Available via license: CC BY 4.0
Content may be subject to copyright.
Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
Applied Mathematics and Nonlinear Sciences
https://www.sciendo.com
† Corresponding Author.
Email address: wangrujun202411@163.com
ISSN 2444-8656
DOI: https://doi.org/10.2478/amns-2025-0129
© 2025 Wang, R. published by Sciendo.
This work is licensed under the Creative Commons Attribution alone 4.0 License.
Application and Analysis of Big Data Analysis in Sports
Rujun Wang1,†
1. Zhumadian Preschool Education College，Zhumadian, Henan463000, china.
Submission Info
Communicated by Z. Sabir
Received October 9, 2024
Accepted January 11, 2025
Available online February 27, 2025
Abstract
In this digital age, with the burst of national sports events and the promotion of relevant policies, the sports industry has
gradually risen, and sports have gradually spread to everyone. With the successful holding of the 2008 Olympic Games
and the Winter Olympics Winter Paralympic Games in 2022, the national movement has gradually formed in everyone's
heart. However, in sports, the physical and mental health of every athlete needs our close attention. And let our country's
sports cause to a higher level, but also as the current core concern. With the rapid development of big data technology, it
has been integrated into all areas of life. Big data technology is used to analyze all aspects of sports, such as sports events,
athletes'competition analysis, athletes' training mode and so on. So as to further enhance the sports performance in the
competition. This paper puts forward the application and analysis of big data analysis in sports. Firstly, under the current
sports prospect, the quality core brought by sports is analyzed by using big data analysis model. Finally, through algorithm
comparison and analysis, the problems and statistical results are processed together to enhance the popularity of sports
and solve its related problems. People of different ages have different understanding of sports. The probability of five
different ways of sports competition at different ages is less than 0.05. Young people have the highest understanding of
sports competition consciousness and the lowest understanding of sports physical and mental health.
Keywords: Big Data; Artificial Neural Network; Sports; Physical and Mental Health
AMS 2020 codes: 62R07
Rujun Wang. Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
2
1 Introduction
With the rapid development of sports in China, especially the successful hosting of the 2008 Olympic
Games, large-scale improvement of athletes'performance in various and special sports has become
the focus of attention of management departments, coaches, athletes and even the whole nation.
Athletes'excellent achievements depend on scientific training methods. In order to successfully
manage and control the training process of athletes, it is necessary to establish a scientific and
reasonable training model for athletes. Athlete training model is a mathematical model which takes
the specific sports performance as the dependent variable and the training index as the independent
variable. This shows that the training index has an impact on the specific performance, and can predict
the specific performance under a certain training level. By using professional training model, coaches
can avoid making training plans, and then arrange the training process scientifically, so as to promote
athletes to create the best results in a relatively short period of time. Modern sports training is a long-
term and complex process, which requires the continuous inflow of information to restrict its
effectiveness. This kind of information is used to train and control [1]. The knowledge of artificial
intelligence is used to diagnose and evaluate the techniques and tactics of Chinese men's table tennis
players in competition, so as to analyze and improve the loop drive techniques of our hands. In terms
of tactics, we should adopt more tactics of serving attack and receiving control [2]. BP neural network
is used to evaluate the quality of volleyball teaching, aiming at the problem that neural network is
easy to fall into local optimum, BP algorithm is improved on the basis of artificial fish swarm
algorithm [3]. BP algorithm is improved on the basis of artificial fish swarm algorithm [4]. Neural
network is used to improve the success rate of volleyball players' blocking [5]. The maximum entropy
neural network model has been successfully applied in the establishment of basketball teaching
quality evaluation model. Researchers have developed a table tennis technical and tactical analysis
system based on artificial neural network model to filter the original data and obtain the technical and
tactical utilization rate and scoring rate to guide table tennis matches [7]. Using BP neural network
and ACSI model to explore the audience's appreciation of the quality of the game, it is concluded that
the table tennis diameter of 39.4mm is the optimal solution for both players and spectators to
experience the quality of the game [8]. The method of using forward-backward neural network to
realize the robot arm table tennis competition was studied [9]. The volleyball robot can be used to
plan motion, such as learning the behavior pattern of the human reader who performs the
corresponding action by collecting experience, and solving the motion planning problem when the
