Table 1.
Accuracy values for each algorithm/test. Kappa, sensitivity and specificity values are also included.
| Complete set (intrinsic and extrinsic variables) | |||||
|---|---|---|---|---|---|
| accuracy | 95%CI | kappa | sensitivity* | specificity* | |
| NN | 100 | 0.99-1 | 1 | (1.0,1.0,1.0,1.0) | (1.0,1.0,1.0,1.0) |
| SVM | 100 | 0.99-1 | 1 | (1.0,1.0,1.0,1.0) | (1.0,1.0,1.0,1.0) |
| KNN | 100 | 0.99-1 | 1 | (1.0,1.0,1.0,1.0) | (1.0,1.0,1.0,1.0) |
| NB | 96,86 | 0.961-0.974 | 0,95 | (0.59,1.0,1.0,0.98) | (1.0,0.99,0.96,0.99) |
| RF | 100 | 0.99-1 | 1 | (1.0,1.0,1.0,1.0) | (1.0,1.0,1.0,1.0) |
| PLS | 96,73 | 0.960-0.973 | 0,95 | (0.72,0.97,0.97,0.99) | (1.0, 0.99, 0.96,0.98) |
| MDA | 99,33 | 0.989-0.995 | 0,99 | (0.92,1.0,1.0,0.99) | (1.0,0.99,0.98,1.0) |
| C5.0 | 100 | 0.99-1 | 1 | (1.0,1.0,1.0,1.0) | (1.0,1.0,1.0,1.0) |
| Partial set (intrinsic variables) | |||||
| NN | 99,63 | 0.993-0.998 | 0,99 | (1.0,1.0,1.0,0.98) | (1.0,0.99,1.0,1.0) |
| SVM | 99,07 | 0.986-0.993 | 0,98 | (0.94,1.0,1.0,0.98) | (1.0,0.99,0.99,1.0) |
| KNN | 99,63 | 0.993-0.998 | 0,99 | (1.0,1.0,1.0,0.98) | (1.0,0.99,1.0,1.0) |
| RF | 99,64 | 0.993-0.998 | 0,99 | (1.0,1.0,1.0,0.98) | (1.0,0.99,1.0,1.0) |
| C5.0 | 99,63 | 0.993-0.998 | 0,99 | (1.0,1.0,1.0,0.98) | (1.0,0.99,1.0,1.0) |
*(croc,rf,sf,tramp). Key: croc, crocodile tooth marks; rf, retouched flakes; sf, simple flakes; tramp, trampling.