Ensembling Methods for Selecting a Splitting Attribute in Decision Trees Learning Algorithms
Table 3: Ensembling efficiency
Task number | Classification accuracy for groups: | |||||
All methods | 3 best | 3 average | 3 worst | BAW | Best method | |
1 | 0.412 | 0.632 | 0.548 | 0.409 | 0.416 | 0.691 |
2 | 0.32 | 0.52 | 0.417 | 0.325 | 0.324 | 0.563 |
3 | 0.164 | 0.309 | 0.258 | 0.163 | 0.163 | 0.745 |
4 | 0.864 | 0.883 | 0.864 | 0.83 | 0.854 | 0.883 |
5 | 0.649 | 0.952 | 0.932 | 0.649 | 0.649 | 0.956 |
6 | 0.444 | 0.691 | 0.693 | 0.442 | 0.471 | 0.752 |
7 | 0.892 | 0.952 | 0.895 | 0.847 | 0.933 | 0.957 |
8 | 0.199 | 0.868 | 0.246 | 0.199 | 0.199 | 0.932 |
9 | 0.683 | 0.757 | 0.694 | 0.614 | 0.611 | 0.766 |
10 | 0.541 | 0.827 | 0.543 | 0.541 | 0.541 | 0.831 |
11 | 0.697 | 0.72 | 0.77 | 0.711 | 0.715 | 0.799 |
12 | 0.209 | 0.378 | 0.234 | 0.209 | 0.209 | 0.426 |