Table 5.

Comfort conditions classification report. ‘Precision’ is the ratio of correct positive predictions to all positive predictions; ‘recall’ is the ratio of correct positive predictions to all positive instances; ‘f1-score’ is the weighted harmonic mean between precision and recall; ‘support’ indicates the number of occurrences of each class in the data set.

N.	Precision	Recall	f1-Score	Support
1	0.92	0.94	0.93	5098
2	0.75	0.79	0.77	317
3	0.92	0.93	0.93	2406
4	0.63	0.64	0.63	233
5	0.89	0.54	0.67	78
6	0.90	0.87	0.89	3571
7	0.91	0.93	0.92	1896
8	0.67	0.50	0.57	4
9	0.31	0.19	0.24	21
11	0.98	0.96	0.97	8524
12	0.88	0.73	0.80	173
13	0.96	0.90	0.93	219
14	0.00	0.00	0.00	1
15	0.00	0.00	0.00	1
16	0.81	0.84	0.82	2090
17	0.70	0.74	0.72	245
18	0.61	0.45	0.52	31
19	0.91	0.89	0.90	2861
20	0.58	0.53	0.55	62
21	0.94	0.94	0.94	17
22	0.89	0.91	0.90	570
24	0.87	0.92	0.90	2865
26	0.75	0.82	0.78	211
27	0.96	0.80	0.87	30
accuracy			0.91	31,524
average	0.74	0.70	0.71	31,524
weighted avg	0.91	0.91	0.91	31,524