Academic Journal of Science and Technology 
ISSN: 2771-3032 | Vol. 5, No. 2, 2023 

 

140 

Smart	Pension	System	Based	on	Big	Data	Mining	
Algorithm	
Bowen Li, Dandan Tan*, Tao Zhang, Jialin Zhao, Zihang Fei, Junhao Zhai 

School of Computer and Software Engineering, University of Science and Technology Liaoning, Anshan 114051, China 
* Corresponding author 

 

Abstract: With the increasingly prominent problem of aging in our country, it has become an urgent task for our society to 
establish and improve a good old-age care system. At present, a new type of community smart elderly care model is gradually 
emerging, and has been actively explored and improved. In order to meet the needs of community elderly care services, big data 
mining algorithms are used to process data, and the smart elderly care system is connected with mobile APP and computer server 
equipment. Open source WebGIS technologies such as OpenLayers/Cesium, PostgreSQL/PostGIS, GeoServer, etc. are used as 
core supports, combined with front-end And back-end development technologies such as Vue and SpringBoot, etc., developed a 
mobile smart pension app and a web smart pension system based on open source WebGIS. This system can be more in line with 
my country's current smart elderly care service system, and accelerate the construction process of my country's intelligent elderly 
care. 

Keywords: Smart Pension System, Big Data Mining Algorithm, WebGIS, Vue, SpringBoot. 
 

1. Introduction 
With the increasingly severe aging problem in our country, 

establishing and perfecting the old-age security system has 
become an urgent task of our society. However, we are 
currently facing the following two practical challenges: First, 
although the construction of the elderly care service system 
has achieved initial results, the quantity and quality of service 
supply still cannot meet the actual needs. With the 
intensification of the population aging trend, the birth rate 
continues to decline and the total population growth is zero or 
even negative, making China likely to enter a stage of 
negative population growth, which will have a profound 
impact on the high-quality development of labor supply and 
consumption levels. Therefore, the most important issue to 
consider at present is how to maintain a long-term balanced 
development of China's population, which has a profound 
impact on the high-quality development of labor supply and 
consumption levels[1]. 

Since entering the 21st century, the Internet revolution is 
one of the most important technological revolutions in human 
society. The rapid development of the Internet has enabled 
industry and service industries to penetrate and integrate with 
each other. The rapid development of electronic technology 
and Internet technology has jointly promoted the 
development of new service industries including elderly care 
services and the innovation of service models. Smart 
technology has been widely used in home, community and 
institutional elderly care, and has achieved deep integration, 
gradually forming a new elderly care service system under the 
smart elderly care model, and realizing the digital, electronic 
and informational construction of the elderly care service 
system. 

2. Main Function Introduction 
The smart pension system is a real-time online monitoring 

platform with high reliability and stability. The system can 
obtain the health status of the monitored person in time, and 

start corresponding online health services through remote 
collaboration. At the same time, the system can also send 
alarm information to guardians to maximize the protection of 
the health and safety of the elderly. The smart elderly care 
online monitoring system realizes multiple functions such as 
elderly information management, daily monitoring 
management, medical doctor's order management, and 
accident management under the elderly account mode, and 
has relatively comprehensive services. The system mainly 
includes three user roles, namely administrator, elderly and 
children[2]. 

2.1. Main Functions of the Administrator 
Module 

2.1.1. Elderly Information Management 
This module mainly realizes the management of the basic 

information of the elderly by the administrator account, 
including the functions of adding, viewing, modifying and 
deleting the information of the elderly, and can also add and 
manage the accounts of the elderly and children. 

2.1.2. Daily Monitoring and Management Functions 
This module aims to realize the collection and monitoring 

of physiological data of the elderly, including heart rate and 
blood oxygen data collection, monitoring of spatial 
displacement changes of the object to be measured, real-time 
recording of geographic location information in abnormal 
situations, and abnormal physiological data alarm. When the 
health of the monitored person is abnormal, the system will 
promptly send information about the abnormal physical 
condition to the monitoring platform, and store and manage 
the data. At the same time, supervisors can remotely control 
terminal equipment on the monitoring platform, including 
setting environmental safety thresholds and turning on 
equipment, to ensure the stability and safety of the system 
operating environment. 

2.1.3. Doctor Order Information Management Function 
The function realized by this module is to add, view and 

delete the medical order information of the elderly. 



 

141 

2.2. Elderly Module Function 
The main functions under this module include the elderly 

viewing the administrator's records of their daily behavior; 
the doctor's order viewing; and the management of their own 
accidents; functions involving accidental addition, viewing, 
and deletion. 

