An empirical study on the improvement of students’ physical fitness and health in college physical education programmes based on big data

With the rapid development of the economy, the continuous change of society, and the improvement of people’s living standards, the level of students’ physical fitness level is declining year by year due to the excess of nutrition. The students have begun to rise in body weight year by year, and the level of physical health status is showing a decline in health year by year [1–2]. Nowadays, various realities are not conducive to the healthy development of the student population, such as sedentary, cell phone addiction, irregular work and rest, poor self-control and other internal factors, as well as busy school work, family environment, exposure to social affairs and other external factors [3–4]. The above situations have resulted in developmental problems such as lack of physical exercise, serious vision problems caused by excessive eye use, poor physical functioning, and overweight, which have become important social problems affecting the healthy development of students [5–7]. As the most important growth stage in the whole life cycle, how to ensure that students’ bodies develop well and reach the level of physical health that the country expects is an issue that society needs to pay strong attention to.

Moreover, the interest in sports activities and the degree of participation in sports during the student period will have a profound effect on whether students participate in sports in adulthood [8–10]. Students are the future and hope of the motherland, and the health of the students is the future health of the motherland. If we can not do a good job of supervising and inspecting students’ health, then the future of the motherland is not a perfect guarantee [11–12]. Therefore, based on the current trend of students’ physical health levels declining year by year, it is necessary to put forward the urgent need to promote the reform of the majority of colleges and universities’ physical education curriculum only through the autonomy of colleges and universities ontology concern, can be a benign intervention on students’ physical health, in order to reverse the adverse development trend of students’ physical health declining year by year and promote the overall healthy development of students [13–15].

For today’s serious trend of decreasing physical fitness of students in colleges and universities, the problem of physical fitness improvement is urgent for colleges and universities. Literature [16] mainly investigated the status of physical fitness level and the reasons for the deterioration of male and female students in grades 5-11 and concluded that exploring innovative technologies for organizing physical education in general education institutions is an urgent problem in the theory and practice of physical education, which can maintain and promote the physical fitness of students. Literature [17] points out that the loss of students’ awareness of self-development creates the problem of healthy lifestyles and argues that health-oriented sports activities can form and maintain students’ physical health. Literature [18] explores the methods of the most accessible forms and methods of physical culture for improving health, aimed at improving the physical development of students, reducing mental, physical and emotional stress, thus enhancing their physical health preventing and treating diseases.

Students’ physical fitness and health status have always been the focus of school physical education, and school physical education programs are in the process of continuous improvement for this purpose. Literature [19] proposed a multi-component school-based intervention (SELF-FIT) based on self-determination theory and designed a controlled experiment to investigate the effects of the intervention on students’ moderate to high intensity physical activity in PE classes, and the results of the experiment showed that SELF-FIT could improve students’ health and well-being by enhancing high intensity activity in PE programs. Literature [20], in response to the finding that lack of free choice of type of exercise is the main reason that prevents most students from engaging in physical fitness and physical activity, suggests the implementation of physical education oriented lessons in physical education in order to increase students’ motivation to engage in physical activity and sport, which in turn promotes the development of morphological functioning and physical fitness. Literature [21] takes students of special medical groups as an example and explores the quality of the process of teaching physical education in special medical groups through teaching observation and mathematical and statistical methods, and empirically analyzes and finds that the quality of the process of teaching physical education in special medical groups depends on the interconnection of all the components affecting their efficiency, and the results provide the scientific basis for the study of the direction of improving the students’ physical fitness and health. Literature [22] examined whether the programs conducted according to the developed methods of teaching physical education contribute to the acquisition of health and fitness by students, and it was experimentally verified that the high efficiency of methods of physical education improves the physical fitness of students and their motivation for a healthy lifestyle and their health.

