Smartwatch-Based Legitimate User Identification for Cloud-Based Secure Services

Table 3

Confusion matrix according to the users using 400 samples per window and magnetometer sensor.

Users	User 1	User 2	User 3	User 4	User 5	User 6
DT classifier
User 1	1230	48	17	151	21	184
User 2	39	1499	63	49	89	43
User 3	29	136	1551	110	282	9
User 4	160	101	125	1476	84	79
User 5	51	63	188	17	1138	1
User 6	263	22	3	37	3	1696
SVM classifier
User 1	1005	73	19	168	52	302
User 2	137	1244	292	69	72	110
User 3	56	284	1124	167	511	15
User 4	147	220	184	1166	107	94
User 5	48	68	238	24	914	29
User 6	343	28	23	178	26	1521
KNN classifier
User 1	1300	41	21	233	45	318
User 2	78	1423	168	93	132	47
User 3	34	182	1311	134	353	48
User 4	198	103	167	1248	87	178
User 5	45	79	226	45	998	4
User 6	168	24	29	64	2	1425
NB classifier
User 1	763	99	49	201	72	354
User 2	149	1047	224	103	75	200
User 3	135	339	1003	190	511	65
User 4	343	258	269	1050	298	368
User 5	109	44	186	54	594	0
User 6	879	108	139	175	121	1078