initial situation changes, such as speed and angle [10]. Literature [11] The purpose of this study is to
explore the effect of cognitive training of working memory tasks on the efficiency of executive
control network of table tennis players. Artificial neural network (ANN) is used to predict the results
of the competition, so that coaches can arrange appropriate and effective training [12]. Researchers
designed two fuzzy neural network classifiers to estimate the rotation pattern and predict the
trajectory of table tennis [13]. In the context of studying the athletic performance of athletes,
considering the corpus of special actions of table tennis shots, the proposed model with attention
block is superior to the model without attention block and our baseline [14]. Using BP neural network
and multiple loop method, the technology and support ability of elite male young athletes were
analyzed [15]. In order to solve the behavior selection problem of the kicker robot in the robot soccer
game, the neural free motion network is combined with reinforcement learning, which has been
successful [16]. Various appropriate PID control structures are established based on BP network,
including traditional PID control, different quantitative processing, NNM system identification
network and NNC system control network. Discussion on the structure of single PID neuron
Application and Analysis of Big Data Analysis in Sports
3
controller for motion control based on BP network and different PID controllers [17]. The technique
and technical ability of elite male table tennis players were analyzed by using neural network and
multiple loop method, and the results were clear [18]. The body shape of different athletes in
trampoline was analyzed and the corresponding scheme was formulated [19]. High-resolution
imaging and recording of the network of colloidal nerve conduits were used to pre-process and
identify the typical basic activities of backbone, release and running in the video stream. Target
detection and fine positioning to effectively deal with the technical characteristics of basketball [20].
2 Big data analysis technology
2.1 Basic Theory of Artificial Bee Colony Algorithm
Artificial bee colony algorithm is a new global optimization algorithm based on clustering
intelligence proposed by karabcga in 2005. 2008. The intuitive background is derived from the
behavior of bees collecting honey. Its main feature is that it does not need to know the specific
information about the problem, but only needs to compare the advantages and disadvantages of the
problem. Through the local optimization behavior of each artificial bee individual, the population
finally appears the global optimal value at a faster approximation speed.
2.1.1 Artificial Bee Colony
The initial solution x is randomly generated, and the definition formula is:
，，， (1)
Of which:
(2)
Leading peak stage：
(3)
Reconnaissance bee stage：
(4)
Rujun Wang. Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
4
2.1.2 Demonstration Process of Algorithm
Initialization phase
Leading bee stage
Following bee stage
First cycle?
Taking the current
optimal solution as the
initial clustering center
of fuzzy C-means
algorithm for
optimization
Begin
Reconnaissance bee
stage
Whether the maximum
number of iterations has
been reached
End
Has the optimal solution
changed?
Yes
No
Yes
No
Yes
No
Figure 1. Artificial bee colony algorithm flow
The system generates random solution X, and judges whether the current optimal solution is regarded
as fuzzy C to enter the cycle through leading the bee colony stage and following the bee colony stage.
After entering the cycle, it reaches the reconnaissance bee stage, and after entering the reconnaissance
bee stage，determine if it has reached the maximum number of iterations.
2.2 Basic Theory of ant colony algorithm
Ant colony algorithm (ACA) is an intelligent optimization algorithm, which can simulate the nearest
route to find food. In general, the real random search for food, which is called pheromone in chemistry,
can enable the ant colony to find the shortest route from the ant nest to the food source in a short time.
2.2.1 Bee Colony Algorithm
Path heuristic information:
Application and Analysis of Big Data Analysis in Sports
5
(5)
Update local pheromone:
(6)
Global pheromone update:
(7)
Build a new solution until the goal is achieved:
(8)
2.2.2 Demonstration Process of Algorithm
Ant colony algorithm process is shown in Figure 2.
Begin
Initialization
parameters
Initialize M female ants
randomly at the initial
position
Each ant uses roulette scheduling
method to select the next service
according to the state transition
probability formula
Each ant updates its
local pheromone
Whether all ants have
selected all services
Calculate fitness
value
The best ant
updates the global
pheromone
Whether the
termination
conditions are
met
Output optimal
solution
End
No
Yes
No
Yes
Figure 2. Ant colony algorithm process
Rujun Wang. Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
6
After initializing the parameters, enter the random position to initialize m ants, select the roulette
scheduling service according to the status of each ant, judge whether each ant has selected all services,
then start to calculate its fitness value, select the best ant to update the pheromone, judge whether the
termination conditions are met, and output the optimal solution to end the cycle after termination.