2.3. Child Module Function 
The main function of this module is to enable children to 

view the information of the elderly, and to know the health 
status and daily activities of the elderly at any time, including 
daily monitoring information viewing, medical doctor's order 
viewing, and accident information viewing[3]. 

3. Realization of Software Functions of 
Smart Pension System 

During the development of this project, we used Vue and 
SpringBoot technologies, and used MySQL database for the 
entire data design. In the stage of designing the database, we 
carefully analyzed the requirements and designed the 
following six data tables: administrator password table 
(t_admin), elderly information record table (t_dengji), daily 
monitoring management table (t_jiankong), medical order 
management table (t_yizhu), accident management table 
(t_yiwai) and child information management table (t_zinv). 
The administrator has the highest authority of the system, and 
can add and delete the information of the elderly, manage 
daily monitoring, and manage medical orders. When the 
administrator logs into the system with his own account, the 
function menu options will be displayed on the left. The 
administrator can click the menu option, select the required 
operation module in the right interface, and further perform 
corresponding submenu operations. 

3.1. Elderly Information Management 
Interface 

In this interface, the administrator user can add the basic 
information of the elderly, including name, gender, age, phone 
number, address and login password. The elderly number will 
be automatically generated, which is the login account of the 
elderly user. When adding elderly information, you must 
ensure that the information is filled in completely, and any 
information cannot be empty, otherwise it will not be able to 
save and an error will be prompted. The main method is to 
check whether the user input data is empty through 
onsubmit="return checkForm()" to ensure the integrity of the 
data. After adding the elderly information, the administrator 
user can jump to the child information management interface 
through the child management button on the elderly 
information management interface. The system queries all the 
children's information of the elderly through the children's 
Action class, binds these data to the request for jumping, 
returns the corresponding JSP interface to display the detailed 
information of the children, and can also delete the children's 
information[4]. 

3.2. Daily Monitoring and Management 
Module 

After the administrator user logs in to the system, he can 
select "daily monitoring information management" in the left 
menu to open the monitoring information management 
interface. The background of the system will automatically 
call the Action class of the daily monitoring information to 

find the recorded daily monitoring information, and transmit 
the data to the foreground to display all the current daily 
monitoring information, so that the administrator can manage 
and view it. 

3.3. Medical Order Information Management 
Module 

After the administrator logs in to the system, he can open 
the medical order management interface by clicking the 
"Medical Order Management" module in the left menu. In this 
interface, the system will automatically call the Action class 
of the medical order to query all the added medical order 
information, and encapsulate the information into a list data 
set. Then, the system turns the page to the JSP page 
corresponding to the Action class, and the front end will 
display all the medical order information that has been stored 
in the database. Administrators can manage medical order 
information on this interface. 

3.4. Modify Login Password Interface 
After logging in to the system, the administrator can select 

the "Change Login Password" option in the function menu on 
the left to modify his own login password, which helps to 
prevent information loss caused by password leakage. When 
changing the login password, you need to provide the original 
password to operate. 

The purpose of the system test is to find out the errors in 
the program and the problems that may occur during the 
operation of the system in the process of simulating the actual 
use of the system. During the development of the system, the 
experimental data in the early stage is only added to test the 
development completion of the function. During a full system 
test, all previously recorded data is cleared. The administrator 
account type has higher authority. During the test, each 
module will be tested in detail to ensure that the function can 
achieve the design effect. Ultimately, the test results should 
show that all functions work well. 

4. Hardware Composition Structure 
and Algorithm Realization of Smart 
Pension System 

4.1. Implementation Process of Detection 
Alarm Algorithm 

The combined use of the main control processor MCU and 
MEMS multi-axis acceleration module in the smart watch (or 
the components in the mobile phone) can realize the detection 
of the elderly's body movement, tilt, fall, vibration, swing and 
other states. MMA7260Q three-axis acceleration sensor, as a 
micro-electromechanical intelligent sensing system, can 
quickly capture and start responding to sudden radio 
frequency signals in smart wearable devices within 1.0ms, 
including blood pressure, blood sugar, blood oxygen, pulse, 
heart rate and Collect data such as ECG, and convert different 
body monitoring accelerations into voltage output. There are 
mainly the following aspects in the implementation process 
of the fall detection algorithm for the elderly group during 
exercise[5]. 