This paper constructs a physical education IoT platform based on heart rate monitoring, which combines radio frequency identification technology to connect mobile smart devices to achieve the purpose of transmitting heart rate data. The wireless network is used by the mobile smart device to upload the heart rate data to the server to ensure that users can obtain the required information in a timely manner. The decision tree algorithm for selecting the content of sports courses for college students classifies the heart rate monitoring level of college students by standardising the metrics of each project, which indicates the different impact of different projects on the physical fitness of college students, and provides suggestions for adjusting the arrangements of sports courses for highly efficient students, so as to improve the physical fitness of students. Finally, the application performance of the methods in this paper is tested. The comparison was made between the execution time and monitoring accuracy of different heart rate monitoring methods. Comparative physical fitness experiments were also conducted on students using the physical fitness enhancement methods presented in this paper.

2

Conceptualising the design of an Internet of Things (IoT) platform for physical education based on heart rate monitoring

2.1

Platform architecture

The author believes that the Internet of Things (IoT) platform for physical education based on heart rate monitoring is a kind of network in the form of IoT from people to things, with the heart rate testing instrument (hereinafter referred to as the heart rate apparatus) as the sensor, and the heart rate data is transmitted through the wireless radio-frequency identification (RFID) technology by connecting the mobile smart device installed with specific software, and the mobile smart device uploads the heart rate data to the server through the wireless network, and the user obtains the required information. The structural design of a single class, for example, is shown in Fig. 1:

2.2

Methods of operation of the platform

1)

Before the lesson, the PE teacher distributes the heart rate device to each student to wear, uses the mobile smart device to identify each heart rate device number, and binds each student’s information to their respective heart rate device through the teaching software.

2)

During the lesson, the heart rate device records and saves the students ’ heart rate data, and when the heart rate exceeds the warning value, the heart rate device emits a vibration alarm to remind the students to take appropriate breaks or reduce the intensity of exercise.

3)

After the lesson, students return the heart rate device to the teacher, who uploads the data of the current heart rate lesson to the mobile smart terminal for saving.

4)

After all the daily lessons are finished, the teacher uploads all the heart rate data of the day to the web server.

5)

In extracurricular sports activities, students in economically developed areas can be asked to each be equipped with a heart rate device, physical education teachers can assign extracurricular homework in physical education, individual students or their parents can install heart rate monitoring software in their mobile phones that integrate physical education teaching, bind the heart rate device with their mobile phones, and record in real time the heart rate data of the students who participate in extracurricular sports activities, including the heart rate curve, the time of the exercise, and the average heart rate. Students should send their heart rate data for participating in extracurricular sports activities to the teacher or upload it directly to the web server as proof of completing the extracurricular sports homework.

6)

Students, teachers and education supervisors can access the heart rate data of individuals, classes, different sports and courses through the web server and then improve the physical health of the corresponding student groups through the physical health optimisation algorithm accordingly.

2.3

Analysis of the construction technology of the Internet of Things platform for physical education based on heart rate monitoring

2.3.1

Application of sensor technology

Currently, the mainstream heart rate devices are heart rate bands and heart rate bracelets, and there are two testing principles: the ECG test method and the photoelectric volumetric pulse wave recording method. 1)

Electrocardiogram (ECG) test method [23].

Heart rate bands are mainly measured using this technique. The sinus node excites the atria and ventricles through a specific pathway and time course, leading to the excitation of the entire heart. The advantages of monitoring by heart rate belt are high accuracy, continuous high precision monitoring during exercise, cheaper due to the production and part of the foreign brands of heart rate belt price is about 200 yuan. The disadvantage is that the heart rate belt needs to be worn near the chest and close to the skin in lower temperatures, and girls will be more complicated to wear.

2)

Photoelectric volumetric pulse wave recording method (PPG)

Existing heart rate watches, or sports bracelets (hereinafter referred to as heart rate bracelets) mostly use this technology. The heart rate bracelet is worn in the same way as a watch, on the wrist, with the core part consisting of two green wavelength light-emitting LEDs and a lightsensitive sensor, which is located on the back of the heart rate watch and is worn close to the skin of the hand and hip. A heart rate signal bracelet has the advantage of being easy to wear. The disadvantage is that the test accuracy and stability are susceptible to interference by external conditions, skin colour shades, vibration, light, sweat, and hair will affect the accuracy of the measurement, the accuracy of the heart rate bracelet is more expensive, the price is generally more than a thousand dollars.

