Skip to main content
. 2014 Apr 2;9(4):e93045. doi: 10.1371/journal.pone.0093045

Table 5. Mean classification accuracies (%) of OSWLDA, OPCALDA, and OLDA evaluated on data set A using Inline graphic cross-validation.

Algorithms Inline graphic Inline graphic Intensification sequences Inline graphic
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
OSWLDA 45 1 40.8 62.6 75.8 81.6 85.0 88.0 90.6 92.6 92.2 93.6 94.0 93.6 93.4 94.0 94.6
5 47.4 67.0 79.0 83.8 88.0 89.4 91.6 94.0 94.6 94.2 94.6 94.8 95.4 95.6 96.0
10 47.2 68.2 80.4 86.2 90.4 90.4 93.2 94.8 95.4 95.4 96.2 96.0 97.0 96.4 96.8
15 47.8 68.6 80.6 85.4 91.4 91.6 94.0 95.4 96.2 96.4 97.0 96.6 97.2 97.8 97.8
20 48.4 68.0 81.2 86.4 91.8 92.8 94.2 96.2 96.0 96.6 97.2 97.2 97.8 97.6 98.2
25 47.8 69.0 81.0 86.2 91.6 92.8 94.6 96.6 96.2 97.0 97.8 97.4 98.4 98.6 98.2
30 47.8 69.2 80.8 85.6 91.0 93.0 94.8 96.4 97.0 96.8 98.2 97.6 98.4 98.6 98.8
35 47.4 68.8 80.0 85.8 91.2 92.8 94.4 96.4 97.2 97.4 98.6 98.0 98.4 98.6 98.8
40 46.2 68.0 79.6 85.0 91.2 92.2 94.8 97.0 97.4 97.2 98.4 98.4 98.4 98.6 98.8
45 43.4 68.2 80.2 86.0 90.2 93.4 95.4 96.4 97.4 97.2 98.2 98.2 98.4 98.6 99.0
OPCALDA 45 1 46.0 63.8 76.6 83.2 87.8 88.8 91.2 92.8 92.6 92.8 93.6 94.2 93.8 94.0 94.2
5 46.0 68.8 78.8 83.8 89.2 90.0 91.2 93.2 94.2 94.2 95.4 95.0 94.6 94.4 95.4
10 46.4 67.6 79.2 84.6 90.0 90.0 92.2 93.4 95.6 95.6 96.0 96.0 95.8 96.4 96.8
15 47.4 68.2 80.0 86.0 90.6 92.0 93.8 95.2 96.2 96.4 96.8 96.8 96.8 97.2 98.0
20 47.0 68.2 80.2 85.4 90.4 93.2 94.2 95.6 96.4 96.6 97.0 97.0 97.4 97.8 98.0
25 47.0 66.6 80.2 85.4 91.0 92.4 93.8 95.6 96.6 97.0 97.4 97.2 97.8 98.2 98.2
30 47.4 67.0 80.6 85.4 91.0 92.2 94.4 95.4 97.0 96.8 97.6 97.4 98.0 98.2 98.6
35 47.0 67.4 80.2 85.4 91.2 92.0 94.0 95.6 97.2 97.2 98.0 97.8 98.0 98.2 98.6
40 46.6 67.0 80.6 85.4 91.4 92.2 94.2 96.0 97.6 97.4 98.4 98.0 98.0 98.2 98.6
45 46.8 67.2 80.4 86.0 91.4 91.8 94.2 96.2 97.4 97.2 98.4 98.0 98.0 98.2 98.6
OLDA 45 1 3.8 3.4 4.4 3.8 2.6 2.8 3.4 2.8 2.6 2.6 2.8 3.6 4.0 4.2 4.2
5 46.6 67.2 79.0 84.4 89.4 90.4 92.2 93.6 94.6 94.6 95.4 95.6 95.2 95.6 96.2
10 47.8 68.4 80.0 84.2 89.8 90.6 93.2 94.6 95.2 95.6 96.8 96.8 96.6 97.0 97.4
15 47.6 68.6 81.8 85.4 90.6 91.8 94.2 95.2 96.0 96.2 96.8 96.8 97.8 97.8 98.6
20 47.2 68.6 81.8 85.6 90.8 92.0 94.4 96.0 96.4 96.8 97.8 97.8 98.4 98.8 99.2
25 46.8 68.2 82.0 86.2 91.2 92.2 94.8 96.4 96.6 97.0 98.0 97.8 98.6 98.8 99.6
30 46.4 67.4 81.8 86.4 91.0 93.0 94.6 96.6 96.8 97.2 98.2 98.2 98.6 98.8 99.4
35 46.4 68.0 81.6 86.8 91.0 93.0 94.6 96.2 97.0 97.6 98.8 98.4 98.6 98.8 99.4
40 46.0 68.0 81.4 86.8 90.8 93.0 94.6 96.4 97.2 97.6 98.8 98.4 98.6 98.6 99.4
45 46.0 68.0 81.0 86.8 90.6 93.0 94.6 96.4 97.2 97.6 98.8 98.6 98.6 98.6 99.4

The best accuracy among all Inline graphic for each algorithm and each repetition is written in bold and the worst is underlined. An overlapped ensemble classifier becomes an ensemble classifier with naive partitioning when Inline graphic and Inline graphic. The classifier is equivalent to a single classifier when Inline graphic and Inline graphic.