Uneingeschränkter Zugang

Performance evaluation of college laboratories based on fusion of decision tree and BP neural network


Zitieren

Fig. 1

Procedure of performance evaluation for laboratories with the use of the decision tree and BP neural network.
Procedure of performance evaluation for laboratories with the use of the decision tree and BP neural network.

Fig. 2

The structure of the BP neural network.
The structure of the BP neural network.

Fig. 3

Neural network error curve comparison.
Neural network error curve comparison.

Evaluation system of college laboratories.

The primary indexes The secondly indexes
Construction

1. Area and the environment

2. Instruments and equipment

3. Operation and maintenance

4. System and management

Laboratory team building

5. Tutors of experiment

6. Laboratory team construction

7. Personnel structure

8. Appraisal mechanism

9. Training mechanism

Experimental teaching

10. Practice ability

11. Exam of experiment

12. Report of experiment

13. Comprehensive and designed experiments

Administration system

14. System and management

15. Management tool

16. Experiment teaching material

17. Service efficiency

Laboratory safety

18. Safety measures

19. Hazmat management

20. Experimental environment protection

21. Clean and tidy

Innovation and entrepreneurship

22. Personnel structure proportions

23. Innovative entrepreneurship

24. Experiment project for college student

Grade system of evaluation indexes.

Laboratory number Index 1 Index 2 Index 3 Index 4 Index 5

1 C B B C B
2 B B A A B
3 A A B A B
4 A C B B B
5 B D A B B
6 C B A B B
7 A A A A A
8 B D C B C
9 B D C C C
10 A B A A A

The scores of the indexes.

Laboratory number Index 1 Index 2 Index 3 Index 4 Index 5

1 72 84 76 72 75.6
2 77 78 85 85 81.2
3 86 94 75 86 84.3
4 87 71 82 82 81.4
5 82 63 88 76.5 80.2
6 71 83 89 80 80.6
7 85 94 92 88.5 89.6
8 81 60 68 74.5 71.6
9 79 55 74 66.5 70.2
10 91 76 90 89 87.3

Laboratory evaluation index system.

Number Indexes Information gain ratio
1 Area and environment 33.51%
2 Instruments and equipment 28.46%
3 Operation and maintenance 27.63%
4 System and management 25.321%
5 Practice ability 21.25%
6 Service efficiency 20.87%
7 Personnel structure 19.32%
8 Comprehensive and designed experiments 19.87%
9 Experiment project for college student 18.56%
10 Hazmat management 17.62%
11 Innovative entrepreneurship 16.93%

Comparison of real values and predicted values.

Laboratory No. 1 2 3 5 6 7
Real values 82 76 96 65 85 92
Predicted values 82.3 76.2 95.8 62.3 85.1 91.8

The data after fuzzy processed.

Index 1 Index 2 Index 3 Index 4
A B C D A B C D A B C D A B C D
0 0 1 0 0.375 0.625 0 0 0 0.625 0.375 0 0 0 1 0
0 0.875 0.125 0 0 0.875 0.125 0 0.625 0.375 0 0 0.625 0.375 0 0
0.75 0.25 0 0 1 0 0 0 0 0.625 0.375 0 0.75 0.25 0 0
0.625 0.375 0 0 0 0 1 0 0.125 0.875 0 0 0.125 0.875 0 0
0 0 0.875 0.125 0 0 0.75 0.25 0.75 0.25 0 0 0 0.625 0.375 0
0 0 1 0 0.75 0.25 0 0 1 0 0 0 0.375 0.625 0 0

Comparison of experimental results.

Sequence Number of training samples Accuracy
Decision tree BP neural network Decision tree and BP neural network
1 300 76.4 73.8 81.6
2 400 78.2 72.5 82.5
3 500 81.6 77.6 84.3
4 600 82.1 79.3 86.1
5 700 81.6 80.4 88.5
6 800 82.1 81.5 86.2
7 900 83.4 82.9 89.4
8 1000 84.5 83.2 91.6
Average value 81.24 78.9 86.28
eISSN:
2444-8656
Sprache:
Englisch
Zeitrahmen der Veröffentlichung:
Volume Open
Fachgebiete der Zeitschrift:
Biologie, andere, Mathematik, Angewandte Mathematik, Allgemeines, Physik