Through the above comparison, the heart rate band is more suitable as a sensor for heart rate monitoring in sports teaching.

2.3.2

Information recognition, connectivity and transmission technologies

1)

Bluetooth

A wireless technology standard that enables the connection of multiple devices and the exchange of information between fixed or mobile devices and personal area networks over short distances. Pairing usually includes a certain degree of user interaction and confirmation of the device ID; after successful completion of the pairing, a connection will be formed between the two devices, and there is no need to repeat the pairing process in order to confirm the device ID when connecting again in the future.

2)

Near Field Communication (NFC) Technology

NFC near-field communication technology has evolved from the integration of non-contact radio frequency identification (RFID) and interconnection technology, combining inductive reader, inductive card and point-to-point functions on a single chip, which can be used to identify and exchange data with compatible devices within a short distance, and generally adopts two reading modes: active and passive.

Bluetooth data transmission speed is fast, but the connection time is long. NFC technology, although the transmission distance is short, can only realise point-to-point data transmission, but the connection speed is fast, and the energy consumption is very low or no energy consumption, through the contact connection and data transmission, in large-scale physical education activities effect and efficiency may be better than Bluetooth. In summary, NFC technology is more suitable for data connection and transmission in physical education.

2.3.3

Mobile smart devices

With mobile smart phones or smart tablets as the main focus, the functions and configuration of current mobile smart phones are highly developed, with CPU processor frequency reaching 2.35GHZ and storage space up to 128GB, whose computing and storage capacity even exceeds that of the computers of a few years ago. The vast majority of mobile smart devices sold in the market are based on the operating platforms of iOS and Android systems. A large number of APP applications have been developed. Android, led and developed by Google and the Open Handset Alliance, is an open source operating system. Currently, most of the world’s smartphones and tablets use the Android operating system. Android mobile devices are inexpensive. Less than 1,000 yuan smart mobile devices can handle heart rate data very well. According to an August 2016 statistic, Android captured 82.2% of the world market share, compared to only 14.6% for IOS. Choosing to develop heart rate monitoring software for physical education based on Android by adding external NFC devices is more universal.

2.3.4

Web-based service platforms

By renting or purchasing a server and designing a website, mobile smart terminals can transmit data to the server cloud, and students and individual teachers can log in through the website to check the heart rate data of their own and their classes’ lessons and adjust the design of the classroom according to the data. Education regulators can monitor the quality of physical education teaching through big data, and overall control the adjustment and formulation of relevant standards and policies.

3

Algorithms for analysing university student fitness data

3.1

Principles of Decision Tree CART Algorithm

The CART classification algorithm has an indispensable role in the computational classification process of heart rate monitoring sample data of students in higher education [24]. If there are K attributes in the data set and the probability that a certain sample value belongs to attribute k is P_k, then the Gini coefficient of the probability distribution of attribute k is calculated by the formula: 1 $Gini (p) = \sum_{k = 1}^{k} p_{k} (1 - p_{k}) = 1 - \sum_{k = 1}^{k} p_{k}^{2}$

For the heart rate monitoring data sample M, the number of data is [M] and Ck denotes the subset of the sample M that belongs to attribute k. The formula calculates the Gini coefficient of the heart rate monitoring data sample M: 2 $Gini (M) = 1 - \sum_{k = 1}^{k} {(\frac{| C_{k} |}{| M |})}^{2}$

Sample M needs to be partitioned according to attribute A at all points that take the value a to get the two parts Mi and Mz then the Gini coefficient of sample M under attribute A is defined as follows: 3 ${Gain}_{G} ini (M, A) = \frac{| M_{1} |}{| M |} Gini (M_{1}) + \frac{| M_{2} |}{| M |} Gini (M_{2})$

For the sample M attribute set {A₁,A₂,…A_n}, calculate the arbitrary attribute values of each attribute separately and divide the sample M into two parts GainGini(M, A), select the smallest value as the optimal dichotomous attribute value obtained from attribute A all the optimal dichotomous attribute values of the entire attribute set, and select the smallest of them as the optimal dichotomous scheme of the sample M: 4 $Min (GainGini (M, A))$

