Improving Accuracy and Simplifying Training in Fingerprinting-Based Indoor Location Algorithms at Room Level

Table 4

Relative accuracy by algorithm based on the number of WAPs when training with all locations (part III).

#WAPs	Random; Tree	REPTree	SVM	NN	NBC
40	90.83; ±; 7.86	93.33; ±; 6.24	99.17; ±; 2.50	94.17; ±; 6.51	85.00; ±; 7.26
5	80.83; ±; 10.57	82.50; ±; 7.86	71.67; ±; 10.00	83.33; ±; 9.86	78.33; ±; 9.28
8	89.17; ±; 8.37	86.67; ±; 11.90	85.00; ±; 11.06	87.50; ±; 11.33	79.17; ±; 11.93
10	94.17; ±; 6.51	94.17; ±; 7.50	95.83; ±; 5.59	95.83; ±; 5.59	96.67; ±; 4.08
16	92.50; ±; 8.70	93.33; ±; 8.16	97.50; ±; 5.34	95.83; ±; 5.59	97.50; ±; 3.82
20	90.00; ±; 10.41	92.50; ±; 9.46	98.33; ±; 5.00	96.67; ±; 5.53	95.00; ±; 6.67
32	87.50; ±; 9.32	93.33; ±; 8.98	98.33; ±; 3.33	95.83; ±; 5.59	86.67; ±; 11.30