Research on Urban Security Management Based on Cluster Analysis

of 110 Alert Quantities

Shiwen Cui

Shandong Police College, Jinan, Shandong, China

Keywords: Urban Security Management, Cluster Analysis, 110 Alert Quantities.

Abstract: With the development of social economy and wide spread of Internet Technology, mobility and openness

level of the whole society has been greatly improved in recent years. Disposal of 110 Alert and fight against

urban crime has become more and more sophisticated, intellective, covert and associated. Police Departments

are required to adjust police force distribution, investigation means, and the operating mechanism, con-

summate 110 Alert Commanding & Allocating System, and improve their ability to deal with the intricate

urban public security situation. In this article, we will analyze functions of 110 Command Center in urban

public security management, make a Cluster Analysis of 110 Alert Quantities based on the Hierarchical

Clustering Method and K-means, and raise reasonable advice in order to help policemen on duty form a

scientific expectation of 110 Alert Quantities and make enough preparation for 110 Alert.

1 INTRODUCTION

In recent years, mobility and openness level of the

whole society has been vastly improved, with the

development of social economy and wide spread of

Internet Technology. While urban crime has become

more and more intellective, covert and associated,

disposal of 110 Alert and combat against urban crime

has become more and more sophisticated. (Yang,

2019) Information Technology being integrated with

human productive activities, worldwide data increase

explosively and gather massively. Data Technology

has become a significant factor determining a city’s

management level of public security. (Xu, 2018)

There exist a certain actual problems in the present

police operation pattern. Public Security Departments

need to grasp the opportunity of Data Technology

development, make full use of substantial data re-

sources, integrate police affairs with Data Technol-

ogy, and make the police operating pattern more

intelligent and advanced. (Li, 2013) They are also

required to adjust police force distribution, investi-

gation means, and the operating mechanism, con-

summate 110 Alert Commanding & Allocating Sys-

tem, and improve their ability to deal with the intri-

cate urban public security situation.

2 CHALLENGES WE MEET IN

URBAN PUBLIC SECURITY

MANAGEMENT

First, more sources of risks have appeared with the

development of urbanization. Some problems that

ever appeared in the past become more challenging in

an information society, since the crowd can easily

acquire and transmit concerning information, make

remarks, which include environment pollution, dis-

putes between employers and workers, medical dis-

putes, accident disputes, and internet financial risks.

Traffic security risks also increase, as grades and

mileage of urban roads improve, vehicle possessing

quantity increasing, traffic safety consciousness of

drivers remaining weak. Procedure of hazardous

articles’ production, storage, transportation, sale,

utilization is sophisticated and intricate. Some prac-

titioners focus only on profit, neglecting safety during

the course of producing, transporting and utilizing

hazardous articles. Without strict supervision and

control, the industry of hazardous articles may bring

about serious risks towards urban public security.

Individual extreme crimes still exist, which may lead

to grievous harm to others since density of urban

population is higher than rural population. A person

who holds a hostile attitude towards the society tends

to make extreme violent crime. Unluckily, this kind

Cui, S.

Research on Urban Security Management Based on Cluster Analysis of 110 Alert Quantities.

DOI: 10.5220/0011737200003607

In Proceedings of the 1st International Conference on Public Management, Digital Economy and Internet Technology (ICPDI 2022), pages 373-377

ISBN: 978-989-758-620-0

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

373

of people exist in the whole human history, so do they

nowadays. They look as normal as others on ordinary

days. Once taking actions, they become violent,

ferocious and cruel. It costs a lot of time and police

power to investigate and prevent an unknown per-

son’s extreme crime. Therefore, police’s pressure of

dealing with terrorism is also very high.

Second, development of we-media makes urban

security management more challenging. Some people

do not exactly understand the balance of social rights

and social responsibilities, pursing their rights

through unreasonably hyping up on the internet and

disturbing the cyber order. These people are selfish,

littered with self ego, paying little attention to others’

feelings, and only caring about their own interest and

feelings. If only their interest is aggrieved, they will

exaggerate and widespread their situation on the

internet, incurring much attention. Some social con-

flicts concern a large quantity of people in various

regions, which is difficult for the police to deal with

effectively. In the past, civil disputes are concentrated

on family, marriage and debt, but now they have

covered environment protection, tort, removing old

houses, inappropriate judgment etc. In the past, con-

flict parties are limited to citizens and citizens. Now,

they include citizens and factories, citizens and cor-

poration, citizens and administrative units, factories

and administrative units etc. Diversification of con-

flict parties bring much work to grass-root police

departments. Conflicts and disputes are closely re-

lated to people’s production and life. If an individu-

al’s appeal is not satisfied, he will turn to the internet,

find people who have the same appeal, and maximize

their interest in an extreme way. People who have

ulterior motives may make use of these conflicts and

disputes to spread cyber rumor and false information.