In CART classification tree, discretisation is required for continuous features, which is based on the principle that the continuous feature A in the sample D has m values {a₁,a₂,a₃,…a_m}, and the sorting operation is performed on A first: 5 $S o r t (A)$

The two neighbouring element values of feature A are averaged to make a discrete feature, taking m-1 values, denoted as: 6 $T_{i} = (a_{i} + a_{i + 1}) / 2$

After calculating the Gini coefficient with T_i as the dichotomy point respectively, and obtaining the minimum Gini coefficient on a_t, the values of the A characteristic at greater than a_t and not greater than a_t are grouped into two categories.

3.2

Decision Tree Algorithm for Analysing College Physical Fitness Data

It is not difficult to find out that in the decision tree CART algorithm, the ratio of frequency to the number of samples is used for the calculation of probability P_k, which requires the continuous features present in the samples to be discretised in order to perform the relevant probability calculation. The average of the two neighbouring elements after sorting is used as the discrete feature for continuous features for probability calculation. This method is applied to college students’ heart rate monitoring data [25] in terms of purely classifying and predicting college students’ heart rate monitoring grades, for the heart rate monitoring data of pull-ups is obviously more advantageous than this kind of data of standing long jump, which can only indicate that the item is good for distinguishing heart rate monitoring grades in the results, but the degree of the item’s influence on the heart rate monitoring grades should also be reflected in the evaluation. Therefore, in the process of discrete processing of continuous characteristics of heart rate monitoring data of college students, the data should be unified, and this study proposes to improve the discrete processing of continuous characteristics of heart rate monitoring data of college students as follows:

For T_i the computation is improved as follows: for test items {A, B, C, …F}, where A has a score a_i and its weighted conversion score ac_i, for continuous features A has m values {a₁,a₂,a₃,…a_m}, the conversion is (ac₁,ac₂,ac₃,…ac_m}, and T_i is computed as: 7 $T_{a i} = \frac{a c i}{a c i + b c i + ... + f a}$

After the measurement standardization operation of each item, the advantage of the range of data types on the results can be avoided, and the results can classify the heart rate monitoring level of college students, and the hierarchical level can also show that the differences of different items have different degrees of influence on the physical fitness of college students from the side, and then convert the weighted conversion score of each item ac_i into a_i for decision tree input when the model outputs, so as to provide guidance and suggestions for the personalized arrangement and adjustment of students’ physical education courses.

3.3

Steps for establishing the algorithm for analysing college students’ fitness data

Algorithm building process:

Input training data set M and output decision tree model. Step 1:

For the input data set M. If the input data set has no features or the number of data is less than a threshold, output the decision tree model.

Step 2:

Calculate the Gini coefficient of the current data set M. If the Gini coefficient of the current set M is less than the preset threshold, return the decision tree subtree of the current node.

Step 3:

For each attribute A in the current data set M, for all values a of attribute A, for each a value, divide M into M₁ and M₂ parts and compute Gain_Gini(M, A).

Step 4:

For all the values a of all the features A, the smallest Gain_Gini(M, A) is selected as the optimal attribute of the current node and the optimal cut-off value under the current attribute. The optimal cut-off value of the current data set M at the current optimal attribute is split into two parts: two sub-data sets are generated sequentially, and the current data set M is assigned to the two subsets according to the current optimal attribute.

Step 5:

Recursively calls the above steps for the two sub-data sets generated by the above steps until the stopping condition for returning the decision tree model is satisfied, the Gini coefficient is less than a predetermined threshold, or there are no more features. The decision tree model T is generated and output.

The implementation of the CART decision tree uses the Python language to manage its excellent data analysis pandas library li visualisation fried matplotlib for modelling, the specific process is shown in Figure 2.

1)

Data preprocessing is performed to analyse the data set for outliers, and the data set is proportionally partitioned using panda’s library.

2)

Project selection of training test data sample set as features in the decision tree according to the decision tree building process.

