Table 3 The total accuracy obtained from running seven SVM methods and X-wrapper validation on the 12 datasets derived from attribute weighting models
From: Prediction of lung tumor types based on protein attributes by machine learning algorithms
SVM models | SVM | SVM Linear | SVM Lib SVM | SVM Evolutionary | SVM POS | SVM Hyper | SVM Fast Large Margin |
|---|---|---|---|---|---|---|---|
Dataset | |||||||
SAM | 70.98% | 68.86% | 51.67% | 41.97% | 45.68% | 41.82% | 66.52% |
MR | 70.38% | 70.00% | 51.67% | 47.35% | 40.15% | 35.15% | 69.32% |
Chi Squared | 69.02% | 70.83% | 51.67% | 43.79% | 44.39% | 32.58% | 71.89% |
Deviation | 67.80% | 71.14% | 51.67% | 47.42% | 39.24% | 40.38% | 67.12% |
Gini Index | 63.94% | 68.18% | 51.67% | 45.76% | 47.12% | 32.73% | 68.56% |
Info Gain | 70.00% | 70.98% | 51.67% | 45.83% | 49.17% | 42.95% | 70.30% |
Info Gain Ratio | 71.97% | 68.48% | 51.67% | 46.59% | 43.79% | 23.86% | 67.50% |
PCA | 70.15% | 68.26% | 51.67% | 45.68% | 41.29% | 29.77% | 72.73% |
Relief | 70.83% | 66.74% | 51.67% | 45.83% | 37.95% | 28.26% | 70.23% |
Rule | 71.89% | 68.33% | 51.67% | 47.58% | 43.71% | 28.03% | 72.88% |
SVM | 70.00% | 68.26% | 51.67% | 47.35% | 44.02% | 33.03% | 67.42% |
Uncertainty | 67.42% | 69.17% | 51.67% | 47.35% | 43.48% | 29.92% | 66.44% |