Some cynical, pessimistic and hostile people are

usually influenced by these people with ulterior

motives, paying special attention to cases concerning

police and other government officials. If police de-

partments make response inactively, they may incur

harsh criticism on the internet, and cyber public

sentiment crisis may turn to public security problems.

Once trapped in a cyber public sentiment crisis,

police departments may lose authority and credibil-

ity. Whatever they say or do, the crowd on the in-

ternet do not believe, causing harm and erosion to the

whole social credit system.

Third, New forms of crime and risks appear in

urban regions. While quantities of traditional forms

of crime have decreased, new forms of crime have

appeared and increased rapidly, including internet

fraud and cyber financial crime, etc, which have

encroached on a large number of people, being dif-

ficult to prevent and deal with. Internet fraud and

telecommunication fraud are two most widespread

forms of urban crime, affecting a large quantity of

people and causing a large amount of property loss.

Most of the victims of internet fraud and telecom-

munication fraud are elderly people, who live alone

and owe a large of amount of money to support

themselves and prepare for possible diseases. They

are liable to believe in others, and easy to cheat.

Young people who indulge in the cyber game may

become victims too. Some criminals go abroad and

make crime of internet fraud and telecommunication

fraud, making it difficult for the police to investigate

and capture them. It needs international police

co-operation if domestic police want to capture and

arrest them. With the development of online shop-

ping, the number of concerned fraud cases increases

rapidly. Posting and delivery employees are able to

collect information of senders and receivers, and may

abuse their individual information, causing potential

safety hazard. Practitioners of hotels, especially

private hotels, also have the chance to collect and let

out the crowd’s private information. Once fraud

criminals have mastered citizens’ private infor-

mation, they may make concerning people believe in

them and transfer accounts to them. Nowadays,

criminals usually make use of smart phones and new

transmission technology to make crime. For example,

they may absorb public saving illegally under cover

of new industry. All the above new-form crime and

risks bring much pressure to police departments.

(Jiao, 2007).

3 CLUSTER ANALYSIS OF 110

ALERT QUANTITIES

Cluster Analysis is mathematical statistics that clas-

sifies objects based on numeric characteristics, which

is used most often in practical work procedure, being

able to classify either samples or variables. With the

development of computers, Internet Technology and

Big Data, interpersonal exchange has become more

and more frequent, and people’s need to analyze,

manage and utilize Big Data has become more and

more urgent, while Cluster Analysis plays a more and

more significant role during the course of Data

Mining, as traditional statistics can not satisfy present

researchers’ need. Under the background of Social

Informatization and Big Data, researchers of any field

need to collect, analyze and utilize data, which has

become more and more difficult. Cluster Analysis is a

ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology

374

significant means for researchers to deal with these

problems. (Annikaer, 2019)

Cluster Statistics is statistics that can reflect na-

ture distance of samples or variables, including Dis-

tance and Similarity Coefficient. Distance can be

used to classify samples, including Absolute Value

Distance, Euclidean Distance, Minkowski Distance,

Mahalanobis Distance, etc. Euclidean Distance is

used most often in practical work procedure. Simi-

larity Coefficient is usually used to classify variables,

including Cosine and Correlation Coefficient, etc.

Pearson Correlation Coefficient can be used to clas-

sify successive materials.

The Hierarchical Clustering Method is easy, ac-

curate and efficient, which is a branch of Cluster

Analysis, and universally applied in practical cases.

Biology researchers have used the Hierarchical

Clustering Method to classify animals, plants and

genes; traffic experts have used the Hierarchical

Clustering Method to classify factors that affect

traffic accidents; economic researchers have used the

Hierarchical Clustering Method to classify economic

indicators of different regions to acquire economic

information and offer suggestions. (Qin, 2017).

In order to decrease the fluctuation range of 110

Alert Quantities of each period, we divide one day

into 12 stages, including 0:01-2: 00, 2:01-4: 00,

4:01-6: 00, 6:01-8: 00, 8:01-10: 00, 10:01-12: 00,

12:01-14: 00, 14:01-16: 00, 16:01-18: 00, 18:01-20:

00, 2:01-22: 00, 22:01-24: 00. Using SPSS, we can

acquire the clustering result as follows.

