|
Summary of cross validation results. All 14 tumor types and their abbreviations are listed in the first column. The sample sizes for each tumor type and the number of successfully classified samples by the SVM algorithm (from ref. [19]), the NB (Naive Bayes) algorithm with cutoff σ2 = 0.06 and the NN (Nearest Neighbor) algorithm with σ2 = 0.05 are listed. |
||||
| Tumor type (Abbreviation) |
Sample Size |
SVM [19] |
NB |
NN |
|
|
||||
| Ovary (OV) |
8 |
3 |
1 |
4 |
| Lung (LU) |
8 |
4 |
1 |
3 |
| Bladder (BL) |
8 |
5 |
7 |
5 |
| Melanoma (ML) |
8 |
5 |
6 |
6 |
| Renal (RE) |
8 |
5 |
5 |
6 |
| Pancreas (PA) |
8 |
5 |
6 |
6 |
| Colorectal (CR) |
8 |
6 |
1 |
3 |
| Prostate (PR) |
8 |
6 |
6 |
6 |
| Breast (BR) |
8 |
7 |
3 |
4 |
| Uterus (UT) |
8 |
7 |
6 |
6 |
| Mesothelioma (ME) |
8 |
8 |
6 |
6 |
| Lymphoma (LY) |
16 |
16 |
14 |
16 |
| Central Nervous System (CNS) |
16 |
16 |
16 |
16 |
| Leukemia (LE) |
24 |
24 |
23 |
24 |
Wang and Huang BMC Bioinformatics 2004 5:121 doi:10.1186/1471-2105-5-121 |
||||