The eigenvector of the human body during motion mainly 
includes acceleration (m/s2), acceleration vector amplitude, 
absolute average value of acceleration differential, and human 
body postures such as pitch angle (pitch), yaw angle (yaw), 
and roll angle (roll) horn. When high-intensity and violent 



 

142 

movements occur in the elderly group, the SVM peak value 
will continue to increase, and the acceleration vector 

amplitude is defined as:  

(x, y, z-axis acceleration respectively). 
The acceleration vector amplitude SVM of the elderly 

when they fall is usually SVM≥1.8gn, so the critical value 
of the first-level fall detection is set as SVM=1.8gn. Later, 
when the elderly are running fast, the acceleration vector 
amplitude SVM in the smart wearable device will also be 
greater than 1.8gn. At this time, it is necessary to set the 
absolute average value of the acceleration differential MADS 
as a secondary fall for the elderly group. Detection threshold. 
When the general old people fall, the absolute mean value of 
the acceleration differential MADS≥0.36g/s, the absolute 
mean value of the acceleration differential is defined 
as(T:time period): 

 

Under the constraints of the above two fall detection 
critical values, through the calculation of the relationship 
between the acceleration of the human body in the x-axis, y-
axis, and z-axis space, and the gravity of the human body, the 
pitch angle (pitch) and yaw angle ( yaw), the horizontal 
inclination angle θ1 and the vertical inclination angle 2 are 
used as the critical value of the third-level fall detection, and 
the definition of the human body posture angle is: 

 

 

4.2. Functional Information Detection Module 
and Algorithm Implementation 

In the smart elderly care system for residents, the 
monitoring system composed of intelligent sensor data 
measurement module, reader, wireless data repeater, 
embedded web server, etc., can realize the collection of blood 
pressure, blood sugar, pulse, body temperature and other data 
of the elderly and monitoring. Among them, the monitoring 
modules in smart wearable devices and handheld mobile 
devices include a variety of serial port/network protocols, test 
chips, and blood pressure, blood sugar, body temperature, 
pulse detection sensors and data transmission modules. 
Through GSM/GPRS communication technology, the 
collected data is transmitted to the community remote 
monitoring center for processing and analysis, and the final 
monitoring results are displayed on the control page. When 
the elderly have abnormal physical conditions, they can notify 
the nursing management staff through the one-key call system 
to carry out corresponding medical and nursing work. In this 
system, the reader and the wireless data repeater connect the 
smart wearable device and the embedded Web server, and the 
embedded Web server receives the transmitted human body 
information, and automatically executes the make fifo 
statement of the monitoring program, and converts different 
channel variables The dynamic numerical value is converted 
into a CGI file. When the user issues an access request for 
stored data such as blood pressure, pulse, body temperature, 
blood sugar, etc., the main program executes the submit 
command to transmit the values of various variables to the 
user port, thereby realizing the user's data access requirements. 
Through the organic combination of smart wearable devices, 

embedded Web server and GSM/GPRS communication 
technology, the elderly group physical condition monitoring 
system can realize remote monitoring and management, and 
provide safer, more comfortable and convenient home care 
services for the elderly. 

In the process of implementing this algorithm, the main 
program needs to use the switch statement to perform data 
transmission and control of different network interfaces. The 
algorithm of the gateway main program is implemented as 
follows: 

int main(int argc, char * * argv){ 
    int up; 
    int oldcarebaojing; 
    int fd; 
    if (argc!=3 || sscanf(argv[1],"%d", &oldcarebaojing) 
!=1 || scanf(argv[2]," %d",&up)!=1 ||up< 0 ||up>1  
|| oldcarebaojing< 0 || oldcare baojing> 3){ 
        fprintf(stderr,"WRUPG\n"); 
        exit(1); 
    } 
    fd=open"( /dev/oldcare_ oldcarebaojings", 0); 
    if (fd<0) { 
        fd=open"( /dev/oldcare_ oldcarebaojing", 0); 
    } 
    if (fd<0) { 
        perror"( open device oldcarebaojing"); 
        exit(1); 
    } 
    ioctl(fd, up, oldcarebaojing); 
    close(fd); 
    return 0; 
} 
According to the main program of the gateway above, it 

can be concluded that oldcarebaojing represents the type of 
alarm, 0 for no alarm, 1 for alarm, and the switches are 
respectively represented as Up and Close. With the help of 
fd=open"(/dev/Oldcare_oldcarebaojings",0) program, you 
can open and view the transfer files of different devices, and 
through ioctl(fd,up,oldcarebaojing), you can realize the 
control of input and output I/O pins of various devices. 
Afterwards, the caregivers of the intelligent elderly care 
system will take relevant measures to provide medical, 
nursing or first aid services for the elderly in the community 
based on the alarm information of different gateway main 
programs. 

5. Application and Promotion of Smart 
Pension System 

In this project, we completed the design and 
implementation of the smart elderly care system, and tested it. 
Although the system works well in a lab environment, there 
is still a lot of room for expansion, and we can consider the 
following: 

1) The location data modeling method for the elderly used 
in this project is just a simple processing method for location 
data. In the future, we can use other data preprocessing 
methods to reconstruct the location data of the elderly. 