2.3 Basic Theory of Cultural Algorithm
The cultural algorithm framework describes cultural evolution as a double inheritance process: at the
micro level (i.e. the population space), the individual evolution of the population forms behavioral
characteristics, which are passed on from generation to generation under the action of a group of
social incentive operators; At the macro level (i.e. belief space), individual experience is evaluated
through collection, consolidation, induction and specialization, so as to retain the above behavior
characteristics, store and share them in the belief space, so as to guide the continuous evolution at the
micro level through communication with the micro level.
2.3.1 Cultural Algorithm
Add covariance matrix:
(9)
Calculate characteristic equation dot product:
(10)
(11)
Simplify Gaussian kernel function:
(12)
2.3.2 Demonstration Process of Algorithm
Cultural algorithm process is shown in Figure 3.
Application and Analysis of Big Data Analysis in Sports
7
Initialize
population
and belief
space
Through the initialized parameter
value, the SVM parameter
combination is randomly generated
Use each parameter combination in the
current parameter combination group to
train the data and obtain the detection
results
Compare the test results with the actual
results and calculate the fitness value
Whether the
termination
conditions are met
According to the acceptance
function, the dominant
individuals are extracted
from the parents to enter the
belief space
Extract the dominant
information from the
dominant individuals and
update the knowledge
Store the updated
knowledge into the
evolutionary knowledge
base
Mutate the population
Select individuals with high
fitness value to form the
next generation group
End
Yes
No
Guided
mutation
Figure 3. Cultural algorithm process
Initialize the population and belief space, detect each parameter of the randomly generated SVM
combination and get the results. Compare it with the actual results to judge whether the conditions
are met. If not, perform mutation on the population, select the individuals with high fitness to enter
the next generation of population, and enter the judgment cycle again until the termination conditions
are met.
2.4 Basic Theory of Genetic Algorithm
Genetic algorithm (GA) is an algorithm that simulates the principle of biological evolution in nature.
In other words, in the process of evolution, we should retain useful individuals, eliminate useless
individuals, and follow the law of survival of the fittest and survival of the fittest. In scientific and
social practice, we should find the most practical solution among all solutions. In a word, it is to find
the best solution.
2.4.1 Genetic Algorithm
Initialize task set:
(13)
Rujun Wang. Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
8
Initialize task length set:
(14)
Collection of virtual machines:
(15)
Initialize virtual machine processing speed:
(16)
Etc matrix:
(17)
Unit cost of virtual machine Runtime:
(18)
2.4.2 Demonstration Process of Algorithm
Genetic algorithm flow is shown in Figure 4.
Application and Analysis of Big Data Analysis in Sports
9
Begin
Input rule set
Count the number of
occurrences of rule set
mode
Arrange patterns in descending
order of sharing degree to
generate sequence S(p)
Use S(p) to initialize the
population of GA, and the
number of individuals is n
Individuals in the population
are not evaluated？
Whether any individual has
reached the target value
Select an individual in the
population as a sequence of
RETE network
Start timing
Build RETE
network
Input instance TPI
Perform rule
matching reasoning
Output matching
results Random elimination
of individuals using
Roulette
Select individual
Hugh to replicate,
so that the
population
reaches n
Randomly selected
individuals for
crossover
Randomly selected
genes for mutation
Save the individuals
that reach the target
value as α Sequence
of networks
End
End timing
No
Yes
Yes
Figure 4. Genetic algorithm flow
By judging the number of occurrences of the rule set, use S(p) to initialize the GA population to judge
whether the individuals in the population have been evaluated. If not, judge whether the individuals
have reached the target value. If so, save the target value. If not, enter the cycle again; If the evaluation
is reached, the rule matching reasoning will be carried out again and the cycle judgment will be
entered again.
2.5 Basic Theory of Tabu Search Algorithm
Tabu search algorithm was born in the late 1970s. It has the ability of fast calculation and can
eliminate In addition to local optimality, it can solve complex large-scale engineering problems. At
present, tabu search algorithm has been applied in many automation fields, such as traveling salesman
problem, locomotive scheduling problem, secondary assignment problem and workflow scheduling
problem, and has been highly praised by a wide range of experts and scholars.
Rujun Wang. Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
10
2.5.1 Tabu Search Algorithm
Domain move initialization:
(19)
A collection that can be reached by domain mobility:
(20)
Initial solution X:
(21)
Find the optimal value:
(22)
2.5.2 Demonstration Process of Algorithm
Tabu search algorithm flow is shown in Figure 5.
