Research on Classification of College Students’ Physical Fitness Test Scores Based on Neural Network

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

The correlation between attributes

	Height	Weight	Vital capacity	50m run	Standing long jump	Sitting forward bending	Endurance Project	Power project
Height	1.0000	0.6106	0.6631	-0.5396	0.5976	-0.2374	0.1305	-0.6750
Weight	0.6106	1.0000	0.5442	-0.2694	0.2901	-0.1876	0.3332	-0.5438
Vital capacity	0.6631	0.5442	1.0000	-0.5372	0.5737	-0.1440	0.1314	-0.6386
50m run	-0.5396	-0.2694	-0.5372	1.0000	-0.7478	0.1766	0.0790	0.6322
Standing long jump	0.5976	0.2901	0.5737	-0.7478	1.0000	-0.1497	-0.0492	-0.6619
Sitting forward bending	-0.2374	-0.1876	-0.1440	0.1766	-0.1497	1.0000	-0.1230	0.3287
Endurance Project	0.1305	0.3332	0.1314	0.0790	-0.0492	-0.1230	1.0000	-0.2490
Power project	-0.6750	-0.5438	-0.6386	0.6322	-0.6619	0.3287	-0.2490	1.0000

The proportion of students in each category from 2016 to 2019

Year	Excellent category	Good class	Passing category	Fail class
2019	0.85%	15.82%	74.89%	8.44%
2018	1.03%	16.73%	72.26%	9.98%
2017	1.34%	20.29%	67.76	10.61%
2016	0.99%	19.71%	69.17%	10.13%

The parameters of the model

Parameter name	Parameter value
Number of Input Layer Neurons	8
Number of Output Layer Neurons	1
Number of Neurons in the Hidden Layer	11
Threshold	0.005
Learning rate	0.1
Maximum Iteration Count	1.0e10
Training algorithm	rprop+
Fault function	sse
Activation function	logistic

Percent of absolute error distribution predicted by the model established 2016 data

	[0, 1)	[1, 2),	[2, 3)	[3, 4)	[4, 5)	[5, ∞)
2016	64.38%	28.57%	6.27%	0.7%	0.04%	0.02%
2019	40.84%	27.16%	15.71%	8.06%	4.06%	4.17%

The correlation between attributes and synthesis score (2019)

	Height	Weight	Vital capacity	50m Run	Standing long jump	Sitting forward bending	Endurance Project	Power project
Consolidated results	—	—	-0.0274	-0.2078	0.1579	0.3992	-0.6848	0.4328
Sort	—	—	6	4	5	3	1	2