Table 2. The classification results of LVQ-ANN with different number of competitive layer neuron based on LVQ1 and LVQ2 learning algorithm.
| Samples | Sample numbers | Learning algorithm | Identification rate of different numbers of competitive layer neuron numbers | ||||||||||
| 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | |||
| Sect. Furfuracea | 5 | LVQ1 | 80.00% | 20.00% | 80.00 | 20.00% | 60.00% | 20.00% | 40.00% | 60.00% | 20.00% | 40.00% | 20.00% |
| LVQ2 | 0.00% | 0.00% | 0.00% | 80.00 | 0.00% | 80.00% | 80.00% | 0.00% | 0.00% | 20.00% | 20.00% | ||
| Sect. Paracamellia | 8 | LVQ1 | 37.50% | 37.50% | 37.50% | 37.50% | 37.50% | 37.50% | 37.50% | 37.50% | 37.50% | 37.50% | 37.50% |
| LVQ2 | 0.00% | 75.00% | 75.00% | 75.00% | 0.00% | 75.00% | 75.00% | 0.00% | 75.00% | 75.00% | 75.00% | ||
| Sect. Tuberculata | 6 | LVQ1 | 0.00% | 16.67% | 16.67% | 16.67% | 16.67% | 16.67% | 16.67% | 0.00% | 16.67% | 0.00% | 16.67% |
| LVQ2 | 0.00% | 0.00% | 0.00% | 0.00% | 0.00% | 0.00% | 0.00% | 0.00% | 0.00% | 0.00% | 0.00% | ||
| Sect. Camellia | 16 | LVQ1 | 62.50% | 68.75% | 62.50% | 68.75% | 68.75% | 68.75% | 68.75% | 68.75% | 62.50% | 62.50% | 62.50% |
| LVQ2 | 75.00% | 75.00% | 87.50% | 68.75% | 81.25% | 93.75% | 81.25% | 81.25% | 81.25% | 75.00% | 75.00% | ||
| Sect. Theopsis | 10 | LVQ1 | 90.00% | 90.00% | 90.00% | 90.00% | 90.00% | 90.00% | 90.00% | 90.00% | 90.00% | 90.00% | 90.00% |
| LVQ2 | 100% | 0.00% | 0.00% | 0.00% | 100% | 0.00% | 0.00% | 100% | 0.00% | 0.00% | 0.00% | ||
| Total accuracy (%) | LVQ1 | 57.78% | 55.56% | 60.00% | 55.56% | 60.00% | 55.56% | 57.78% | 57.78% | 53.33% | 53.33% | 53.33% | |
| LVQ2 | 48.89% | 40.00% | 44.44% | 46.67% | 51.11% | 55.56% | 51.11% | 51.11% | 42.22% | 48.89% | 42.22% | ||