Research Article
The Study of Misclassification Probability in Discriminant Model of Pattern Identification for Stroke
Table 9
Discriminant rate distribution of the selected data according to four measures.
| | Discriminant rate | | | | | | |
| 100% | 3306 | 66.82 | 66.82 | 66.82 | 66.82 | 90% | 2975 | 68.24 (+1.42) | 67.63 (+0.81) | 66.92 (+0.10) | 67.53 (+0.71) | 80% | 2645 | 68.62 (+0.38) | 68.47 (+0.84) | 67.15 (+0.23) | 69.04 (+1.51) | 70% | 2314 | 69.53 (+0.91) | 69.97 (+1.50) | 66.98 (−0.17) | 69.49 (+0.45) | 60% | 1984 | 71.98 (+2.45) | 70.82 (+0.85) | 66.94 (−0.04) | 71.22 (+1.73) | 50% | 1653 | 73.32 (+1.34) | 73.08 (+2.26) |
69.03 (+2.09) | 71.81 (+0.59) | 40% | 1322 | 75.34 (+2.02) | 74.28 (+1.20) | 68.68 (−0.35) | 73.75 (+1.94) | 30% | 992 | 77.32 (+1.98) | 76.81 (+2.53) |
70.26 (+1.58) | 75.81 (+2.06) | 20% | 661 | 83.36 (+6.04) | 80.94 (+4.13) |
73.83 (+3.57) | 77.61 (+1.80) | 10% | 331 | 89.12 (+5.76) | 87.01 (+6.07) |
75.83 (+2.00) | 82.78 (+5.17) |
|
|