Construction of Multi-Channel Teaching Effect Evaluation System Based on Deep Learning in the Era of Education Informatization

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Channel teaching evaluation index system

Primary indicator	Secondary indicator
Background evaluation	Target setting
	Faculty
	Student capacity
Input evaluation	Facility resources
	Teacher reserve
	Course preparation
Process evaluation	Teacher performance
	Student performance
	Programme implementation
Result evaluation	Teacher development
	Student growth
	Overall effect

The number of questionnaires and the results of experts’ evaluation

Course name	Questionnaire distribution	Expert evaluation results
3D animation design and production	100	Excellence
China modern history	100	Good
College students mental health education	100	Medium
Gem appreciation	100	Qualify
National music appreciation	100	Out of line
College students’ artistic appreciation	100	Good
Tot	600

According to the questionnaire obtained

Courses	Serial number	Evaluation index					Evaluation grade	Quantitative result
		X1	X2	X3	……	X12
3D animation design and production	1	3	4	3	……	3	Excellence	0.95
	2	4	4	3	……	3	Excellence	0.95
	3	4	3	3	……	4	Excellence	0.95
	……	……	……	……	……	……	……	……
	100	3	3	4	……	3	Excellence	0.95
China modern history	1	4	3	3	……	3	Good	0.85
	2	4	4	3	……	3	Good	0.85
	3	3	4	4	……	3	Good	0.85
	……	……	……	……	……	……	……	……
	100	3	4	3	……	4	Good	0.85
College students mental health education	1	2	2	3	……	3	Medium	0.75
	2	3	3	2	……	2	Medium	0.75
	3	3	3	3	……	2	Medium	0.75
	……	……	……	……	……	……	……	……
	100	2	3	2	……	2	Medium	0.75
Gem appreciation	1	1	1	2	……	1	Qualify	0.65
	2	1	1	2	……	2	Qualify	0.65
	3	1	1	1	……	2	Qualify	0.65
	……	……	……	……	……	……	……	……
	100	2	2	2	……	1	Qualify	0.65
National music appreciation	1	1	0	1	……	1	Out of line	0.3
	2	1	1	0	……	1	Out of line	0.3
	3	1	0	0	……	0	Out of line	0.3
	……	……	……	……	……	……	……	……
	100	0	0	1	……	0	Out of line	0.3
College students’ artistic appreciation	1	4	3	2	……	3	Good	0.85
	2	4	2	2	……	4	Good	0.85
	3	4	3	3	……	3	Good	0.85
	……	……	……	……	……	……	……	……
	100	4	3	2	……	2	Good	0.85

Comparison of performance indicators of different algorithms

Algorithm	Training error (RMSE)	Test error (RMSE)	The number of hidden layers of neurons	Training time/s	Test time/s
GA-RBF	10.7415	10.9745	12	955.4	0.0033
APSO-RBF	8.1544	8.1145	11	914.2	0.0032
Improved PSO-RBF	7.0025	7.0128	9	905.1	0.0028

Sample training results

Courses	Serial number	Actual output	Expected output	Training results	Expert outcome
College students’ artistic appreciation	1	0.8541	0.85	Good	Good
	2	0.8564	0.85	Good	Good
	3	0.8451	0.85	Good	Good
	……	……	……	……	……
	100	0.8459	0.85	Good	Good