Research Article
Applying Randomness Effectively Based on Random Forests for Classification Task of Datasets of Insufficient Information
Table 7
The accuracy of trial random forests (RF) for each class with different parameters.
| Bridges | (, ) | (, : def.) | (, ) |
| Accuracy per class | | | | 1 | 0% | 0% | 0% | 2 | 0% | 100% | 100% | 3 | 0% | 0% | 27.3% | 4 | 100% | 100% | 93.2% | 5 | 0% | 15.4% | 23% | 6 | 0% | 0% | 0% | 7 | 0% | 50% | 60% | Accuracy of RF | 41.5094% | 63.2075% | 65.0943% |
| Dermatology | (, ) | (, : def.) | (, ) |
| Accuracy per class | | | | 1 | 97.3% | 100% | 100% | 2 | 77% | 85.2% | 85.2% | 3 | 97.2% | 100% | 100% | 4 | 87.8% | 87.8% | 87.8% | 5 | 90.4% | 100% | 100% | 6 | 85.0% | 95.0% | 100% | Accuracy of RF | 90.9836% | 95.6284% | 96.4881% |
| Post Operative | (, ) | (, : def.) | (, ) |
| Accuracy per class | | | | 1 | 0% | 0% | 0% | 2 | 26.7% | 4.2% | 0% | 3 | 58.1% | 85.9% | 92.2% | Accuracy of RF | 52.2222% | 62.2222% | 65.5556% |
|
|