From the above figure, we can see that the period

0:01-6:00 can bee regarded as one category, in which

110 Alert Quantity is relatively smaller; periods

including 8:01-12:00 and 14:01-18:00 can bee re-

garded as one category. The Diagram Based on Hi-

erarchical Clustering Method is like a tree, from

which we can judge which two periods are closest in

distance. K means is usually utilized to classify

big-sample data quickly, in which we can set initial

cluster centers artificially, making full use of previ-

ous research achievement, and saving much time.

However, this means has some limitation. It can only

be used to cluster samples instead of variables; var-

iables we use should be continuous. It is supposed

that there are n variables to be clustered into K cat-

egories, which make up a n–dimensional space, and

each sample is one point in the space. First, choose K

points as initial cluster centers; second, make other

sample points clustered to category centers based on

the principle of minimum Euclidean distance and we

will acquire a scheme of initial clusters, means of

Figure 1: Diagram Based on Hierarchical Clustering Method.

Figure 2: Clustering Members of 2 Categories.

Figure 3: Clustering Members of 3 Categories.

Research on Urban Security Management Based on Cluster Analysis of 110 Alert Quantities

375

each initial cluster having been calculated; at last,

re-cluster samples based on calculated means until

reaching the standard of converging. Using SPSS, we

can acquire the following result of two clusters and

three clusters. In the following figure, 1 stands for the

period of 0:01-2:00, and 2 stands for the period of

2:01-4:00, and so on.

From the above figure, we can see that if samples

are divided into two categories, the higher period

includes 08:01-20:00, while the lower period in-

cludes 0:01-08:00 and 20:01-24:00.

From Figure 3, we can see that if samples are di-

vided into three categories, the higher period includes

08:01-10:00 and 18:01-20:00, the lower period in-

cludes 0:01-8:00 and 20:01-24:00, and the ordinary

period includes 10:01-16:00.

4 STRATEGIES OF URBAN

SECURITY MANAGEMENT

BASED ON CLUSTER

ANALYSIS OF 110 ALERT

QUANTITIES

First, it is required to allocate police power appro-

priately. An instant monitoring and judgment system

needs to be established in order to estimate and al-

locate police power appropriately, as police power is

to be allocated according to public security situation

of different districts. Through Cluster Analysis, we

have mastered the objective fluctuation law of 110

Alert Quantities, according to which the command

center can allocate police power dynamically. In the

period from 0:01-6:00, 110 Alert Quantity is rela-

tively smaller, and the corresponding police power

can be set relatively smaller. In periods from 08:01 to

10:00 and from 18:01 to 20:00, 110 Alert Quantity is

relatively larger, and the corresponding police power

can be set relatively larger. Police personnel de-

partments need to master data and keep track of

categories, ages, educational backgrounds of po-

licemen. Police power exchange is a significant

supplement of police power allocation, the purpose of

which is to encourage fluidity of police power, and

make time distribution, personnel structure and work

quality more balanced and coordinated. Allocation of

police power should also be based on policemen’s

individual career development in order to realize

balanced and effective distribution. Local Police

Stations do not necessarily need policemen with

high-level knowledge, technique and specialty. Elder

policemen, middle-age policemen and young po-

licemen need to cooperate closely and collocate

reasonably. When a policeman has worked in the

same Local Police Station for 10 years, he will be

asked to work in another position. However, the

longer time a community policeman works, the bet-

ter, since he has known well almost all residents and

surroundings. When policemen retire, young po-

licemen must be supplemented in order to keep the

balance of police power. (Zhao, 2011)

Second, it is required to improve patrolling pow-

er. The most effective way is to improve police visi-

bility in order to prevent city violent crime. Compo-

sition of police power is like a pyramid, which means

that if more police power is distributed to streets,

control level of city security will be enhanced. Young

policemen are dynamic, enthusiastic, and if the

command center sends them to streets, patrolling

power will be greatly improved. They can also ac-

quire rich practical experience, which is beneficial to

their career development. A policeman without

grass-root work experience surely will not be fit for

any position in police department. Therefore, fresh

policemen should be strictly demanded to work for

one or two years in Local Police Stations. On one

hand, grass-root police stations have more police

power to distribute, on the other hand, young po-

licemen acquire practical experience, which is bene-

ficial to both sides. It must be pointed out that some

non-police alert takes up much patrolling power.

Some alert needs to be handled by other government

sectors. However, they call the police. Usually pa-

trolling police force go to the spot and investigate

facts. The whole procedure cost much time and en-

ergy of patrolling police power. An effective way is

to shunt these non-police alert tasks to other gov-

ernment sectors. Police departments can not play the

role of other government sectors. If an alert should

not be handled by the police, 110 command center is

responsible for telling the person calling the police to

turn to other government sectors for help instead of

calling the police. Folk public security power is also

to be combined with official police power in order to

set up a city security net and avoid shortage of pa-

trolling police power, which includes social volun-

teers, security guards of schools, enterprises, facto-

ries and communities. (Liu, 2015)

With the help of

modern internet and telecommunication technology,

folk public security power can be easily integrated

into the social public security net. When one alert

happens, the neighboring folk public security power

will actively respond and reach the spot instantly.