Begin
Setting
parameters
Generate
domain solution
PT 500
PD>200
Generate
domain solution
Calculate L (X') to meet
the taboo condition
Move n belongs to taboo
table
Tabu table
update，
PD=PD+1
CalculateF(X'),to meet the
taboo condition
Moving m belongs to
the taboo table
Tabu table
update，PT=PT +1
Output optimal
solution and
objective
function
End
Break the
prohibition and
update the taboo
list，PT=PT+1
Break the
prohibition,
update the taboo
list, PD=PD+1
Set initial solution X,
objective function value
F(x) iteration times PT=0,
PD=0
Yes
NO
Yes
No
No
Yes
No
No
Yes
No
Yes
Figure 5. Tabu search algorithm flow
Application and Analysis of Big Data Analysis in Sports
11
Set PD and PT parameters, generate domain solutions, and judge whether the tabu table is updated.
If it is updated, judge whether M meets the tabu conditions. If it is, output the objective function of
the optimal solution.
3 Analysis on The Current Situation and Problems of Sports
3.1 Problems in Sports
3.1.1 Physical exercise is still in the position of resistance in most people's minds
With the popularity of sports, more and more people have joined in sports. However, there are still
many people who refuse sports. One of the reasons is that their own physical condition is not suitable
for sports. The second is the psychological construction of some people on sports. When we mention
sports, we will think of the word "tired", and then produce resistance psychology. This kind of
problem can be encouraged. We should also respect their wishes, and solving this problem is the key
to driving the all member movement.
3.1.2 Negative effects of sports on mental health
After research, only according to their own physical, mental health and other aspects of the situation
to participate in appropriate sports can promote mental health. If the way of exercise is unreasonable,
it will not only damage the body, but also have a bad impact on their mental health, which is mainly
manifested in physical and mental loss and exercise dependence. Physical and mental attrition refers
to a kind of psychological and physiological reaction that can not be avoided during exercise for a
long time. It is a kind of psychological and physiological law that can not be avoided during long-
term exercise. This symptom will not only damage their mental health, but also indirectly lead them
to withdraw from exercise. Dependence refers to the athletes' psychological and physiological
dependence on the normal sports lifestyle, including positive and negative. Positive people can
control exercise behavior, while negative people tend to have more negative emotions after exercise,
such as depression, anxiety, anger and so on. This shows that the physical and mental health of
national sports also needs attention.
3.1.3 It is difficult to develop special sports athletes in China
As we all know, our table tennis level is in the forefront of the world, and our table tennis players
have also made many excellent achievements. However, there are still many reasons restricting its
development. First of all, retired athletes cannot get good arrangements. In the early stage, with the
strong support of the government, this problem will still be properly arranged. However, with the
development of the market economy, retired players need to find their own way out, and it is still
difficult to gain a foothold in the society. The second is logistics. As the saying goes, if God opens a
door for you, it will close a window for you. Professional athletes can get more training but have no
time to learn cultural knowledge. Amateur athletes have the opposite. Third, due to the limited
number of provincial teams and national teams, the competition is very fierce. Many excellent athletes
will be in a state of no ball to play, and the national team will be forced to retire or go abroad for
development.
Rujun Wang. Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
12
3.2 Strategies for Promoting Sports Popularity
The popularization of sports is imminent. With the improvement of people's living standards, more
and more old people and young people like sports. For the middle-aged people who are "not up or
down", even some office workers who are sedentary, they need sports more. For how to popularize,
we can probably draw the public's attention from the benefits of sports and the popularity of sports
events. For example, the 2022 Winter Olympic Games and the winter Paralympic Games also
launched a National Snow Sports, which is a good measure. The advantage of exercise is to enhance
one's own immunity and resistance. Of course, it is not only to strengthen the body, but also to
enhance the relationship between relatives and friends. Driven by the all people Olympic Games, the
all people movement is becoming more and more popular. I believe that in the near future, the all
people movement will also be realized in everyone.
3.3 Analysis of Athlete's Performance and Training Optimization
Evaluate the performance of athletes: By collecting various data of athletes in training and
competition, such as speed, strength, endurance, heart rate, oxygen saturation and other physiological
indicators, as well as the completion quality of technical movements, tactical implementation, etc.,
establish an evaluation model to comprehensively and objectively evaluate the performance level of
athletes, and help coaches and athletes understand their strengths and weaknesses.