3)

Train the training dataset on the feature set selected in step 2.

4)

Cross-validate the initially formed decision tree collection using the validation dataset for evaluation and tuning parameters.

5)

Use the prediction dataset to predict and derive a prediction model for the cross-validated decision tree.

6)

Visualise the decision tree model.

4

Analysis of simulation experiments

In order to test the performance of this paper’s method in achieving the application of heart rate monitoring for college students, the simulation test is carried out in the MATLAB platform in combination with electronic measurement and control technology. The running environment of the experiment is:

Processor: Intel(R)Core(TM)i5-2520M CPU@3.50 GHz (4 CPUs)

Memory: 3500 MB RAM

Operating system: Microsoft Windows 2010 Professional (5.1, version 2600)

The data was obtained from students of one of the majors at a university, including all students in any three classes, totaling 150 people and a volume of 1500 MB of physical data. The tester developed by the university was used to measure the body mass data of the study subjects over a period of one month, the privacy information of the initial data was anonymised and confidential, and only the relevant body mass record information used for the experiment was used. The execution times and monitoring accuracy of different methods were compared.

4.1

Execution time test

In order to verify the advantages of this paper’s method in heart rate monitoring data, the execution time is used as an experimental index. Based on the above experimental environment, the 1500Mb body data volume was examined. The shorter the execution time, the higher the detection efficiency. Three methods of manual measurement, stethoscope measurement and ECG measurement were selected to compare with the method of this paper, and the specific experimental results are shown in Figure 3. The heart rate monitoring method in this paper can complete the detection of all the data in a short execution time, and the execution time is only 57.5s when the volume of body data reaches 1500Mb, which is smaller than the execution time of several other monitoring algorithms. It shows that the method can efficiently achieve the monitoring of heart rate data of college students in college physical education courses. The method reduces monitoring data execution time by performing fusional scheduling of students’ information.

4.2

Data monitoring accuracy tests

Taking the actual measurement values as a reference, the accuracy of manual measurement, stethoscope measurement, ECG measurement and this paper’s method for monitoring college students’ physical education course centre rate data was tested, and the higher the accuracy, the better the monitoring effect was, and the comparison results were obtained as shown in Fig. 4. The serial numbers 1-4 on the Y-axis represent manual measurement, stethoscope measurement, ECG measurement and this paper’s method, respectively. According to the results in Fig. 4, it can be seen that comparing the other three methods, the accuracy of this paper’s method in monitoring college students’ health data is higher than 85%. The accuracy of several other comparison methods is slightly lower than that of this paper’s research method, which indicates that the data monitoring effect of this paper’s method is better and more credible.

5

Analysis of physical fitness test results

The number of cycle times of physical education classes in the experimental and control groups was the same, and the experimental group had the same teaching plan as the control group, but the experimental group would make corresponding adjustments in the teaching content according to the student’s heart rate performance during the course, based on the algorithm for analysing college students’ physical fitness data.

Students in each group wore heart rate monitoring bracelets. Students in the control group wore the bracelets only to provide heart rate data for comparison and analysis at a later stage, and there was no intervention with the college physical fitness data analysis algorithm, i.e., the control group did not receive any feedback or intervention. The students in the experimental group wore the heart rate bracelets to monitor and collect their heart rate data and also to provide real-time feedback so that the physical education teachers could implement personalised training for different students according to the real-time heart rate data and according to the algorithms of college students’ physical fitness and health data analysis.

At the same time, during the experimental period to communicate with students and parents, the groups of students, in addition to physical education classes, do not carry out other strenuous sports, as usual life. During the experiment, we also need to know more about the physical activity of the experimental group of students after school and compare between different groups to reduce the experimental interference factors as much as possible.

5.1

Analysis of the results of the pre-test physical fitness test

Before starting the experiment, a test of physical health indicators was done for both groups of students (experimental group, control group). To know the physical health status of each student, the sample size of this measurement was 100. The measurement data is shown in Table 1. As can be seen from Table 1, in the test conducted before the experiment, there is not much difference between the experimental group and the control group in all indicators, and the P-value is greater than 0.05. The test data shows that there was no significant difference between the two groups of students before experimenting, and there was no significant difference in the physical health indicators. Note: There is a statistical difference of P<0.05 and a statistically significant difference of P<0.01.

