Table 2 Prediction accuracy on three clustering groups. Since different distance threshold and different clustering groups generate distinct prediction accuracy, Table 2 shows a detailed report.
From: Protein local 3D structure prediction by Super Granule Support Vector Machines (Super GSVM)
Distance Threshold | Excellent | Good | Fair |
|---|---|---|---|
550 | 71.98% | 58.42% | 52.89% |
600 | 69.07% | 57.16% | 52.33% |
650 | 69.08% | 57.77% | 51.49% |
700 | 69.47% | 57.08% | 50.31% |
750 | 69.40% | 56.82% | 49.98% |
800 | 69.85% | 56.30% | 49.65% |
850 | 70.02% | 55.34% | 49.37% |
900 | 69.75% | 54.78% | 48.95% |
950 | 69.53% | 53.83% | 48.51% |
1000 | 68.88% | 53.06% | 48.15% |
1050 | 68.26% | 52.34% | 47.78% |
1100 | 67.63% | 51.56% | 47.46% |
1150 | 67.09% | 51.65% | 47.14% |
1200 | 66.54% | 50.75% | 46.92% |
1250 | 66.11% | 50.45% | 46.69% |
1300 | 65.73% | 50.20% | 46.48% |