Table 5. The validation results of the classifiers based on the top rank selected genes from lung cancer dataset.
In bold–the best performance.
| Dataset | Method | SVM with the top genes | ||||||
|---|---|---|---|---|---|---|---|---|
| 2 | 5 | 10 | ||||||
| GSE19804 | Lasso | 89.17% | 93.33% | 92.50% | ||||
| L1/2 | 85.83% | 90.83% | 91.67% | |||||
| SCAD − L2 | 89.17% | 89.17% | 93.33% | |||||
| ElasticNet | 86.67% | 87.50% | 89.17% | |||||
| HLR | 90.83% | 92.50% | 94.17% | |||||
| GSE32863 | Lasso | 93.10% | 95.69% | 93.97% | ||||
| L1/2 | 93.97% | 94.83% | 95.69% | |||||
| SCAD − L2 | 90.28% | 92.24% | 94.83% | |||||
| ElasticNet | 89.66% | 91.38% | 93.97% | |||||
| HLR | 94.83% | 96.55% | 97.41% | |||||
We used the SVM approach to build the classifiers based on the first two, first five and first ten genes selected by the different regularization approaches from the lung cancer dataset (Table 4), and were trained on the lung cancer dataset (Table 2) respectively. These classifiers then were applied to the two independent lung cancer datasets, GSE19804 and GSE32863, respectively.