Personalized training plan formulation: according to the athlete's physical condition, technical
characteristics and training objectives, combined with historical training data and competition data,
use big data analysis to formulate personalized training plan. For example, the best training intensity
and interval time are determined according to the changing law of athletes'heart rate, and the targeted
technical improvement program is provided according to the data analysis of technical movements,
so as to improve the training effect and competitive level.
Training load monitoring and adjustment: real-time monitoring of athletes in the training process of
physiological data and exercise data, analysis of the size of the training load and trends. Through the
quantitative assessment of training load, the risk of fatigue and injury caused by overtraining can be
avoided, while ensuring that the training intensity is sufficient to promote the physical and technical
improvement of athletes.
4 Experimental Results and Analysis
4.1 Comparative Experiment
Using the research of artificial bee colony algorithm, we counted the current people's familiarity with
sports types, compared with the situation without artificial bee colony algorithm, and observed the
advantages of artificial bee colony algorithm to sports. We have selected pupils, middle school
students, high school students, college students and the middle-aged and elderly. From the perspective
of physical education, we believe that sports should have five core qualities for people, and make a
systematic analysis. The specific survey data are shown in table and table:
Application and Analysis of Big Data Analysis in Sports
13
Table 1. Five benefits of sports statistics by traditional methods
Dimension
Minimum
value
maximum
value
average
standard
deviation
Question
mean
Competitive
consciousness
6
80
42.11
11.21
3.52
Sense of cooperation
7
50
37.42
13.42
3.51
Rule awareness
9
30
10.98
3.21
3.11
innovation ability
5
30
18.11
6.26
3.09
Physical and mental
health
10
60
12.43
5.11
3.02
Total score
37
250
121.05
39.21
16.25
Figure 6: Statistical chart of distribution of different dimensions
According to the data in Table 1 and Figure 6, we can conclude that in sports, the average value of
competitive awareness is the largest, which is 42.11, and the average value of rule awareness is the
smallest, which is 10.98. The average value of sports literacy in five different dimensions is 3.25,
which is basically lower than the average dimension value of 3, indicating that the popularity of core
literacy in sports is not very good.
Table 2. Artificial neural network statistics five benefits of sports
Dimension
Minimum
value
Maximum
value
Average
Standard
deviation
Question
mean
Competitive consciousness
12
70
52.65
14.11
3.76
Sense of cooperation
14
55
41.81
10.53
3.8
Rule awareness
9
30
22.54
6.02
3.75
innovation ability
9
30
22.42
5.78
3.73
Physical and mental health
11
60
15.1
3.85
3.79
0
10
20
30
40
50
60
70
80
90
Competitive
consciousness
Sense of
cooperation
Rule awareness innovation ability Physical and
mental health
Minimum value maximum value
Question mean average
Rujun Wang. Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
14
Figure 7. Statistical chart of distribution of different dimensions
According to the data in Table 2 and Figure 7, we can see that under the statistical mode of artificial
neural network algorithm, sports presents a good situation for different dimensions, and the average
values of the five different dimensions have been improved. The average value of competitive
consciousness is the largest, reaching 52.65. Compared with the traditional statistical mode, the
average value has increased by 10.57, and the average value of physical health is the smallest,
reaching 15.10, an increase of 2.67. The average value of the five different models of sports is 3.77,
which is 0.52 higher than the traditional statistical model, both higher than the intermediate critical
value of 3, indicating that the popularity of sports is very good
4.2 Factor Analysis
4.2.1 Evaluation Criterion
Evaluation standard is shown in table 3.
Table 3. Evaluation standard table
The closer the value of I is to 1, the better. A value between 3 and 6 indicates that the model
is acceptable.
A region value between 0.07 and 0.09 indicates that the model is acceptable.
ess than 0.1, the smaller the value, the better。
,,,
, ,, the value of is between 0 and 1, and the closer it is to 1, the better.
4.2.2 Experimental Results and Analysis
According to the results of the comparative experiment, we can find that there are certain differences
between the core qualities of sports in various dimensions. In order to detect the different influencing
factors of sports on physical fitness, the experiment started from three different aspects: age, gender
0
20
40
60
80
100
120
Competitive
consciousness
Sense of
cooperation
Rule
awareness
innovation
ability
Physical and
mental health
average Question mean
Application and Analysis of Big Data Analysis in Sports
15
and occupation, with age, gender and occupation as independent variables and sports as dependent
variables, and conducted factor analysis. The experimental data are as follows:
a.Age
The difference of sports quality in different ages is shown in table 4.
