0.939394 0.861111 0.898551 36.0 0.166667 0.333333 0.222222 3.0 0.820513 0.597222 0.847222 0.535758
2 0.900000 0.794118 0.843750 34.0 0.125000 0.250000 0.166667 4.0 0.736842 0.522059 0.621324 0.445566
3 0.857143 0.750000 0.800000 32.0 0.200000 0.333333 0.250000 6.0 0.684211 0.541667 0.671875 0.500000
4 0.909091 0.937500 0.923077 32.0 0.600000 0.500000 0.545455 6.0 0.868421 0.718750 0.822917 0.684653
5 0.882353 0.937500 0.909091 32.0 0.500000 0.333333 0.400000 6.0 0.842105 0.635417 0.828125 0.559017
6 0.939394 0.911765 0.925373 34.0 0.400000 0.500000 0.444444 4.0 0.868421 0.705882 0.860294 0.675191
7 0.892857 0.714286 0.793651 35.0 0.000000 0.000000 0.000000 3.0 0.657895 0.357143 0.533333 0.000000
8 0.857143 0.937500 0.895522 32.0 0.333333 0.166667 0.222222 6.0 0.815789 0.552083 0.833333 0.395285
9 0.939394 0.911765 0.925373 34.0 0.400000 0.500000 0.444444 4.0 0.868421 0.705882 0.794118 0.675191
mean 0.908229 0.855554 0.881105 33.6 0.302500 0.366667 0.331507 4.6 0.795749 0.611111 0.771254 0.560092
dataset = AG-S.cerevisiae.csv | model = ENB-EV
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.916667 0.942857 0.929577 35.0 0.333333 0.25 0.285714 4.0 0.871795 0.596429 0.871429 0.485504
1 0.923077 1.000000 0.960000 36.0 0.000000 0.00 0.000000 3.0 0.923077 0.500000 0.861111 0.000000
2 0.894737 1.000000 0.944444 34.0 0.000000 0.00 0.000000 4.0 0.894737 0.500000 0.588235 0.000000
3 0.837838 0.968750 0.898551 32.0 0.000000 0.00 0.000000 6.0 0.815789 0.484375 0.656250 0.000000
4 0.842105 1.000000 0.914286 32.0 0.000000 0.00 0.000000 6.0 0.842105 0.500000 0.796875 0.000000
5 0.842105 1.000000 0.914286 32.0 0.000000 0.00 0.000000 6.0 0.842105 0.500000 0.781250 0.000000
6 0.918919 1.000000 0.957746 34.0 1.000000 0.25 0.400000 4.0 0.921053 0.625000 0.882353 0.500000
7 0.918919 0.971429 0.944444 35.0 0.000000 0.00 0.000000 3.0 0.894737 0.485714 0.523810 0.000000
8 0.842105 1.000000 0.914286 32.0 0.000000 0.00 0.000000 6.0 0.842105 0.500000 0.807292 0.000000
9 0.894737 1.000000 0.944444 34.0 0.000000 0.00 0.000000 4.0 0.894737 0.500000 0.786765 0.000000
mean 0.883121 0.988304 0.932756 33.6 0.133333 0.05 0.072727 4.6 0.874224 0.519152 0.755537 0.222295
dataset = AG-S.cerevisiae.csv | model = ENB-EV+BRS
precision(anti) recall(anti) f1-score(anti) support(anti) precision(pro) recall(pro) f1-score(pro) support(pro) accuracy b-accuracy roc-auc g-mean
0 0.914286 0.914286 0.914286 35.0 0.250000 0.250000 0.250000 4.0 0.846154 0.582143 0.871429 0.478091
1 0.942857 0.916667 0.929577 36.0 0.250000 0.333333 0.285714 3.0 0.871795 0.625000 0.824074 0.552771
2 0.888889 0.941176 0.914286 34.0 0.000000 0.000000 0.000000 4.0 0.842105 0.470588 0.588235 0.000000
3 0.878788 0.906250 0.892308 32.0 0.400000 0.333333 0.363636 6.0 0.815789 0.619792 0.656250 0.549621
4 0.864865 1.000000 0.927536 32.0 1.000000 0.166667 0.285714 6.0 0.868421 0.583333 0.838542 0.408248
5 0.885714 0.968750 0.925373 32.0 0.666667 0.333333 0.444444 6.0 0.868421 0.651042 0.885417 0.568258
6 0.942857 0.970588 0.956522 34.0 0.666667 0.500000 0.571429 4.0 0.921053 0.735294 0.838235 0.696631
7 0.909091 0.857143 0.882353 35.0 0.000000 0.000000 0.000000 3.0 0.789474 0.428571 0.457143 0.000000
8 0.842105 1.000000 0.914286 32.0 0.000000 0.000000 0.000000 6.0 0.842105 0.500000 0.859375 0.000000
9 0.942857 0.970588 0.956522 34.0 0.666667 0.500000 0.571429 4.0 0.921053 0.735294 0.860294 0.696631
mean 0.901231 0.944545 0.922380 33.6 0.390000 0.241667 0.298417 4.6 0.858637 0.593106 0.767899 0.477771