Third, it is necessary to enhance intensity of

community police. In a traditional city management

mode, police departments are major suppliers of

public security, while other folk public security

ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology

376

power is limited. The traditional management mode

focuses on dealing with city violent crime instead of

prevent and avoid crime, which has brought more and

more challenges to police departments, and can not

assure satisfying management effects. Therefore,

police departments need to realize that the traditional

management mode is expensive and not fit for the

present social structure under the background of a

transitional city mode. Transition from combat of

crime to prevention of crime is the basic trend and

fundamental strategy. Police departments need to

switch roles and lay more emphasis on preventing

crimes based on community police. The community

is a tie of social development and public security,

based on which police departments take actions to

prevent city crime. Social power is a significant

resource to absorb more residents to take part in

resolving conflicts and preventing violent crimes. (Ji,

2020)

Community policemen know well people and

environment in this district. When strangers appear,

they may realize. Once illegal cases take place in his

community, he can find it out at once and report to

the command center. Cooperating with folk public

security power, and making an intense social securi-

ty net, community police can play a more significant

role, and level of a city’s public security manage-

ment will be improved.

5 CONCLUSION

We have analyzed functions of 110 Command Center

in urban public security management, made a Cluster

Analysis of 110 Alert Quantities based on the Hier-

archical Clustering Method and K-means, and raised

reasonable advice. From the research, we have dis-

covered that there exists different 110 Alert Quanti-

ties in different periods of one whole day, based on

which police departments can scientifically adjust

police power. The research helps policemen on duty

form a scientific expectation of 110 Alert Quantities

and make enough preparation for 110 Alert. Limited

by samples we collect and the city we choose, the

research result may be inappropriate for other cities.

If we want to find out national rules of 110 Alert

Quantities, we need to collect and make a cluster

analysis of other cities. That is, rules we have found

out in this city may not apply to other cities. In the

future, research on this field should focus on factors

that affect 110 Alert Quantities of a city and make

more precise prediction of 110 Alert Quantities based

on weather factors. Change of 110 Alert Quantities is

instant and dynamic. In future research, we may

make full use of advanced algorithm and visualiza-

tion to help 110 Command Center make scientific

determination and adjust police power reasonably

and efficiently.

ACKNOWLEDGEMENT

This article is a phased achievement of “Coping

Strategies of Cyber Public Sentiment Crisis Based on

SWOT”, the research project of Shandong Police

College (Project Number: YSKYB202103).

Cui Shiwen (1980-), a vice professor of Shandong

Police College, from Jiaozhou of Shandong Province,

majors in Cyber Public Sentiment Crisis Manage-

ment, Police English Teaching Reform and 110 Alert

Amount.

REFERENCES

Annikaer. aisikaer, Zulaikezi.mijiti. Systematic Clustering

Method and Its Application [J]. Engineering of Values.

2019, 38(17):254-258.

Jiao Daotong. Construction of Urban Crime Prevention

System [J]. Journal of Jilin Public Security College.

2007(02):57-61.

Ji Naili, Xu Xiao. Management of Urban Public Security

Based on the Community [J]. Journal of Theo-

ry.2020(02):43-52.

Li Haitao. Operating Mechanism and Countermeasures of

110 Alert [J]. Law and Economy, 2013(2):119-121.

Liu xia. Shortage and Consummation of Police Patrolling

under the Severe Counter-terrorism Situation [J].

Journal of Hubei Police College, 2015(9):53-56.

Qin Ming, Ning Jianbiao, Deng Mingjun. Appliance of the

Hierarchical Clustering Method in Analysis of Drivers’

Illegal Behavior [J]. Highway and Tranportation, 2017

(04): 45-51.

Xu Zaihua. Optimization and Path Design of 110 Com-

manding System [J]. Journal of Shanxi Police College,

2018 (01):76-79.

Yang yang, Zheng Hanjun. New Challenges of Local

Police Stations and Coping Strategies of 110 Alert [J].

Journal of Law, 2019(20):126-127.

Zhao Weixiong. To Develop Scientifically and Allocate

Police Power Reasonably [J]. Journal of Shanghai

Public Security Vocational College. 2011, 21(05):

18-22.

Research on Urban Security Management Based on Cluster Analysis of 110 Alert Quantities

377