Table 4.The difference of sports quality in different ages
Age
<18
<45
<60
<90
Factor
M
SD
M
SD
M
SD
M
SD
F
Sig
Competitive consciousness
48.21
17.29
55.81
9.9
55.34
11.62
52.32
14.24
2.70*
0.045
Sense of cooperation
39.21
12.97
44.04
7.99
44.08
8.95
41.53
10.48
2.87*
0.036
Rule awareness
21.08
7.69
23.81
4.62
23.51
5.31
22.44
5.87
2.38
0.069
Innovation ability
21.04
7.08
23.6
4.46
23.79
4.96
22.24
5.73
2.95*
0.032
Physical and mental health
14.06
4.78
16
2.53
15.65
3.45
15.08
3.82
2.68*
0.046
Figure 8. Statistical chart of sports quality difference under different ages
From the data in figure 8, we can see that people of different ages have different understanding of
sports. The probability of five different ways of sports competition at different ages is less than 0.05.
Young people have the highest understanding of sports competition consciousness and the lowest
understanding of sports physical and mental health. Age is in direct proportion to people's
understanding of the benefits of sports. When they are older than 18, these young people have the
best understanding of all aspects of sports. Because the education and knowledge received by each
age group are different, in terms of physical and mental health, people over the age of 45 are higher
than those under the age of 45. Generally speaking, with the increase of age, the more experience you
get, you can have a more comprehensive understanding of the benefits of sports.
b. Gender
The difference of sports quality between different genders is shown in table 5.
0
10
20
30
40
50
60
MSD MSD MSD MSD
<18 <45 <60 <90
Competitive consciousness Sense of cooperation
Rule awareness Innovation ability
Rujun Wang. Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
16
Table 5. The difference of sports quality between different genders
Gender
Male
Female
Male
Female
Factor
M
SD
M
SD
M
SD
M
SD
F
Sig
Competitive consciousness
44.5
16.53
48.9
17.57
53.17
14.51
55.48
7.5
4.96*
0.002
Sense of cooperation
35.83
13.92
38.6
12.78
42.09
10.8
44.43
5.26
6.05*
0
Rule awareness
19
7.42
20.5
7.86
22.81
5.99
23.91
3.15
6.46*
0
Innovation ability
19
7.44
20.85
7.13
22.47
5.89
24.04
2.9
6.29*
0
Physical and mental health
12.88
4.66
13.66
4.93
15.3
3.89
16.01
1.83
7.14*
0
Figure 9. Statistical chart of sports quality differences under different genders
From the data in Figure 9, we can see that there are certain differences between men and women in
their cognition and understanding of all aspects of sports. The accompanying probability of five
different qualities in different genders is less than 0.01. Women have a high level of understanding
of physical and mental health brought about by sports, and a low level of understanding of rule
awareness. The tea friends brought by gender are out of proportion to people's understanding of sports.
Gender cannot determine whether people have a better understanding of sports. When in the role of
women, women pay more attention to the benefits of sports. Generally speaking, men and women are
equal, and women are more active in sports.
c. Occupation
The difference of sports quality in different occupations is shown in table 6.
0
10
20
30
40
50
60
MSD MSD MSD MSD
Male Female Male Female
Competitive consciousness Sense of cooperation Rule awareness
Innovation ability Physical and mental health
Application and Analysis of Big Data Analysis in Sports
17
Table 6 The difference of sports quality in different occupations
Occupation
White
collar
Retire
Student
Athletes
Factor
M
SD
M
SD
M
SD
M
SD
F
Sig
Competitive
consciousness
48.32
17.36
53.56
11.17
56.81
10.18
55.48
7.5
14.61*
0
Sense of cooperation
38.88
12.69
42.41
8.46
44.81
8.05
44.43
5.26
12.34*
0
Rule awareness
20.76
7.44
23.05
4.62
24.22
4.49
23.91
3.15
13.26*
0
Innovation ability
20.83
7.04
22.8
4.85
24.02
4.17
24.04
2.9
11.98*
0
Physical and mental
health
13.91
4.6
15.43
3.27
16.25
2.79
16.01
1.83
14.80*
0
Figure 10 Statistical chart of sports quality difference under different occupations
From the data in Figure 10, we can see that people of different occupations have different
understanding of the five qualities of sports. The accompanying probability of the benefits of the five
different dimensions under different occupations is less than 0.02. Professional athletes are at the
highest level in terms of competitive awareness and low in terms of physical and mental health. The
reason is also the health problems of athletes caused by high-intensity sports. The popularity of
students is relatively better than that of the lower level, and the popularity of white-collar workers is
relatively low. Generally speaking, the office workers who enter the society generally do not accept
sports, and they need to strengthen their own exercise; At the same time, athletes' physical and mental
health problems brought by high-intensity training also need attention.
