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

Table 6

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

#WAPs	J-48graft	LADTree	LMT	NBTree	Random; Forest
40	92.17; ±; 3.73	88.00; ±; 4.00	99.00; ±; 1.11	97.67; ±; 1.53	99.00; ±; 0.82
5	85.33; ±; 3.48	73.33; ±; 15.26	77.83; ±; 4.02	91.00; ±; 2.71	90.00; ±; 1.83
8	90.17; ±; 2.63	91.50; ±; 0.50	93.67; ±; 1.63	92.00; ±; 2.08	94.50; ±; 0.76
10	95.83; ±; 1.71	91.83; ±; 2.17	97.83; ±; 1.67	97.67; ±; 1.33	98.50; ±; 1.17
16	95.17; ±; 1.74	92.33; ±; 2.13	98.33; ±; 1.83	98.83; ±; 1.30	99.00; ±; 0.82
20	96.17; ±; 1.67	93.50; ±; 2.52	98.83; ±; 1.07	99.67; ±; 0.67	98.33; ±; 1.29
32	92.17; ±; 2.59	87.33; ±; 4.03	99.33; ±; 1.11	98.00; ±; 1.63	99.00; ±; 0.82