|
| 作者 |
方法 |
特征数 |
精度 |
灵敏度 |
|
| Krishnan et al。36] |
40%的测试数据,支持向量机(聚)。 |
30. |
92.62% |
92.69% |
| 40%的测试数据,支持向量机(RBF) |
93.72% |
94.50% |
|
| Bagui et al。37] |
64%的测试数据,-RNN |
30. |
96.00% |
95.09% |
| 64%的测试数据,-RNN |
最好3 |
98.10% |
98.05% |
|
| Sweilam et al。38] |
算法+支持向量机 |
30. |
93.52% |
91.52% |
| QPSO +支持向量机 |
93.06% |
90.00% |
|
| Mangasarian et al。39] |
10-CV, MSM-T |
最好3 |
97.50% |
- - - - - - |
|
| 莫特et al。40] |
10-CV,并通过 |
3(2集成电路+ DWT) |
96.31% |
98.88% |
| 厕所,并通过 |
97.01% |
97.78% |
|
| 郑et al。41] |
支持向量机 |
6 |
97.38% |
- - - - - - |
|
| 本研究 |
10-CV,神经网络 |
1功能减少了ICA |
91.03% |
94.67% |
| 40%的测试,神经网络 |
92.56% |
94.02% |
| 10-CV,安 |
90.50% |
96.91% |
| 40%的测试,安 |
90.89% |
97.00% |
| 10-CV, RBFNN |
90.49% |
96.63% |
| 40%的测试,RBFNN |
89.98% |
96.01% |
| 10-CV, SVM(线性) |
90.33% |
96.35% |
| 40%的测试,支持向量机(线性) |
90.01% |
95.00% |
| 10-CV, SVM(二次) |
89.98% |
95.24% |
| 40%的测试,支持向量机(二次) |
91.01% |
96.42% |
| 10-CV, SVM (RBF) |
90.86% |
97.47% |
| 40%的测试,支持向量机(RBF) |
91.03% |
97.56% |
|
|