Table 1.

Student physical health data

Survey index	Control group	Experimental group	P
Height(cm)	140.22±4.15	141.35±5.45	0.208
Weight(kg)	39.27±6.96	39.35±7.44	0.438
Lung capacity(ml)	1844.47±346.97	1904.12±418.15	0.189
50m run(s)	10.24±0.18	10.27±0.46	0.957
Preflexion(cm)	12.55±4.43	12.28±5.42	0.784
Jumping rope in a minute(per)	136.41±25.46	138.45±24.36	0.136
One minute sit-ups(per)	30.64±6.79	31.75±9.24	0.217

5.2

Comparative Analysis of Overall Average Heart Rate of Students

The average heart rate is measured by dividing the total number of times a student’s heart rate went up during a physical education class by the total time spent in that class. The average heart rate of students in PE class is 120-200 beats/minute, and the average heart rate is also a key indicator for determining the amount of exercise load in PE class.

In the new standard, the average heart rate of the physical education class reaches 130 beats/minute is in line with the standard. Research has shown that the heart rate range for aerobic exercise is 110-150 beats/min, and the more accurate range is 120-140 beats/min, so the optimal aerobic metabolism is the middle point of 130 beats/min, which is a characteristic indicator, and is also a general indicator of the intensity of exercise load for physical education students.

The graph of changes in the overall average heart rate of the students in this experiment is shown in Figure 5. Ten sports were conducted, and each sport was conducted for approximately 7-8 physical education lessons. The experimental and control groups had the same teaching program, and from the monitored heart rate data change graphs, it was found that the overall average heart rate of the students in the control group was at 125 beats/minute and less frequently up to 130 beats/minute. From the graph, it was seen that the average heart rate of the students in the 60th and 61st sessions was at its peak, as these two physical education sessions were football matches, one with a combination of men and women in two groups against each other, and the other with a group of boys in a group of girls in a group of men and women, and it was found that the students invested more energy in the process of the matches, and the intensity and density of the exercise was enhanced, and the average heart rate increased a lot accordingly. The overall average heart rate of the students in the experimental group was in the range of 135-145 beats/min, and with the intervention of the college students’ physical fitness data analysis algorithm, the students’ average heart rates were all above 130 beats/min. The monitoring tools proposed in this paper are effective in improving the average heart rate of students during physical education classes. The overall average heart rate of students in different programmes is shown in Figure 6, in which it can be seen that the average heart rate of students in Programme 1 is higher than that of students in other programmes, and the average heart rate of students in the experimental group reaches 140 beats/minute. The reason is that the curriculum of Project 1 not only includes the study of basic skills and basic movements but also includes confrontational competitions, which makes the average heart rate of the students higher than that of other projects.

5.3

Data analysis of post-test fitness test results

At the end of the experiment, the students of the two groups were post-tested on the physical fitness test, and Table 2 shows the results of the comparative analysis of the physical fitness test data of the two groups of students after the teaching experiment. From the point of view of the lung capacity index, the p-value of the experimental group and the control group is 0.0079, indicating that there is a significant difference between the two (P<0.01) from the data obtained from the seated forward bending, there is no big difference between the two groups, but the experimental group’s performance improves in comparison than the control group improves significantly. In the 50-metre run, the experimental group’s performance was better, but there was no significant difference from the control group. In the sit-up programme, it can be seen that there is still a difference between the experimental group and the control group (P<0.05). Similarly, in the 1-minute rope skipping programme, there was a difference between the post-test scores of the experimental group and the control group (P<0.05). Therefore, from the analysis, it can be seen that in sports, according to the college students’ physical health data analysis algorithm proposed in this paper for students to carry out scientific guidance to control the time of exercise and load so that the students’ physical function aspects to get effective exercise.

Table 2.