2) In the input data of model training, environmental 
factors (such as temperature and humidity) in nursing homes 
can be added as influencing factors. 

3) In practical applications, doctors can be asked to classify 
the health status of the elderly in more detail (such as "cold", 
"fever", "insomnia", etc.), and then combined with machine 



 

143 

learning algorithms for modeling. 

4) In addition to supervised learning algorithms, we can 
also use unsupervised learning algorithms (such as KNN) to 
cluster data, observe the accuracy of prediction results and 
discover new activity characteristics of the elderly. 

5) The system requires a large-scale user trial to optimize 
system performance and fine art design for the user interface. 

6) Apply wireless sensor technology to home care, use 
Bluetooth and wireless wifi technology to make sensor 
information collection devices wireless, standardized, and 
low-cost, so as to broaden product audiences and consumer 
groups. 

7) Extend the health status prediction algorithm from 
specifically targeting the elderly population living in nursing 
homes to other populations (such as patients with Parkinson's 
disease). More types of sensor devices can be used to collect 
more types of user sign data to expand Scope of application 
of health status prediction models. 

6. Operating/Running Environment 
CPU: Intel Core i7-13700KF 5.4GHz; 
Running Memory: 4G and above; 
Operating System: Windows11; 
Browser: Chrome browser; 
Programming Software: Microsoft VSCode; 

7. Summary 
With the gradual increase of the aging population of urban 

and rural residents in our country, a nursing system covering 
location positioning, path tracking, vital sign detection and 
call alarm service will be constructed under the guidance of 
"big data mining algorithm" to monitor the elderly population 
at any time According to the living environment, physical 
condition, fall situation and location of the elderly, rescue and 
care for emergencies and accidental injuries of the elderly. In 
addition, based on Vue/SpringBoot, a smart elderly care 
system based on big data mining algorithms is designed and 
implemented, which can analyze and warn of various elderly 
body information, and solve the medical or nursing problems 
of urban and rural residents' elderly care management. 

In the future, the system will be researched in conjunction 
with the medical Internet of Things. According to the top five 
medical Internet of Things (IOMT) applications published by 
NIX Solutions on August 5, 2021, such as Wearable 
Biosensors, Automated Insulin Delivery, Connected Inhalers, 

Smart Thermometers (Smart Thermometers) and virtual 
hospitals/wards (Virtual Hospitals/Wards) meet medical 
needs. 

Acknowledgment 
Fund support: 2023 College Students Innovation and 

Entrepreneurship Training Program Project. 

References 
[1] Zhang Wei. Design of Smart Elderly Care System for Urban 

and Rural Residents Based on Internet of Things + Big Data 
Mining Algorithm [J]. Information Technology and 
Informatization, 2021(11): 228-230. 

[2] Wei Yufeng, Lu Min, Li Caihong, Wu Yuan. Research on Key 
Technologies of Smart Pension System Based on Seamless 
Scenarios [J]. Computer Application and Software, 2018, 35(6): 
330-333. 

[3] Zhou Shaoyan, Tan Qinhong. Design and Implementation of 
Safety Supervision Software for Smart Pension System [J]. 
Information Communication, 2018, 0(1): 84-86. 

[4] Yang Meiyan, Du Guangjie, Liu Yanjiao. Design and 
Development of Smart Elderly Care System [J]. Science and 
Technology Innovation, 2020, 0(9): 105-106. 

[5] Chen Xu. Research on Key Technologies of Smart Pension 
System Based on Internet of Things [J]. Wireless Internet 
Technology, 2017, 0(15): 134-135140. 

[6] Du Danyang, Li Honghai. Design of voice control terminal for 
smart elderly care system [J]. Industrial Control Computer, 
2018, 31(12): 161-162164. 

[7] Zhang Yufeng, Qi Guoyu, Zhou Yaojian, Yuan Chenxun. 
Research on Smart Pension System Based on Artificial 
Intelligence and Internet of Things [J]. Electronic Technology 
and Software Engineering, 2021(13): 206-207. 

[8] Li Shiqi, Duan Luyang, Liu He, Chen Yan. Design of New 
Smart Pension System Based on Internet of Things [J]. Chinese 
Science and Technology Journal Database (Full Text Version) 
Engineering Technology, 2022(10): 0050-0053. 

[9] Lin Zhengmin, Wang Shue, Yu Xiaoyi, Wei Dequan, Yang 
Weilan. Design of Smart Pension System Based on Django [J]. 
Computer Knowledge and Technology: Academic Edition, 
2023, 19(2): 25-28. 

[10] Zhang Jinyao, Zhang Jinrun. Design of Smart Pension Online 
Monitoring System [J]. Internet of Things Technology, 2022, 
12(7):20-21.