0
10
20
30
40
50
60
MSD MSD MSD MSD
White collar Retire Student Athletes
Innovation ability Physical and mental health Competitive consciousness
Sense of cooperation Rule awareness
Rujun Wang. Applied Mathematics and Nonlinear Sciences, 10(1) (2025) 1-19
18
5 Conclusions
To sum up, sports are also a vital part of maintaining the health of the public. Both people and animals
are exercising. As the saying goes, "life goes on and sports goes on", it is very important to develop
and popularize sports in the rapid development of today's era. Both the traditional popularization and
the Popularization Based on artificial neural network tell us that "time waits for no man". Article from
age. The three aspects of gender and occupation are analyzed by using artificial neural algorithm.
According to the problems revealed, corresponding change measures are taken, and some help is
given to relevant personnel. After this analysis, we will try our best to change the current situation.
We believe that in the near future, sports will become the hottest sport for the whole people.
References
[1] Roczniok, R., Rygua, I., & Kwaniewska, A. (2007). The use of Kohonen’s neural networks in the
recruitment process for sport swimming. Journal of Human Kinetics, 17, 75-88.
[2] Sun, D. (2015). Neural network-based table tennis competition technique and tactics diagnosis and
evaluation. Open Cybernetics & Systemics Journal, 9(1), 1722-1727.
[3] Huang, W., Lu, M., Zeng, Y., et al. (2021). Technical and tactical diagnosis model of table tennis
matches based on BP neural network. BMC Sports Science, Medicine and Rehabilitation, 13(1).
[4] Li, L., & Li, S. (2016). Volleyball teaching quality evaluation research based on artificial fish
optimization neural network.
[5] Chen, C. H., Chen, C. F., & Hsu, M. H., et al. (2013). A blocking prediction for volleyball using neural
networks. Applied Mechanics & Materials, 373-375, 1224-1230.
[6] Zhang, X. Evaluation model and simulation of basketball teaching quality based on maximum entropy
neural network.
[7] Wang, C. (2015). A table tennis technical and tactical analysis system based on back-propagation
artificial neural network. Modern Computer.
[8] Bo, P. Based on the BP neural network and the ACSI model to research on the effects of quality of table
tennis wrong big game experience.
[9] Peak, M. Implement the table tennis game by robot arm with forward-backward neural network.
In IEEE International Conference on Software Engineering & Service Science.
[10] Wu, M., & Liu, W. (2020). Case learning decision system for agricultural machinery based on
volleyball robot motion planning. Journal of Agricultural Mechanization Research.
[11] Fathirezaie, Z., Farsi, A., & Kazem, M., et al. (2014). Impact of cognitive training on efficiency of the
executive control network of attention on the table tennis players.
[12] Jie, W., & Yu, L. (2010). Use the combination of the decision tree and the artificial neural networks to
predict the outcome of table tennis matches. In Sixth International Conference on Natural
Computation. IEEE.
[13] Ren, Y. Q., Fang, Z. J., Xu, D., et al. (2014). Spinning pattern classification of table tennis ball's flying
trajectory based on fuzzy neural network. Kongzhi yu Juece/Control and Decision, 29(2), 263-269.
[14] Martin, P. E., Benois-Pineau, J., Péteri, R., et al. (2020). 3D attention mechanism for fine-grained
classification of table tennis strokes using a twin spatio-temporal convolutional neural network.
[15] Yang, Q., & Zhang, H. (2016). Application of BP neural network and multiple regression in table tennis
technical and tactical ability analysis. Journal of Chengdu Sport University.
Application and Analysis of Big Data Analysis in Sports
19
About the Author
Wang Rujun was born in Henan, China in 1988. He studied at Nanyang Normal University from 2007
to 2008 and obtained a bachelor's degree in 2008. He studied at the Henan Provincial Party School of
the Communist Party of China from 2015 to 2018 and graduated in 2018. Currently, he is working at
Zhumadian Preschool Teachers College. His research interests include physical education and
economics.
E-mail:wangrujun202411@163.com