Student health after health survey data

Survey index	Control group	Experimental group	P
Height(cm)	140.44±3.15	141.25±5.13	0.211
Weight(kg)	38.27±6.14	38.78±7.12	0.432
Lung capacity(ml)	1944.35±345.13	2379.49±456.37	0.0079
50m run(s)	9.24±0.23	9.11±0.42	0.716
Preflexion(cm)	12.55±4.43	12.28±5.42	0.651
Jumping rope in a minute(per)	139.47±26.47	145.13±27.38	0.042
One minute sit-ups(per)	33.64±5.46	39.75±8.79	0.041

From this point of view, in the physical education classroom, the teacher adopts the decision tree algorithm of college students’ physical fitness optimisation to provide students with scientific guidance and effective monitoring of students’ heart rate values as a basis for adjusting the time and load of exercise, so that students can carry out scientific exercise and effectively improve their physical fitness.

From the experiment, it can be seen that effective preparation before class can make students more effective at exercising, enhance lung capacity, and improve lung capacity. Many aerobic exercises are widely used, such as middle and long-distance running exercises, ball sports, and jumping rope. The waist is the distribution of human body strength. The key point is to play the role of the bond connection, bearing the top and bottom is the powerpoint of all the action, and if there is no this bond or some problems, such as the bond to withstand the strength, is not enough, will lead to the exercise of the power, action to do not standard, and may even lead to the body of the force point of the change so as to cause the wear and tear of the body. The symptoms are often concealed in the body’s depths, usually unnoticed, but the outbreak is the result of cumulative damage. Therefore, this part of the body is the most important part of the person’s daily life, and scientific research has shown that situps can improve this muscle strength significantly and at the same time, feed on this exercise to improve the quality of movement.

6

Conclusion

1)

The heart rate monitoring method in this paper can complete the monitoring of heart rate data in a shorter execution time. Not only the execution time is less than several other algorithms when the data volume is small, but also when the volume of body mass data reaches 1500Mb, the execution time is 57.5s, which is still much smaller than the execution time of several other monitoring algorithms. The accuracy of the monitoring methods in this article for monitoring the health data of university students is above 85%. The accuracy of several other comparison methods mentioned in the article is lower than that of the heart rate monitoring method proposed in this paper, which indicates that the monitoring heart rate method in this paper can efficiently achieve the monitoring of the heart rate of college students in college sports courses.

2)

The overall average heart rate of the students in the control group was 125 beats/minute, which is less likely to reach the heart rate index for optimal aerobic exercise. Under the intervention of the decision tree algorithm for college student fitness optimisation, the average heart rate of the experimental group was above 130 beats/min, and all of them reached the heart rate index for optimal aerobic exercise. The decision tree algorithm intervention in college students’ fitness optimisation based on heart rate monitoring has been shown to be effective in improving the average heart rate of students in physical education classes.

3)

The p-value of the lung capacity index of the experimental group and the control group is 0.0079, which is a significant difference. Although there is no big difference in the performance between the two groups of sitting forward bends and the 50-meter run, the performance of the experimental group is improved significantly more than that of the control group. In the sit-ups and one-minute rope skipping, there is a difference between the experimental group and the control group compared to the experimental group. This shows that scientific exercise guidance for college students in physical education courses based on the heart rate monitoring method and college students’ physical fitness optimisation decision tree algorithm proposed in this paper can effectively improve students’ physical function and physical fitness.

Language:: English

Publication timeframe:: 1 times per year
Journal Subjects:: Life Sciences, Life Sciences, other, Mathematics, Applied Mathematics, General Mathematics, Physics, Physics, other

Journal RSS Feed

An empirical study on the improvement of students’ physical fitness and health in college physical education programmes based on big data

Jun Chen

Published Online: Feb 05, 2025

Received: Sep 21, 2024

Accepted: Dec 30, 2024

DOI: https://doi.org/10.2478/amns-2025-0050

KeywordsHeart rate monitoring, Decision tree algorithm, Physical fitness, Mobile smart device

© 2025 Jun Chen, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Keywords
Heart rate monitoring, Decision tree algorithm, Physical fitness, Mobile smart device