Research Article
Improving Accuracy and Simplifying Training in Fingerprinting-Based Indoor Location Algorithms at Room Level
Table 5
Relative accuracy by algorithm based on the number of WAPs when training with border locations (part I).
| #WAPs | ADTree | BFTree | Decision Stump | FT | J-48 |
| 40 | 98.17 ± 0.90 | 95.17 ± 1.57 | 94.50 ± 3.95 | 99.83 ± 0.50 | 91.50 ± 4.31 |
| 5 | 91.33 ± 1.94 | 87.67 ± 2.13 | 85.50 ± 1.07 | 70.50 ± 5.00 | 85.00 ± 3.07 |
| 8 | 93.33 ± 0.75 | 94.17 ± 0.83 | 89.83 ± 0.50 | 88.17 ± 4.74 | 90.17 ± 2.63 |
| 10 | 99.00 ± 1.11 | 96.83 ± 0.50 | 96.67 ± 0.00 | 99.17 ± 0.83 | 95.83 ± 1.34 |
| 16 | 97.33 ± 1.33 | 95.50 ± 0.76 | 94.83 ± 0.50 | 99.17 ± 1.12 | 95.17 ± 1.57 |
| 20 | 97.50 ± 1.12 | 95.83 ± 1.34 | 94.83 ± 0.50 | 98.83 ± 1.07 | 96.17 ± 1.67 |
| 32 | 98.00 ± 1.63 | 94.17 ± 2.71 | 94.83 ± 3.61 | 99.00 ± 1.33 | 91.50 ± 3.02 |
|
|