Applying Deep Learning Technologies to Evaluate the Patent Quality with the Collaborative Training

Table 8

Data transfer learning results of the initial 208 Chinese patent English.

Category	Training set				Test set
	Precision	Recall	-score	Support	Precision	Recall	-score	Support
1	1.00	0.92	0.96	61	1.00	0.77	0.87	35
2	0.83	0.91	0.87	33	0.54	0.78	0.64	18
3	0.82	0.92	0.87	49	0.78	0.67	0.72	21
4	0.81	0.72	0.76	29	0.20	0.20	0.20	10
5	0.50	0.76	0.60	17	0.35	0.75	0.48	8
6	0.40	0.40	0.40	5	0.33	0.33	0.33	3
7	0.60	0.33	0.43	9	0.67	0.33	0.44	3
8	0.75	0.60	0.67	5	1.00	0.50	0.67	2
Microavg	0.83	0.83	0.83	208	0.66	0.66	0.66	100
Macroavg	0.71	0.70	0.70	208	0.61	0.54	0.57	100