Research on the Realization of Water-saving Social Service

Application based on Knowledge Cloud Service

Wei Ma

1,*

, Yonggang Liu

, Ke Zhang

, Kang Jing

, Tianfan Wang

and Jiancang Xie

State Key Laboratory of Eco-hydraulics in Northwest Arid Region, Xi’an University of Technology, Xi’an, Shaanxi

710048, China

Shaanxi Provincial Water Resources and River bank Dispatch Management Center, Xi’an, Shaanxi 710004, China

Keywords: Cloud service, Water saving technology, Knowledge service, Water resources management, Information

technology, Knowledge representation

Abstract: At present, the water problem is prominent, and the whole society needs to adhere to long-term water-saving

behavior. This paper combines three knowledge service methods: keyword retrieval, knowledge

recommendation and knowledge customization to improve the quality of knowledge service; The four

knowledge recommendation algorithms of collaborative filtering recommendation, demographics based

recommendation, content-based recommendation and rule-based recommendation are realized by using java

language, which improves the accuracy of knowledge recommendation; At the cloud server, the knowledge

is evaluated and scored according to a certain algorithm according to the statistics of users' behavior data such

as knowledge clicks, so as to ensure the quality of water-saving knowledge. Take cloud service as the

intermediary condition to realize the combination of website knowledge extraction and knowledge base; At

the same time, update and supplement the knowledge in the knowledge base according to the evaluation score

of the cloud server on the provided knowledge. The water-saving service website has been developed to

complete the construction of the website through micro service technology and realize different functional

modules of the website, which plays a certain role in promoting the people's participation in water-saving

awareness, water-saving publicity and the construction of water-saving society.

1 INTRODUCTION

Based on the current shortage of water resources in

China, since 2016, various parts of China have

successively issued policy documents based on water

security, water pollution and water shortage,

emphasizing the establishment of a water resources

management mechanism centered on water rights and

water market, so that the market can restrict and

regulate water-saving behavior, give full play to the

role of economic means and improve water

efficiency. At the same time, the state has also

established several water resources management

systems, such as the water pricing system and the

sewage discharge permit system, and the water saving

system has been gradually improved. The National

Water-saving Action Plan issued by The State

Council in 2019 clearly points out that to solve the

shortage of water resources in China, the first step is

to adhere to the water-saving priority policy,

gradually improve the social awareness of water-

saving, improve the water-saving system, and

accelerate the innovation of water-saving technology

(Kang, 2019). Water saving is related to the

production and life of the whole society, involving

many industries and sectors. It is not enough to rely

on the government's unilateral efforts and appeals.

Only by water-saving behaviors of the whole society

we can accelerate the solution of the shortage of water

resources supply and demand (Sun et al., 2018). In

order to achieve better water-saving effect and

improve the water-saving behavior of the whole

society, it is necessary to combine with information

technology and build a social water-saving service

website (Qiao et al., 2017). Although there were

many water-saving websites in the past, most of them

contained major events and news at home and abroad,

with a long updating cycle and a large length of text,

without relevant pictures and videos. The functional

module of the website water saving forum is also

empty and no one cares. Figure 1 shows the main

problems of traditional water-saving websites.

Ma, W., Liu, Y., Zhang, K., Jing, K., Wang, T. and Xie, J.

Research on the Realization of Water-saving Social Service Application based on Knowledge Cloud Service.

In Proceedings of the 7th International Conference on Water Resource and Environment (WRE 2021), pages 397-413

ISBN: 978-989-758-560-9; ISSN: 1755-1315

397

Figure 1: Problems of traditional water-saving websites.

In terms of website construction technology, this

paper uses springboot, springcloud and docker to

realize the development of water-saving socialized

website by micro service; In terms of website content,

user objects are divided, and targeted knowledge

services are provided according to different user

objects; In terms of service form, it is mainly based

on cloud services to sort and recommend knowledge,

so as to realize the water-saving socialized service of

the website. Through the above research, the

problems existing in the past traditional websites have

been solved, the knowledge quality is higher, the

service is more targeted and humanized, the publicity

role of traditional websites has been strengthened,

and the social consciousness of the whole people to

participate in water saving has been improved.

2 KNOWLEDGE CLOUD

SERVICE FRAMEWORK

In essence, knowledge cloud service is to provide

socialized services in the form of knowledge services.

In the past, socialized services usually refer to

agricultural socialized services. This socialized

service is public welfare, mainly to alleviate and

eliminate the asymmetry of information technology

in production and promote the circulation of

resources (Gu et al., 2019). The socialized services

studied in this paper are mainly in water saving, that

is, to provide socialized services on water saving for

the whole society and alleviate the information

asymmetry in water saving in production and life.

2.1 Composition of Knowledge Services

Knowledge service refers to a series of links such as

extracting, refining, integrating and innovating the

explicit and tacit knowledge owned by individuals or

social organizations based on the knowledge needs of

objects and the analysis of the objective environment,

and finally providing users with highly targeted

services for solving problems (Yang & Yang, 2021).

The beneficiary groups of knowledge services are

diverse, with a large number of users and levels;

However, the water-saving socialized service studied

in this paper makes the beneficiary groups have

obvious user characteristics and can provide targeted

knowledge services according to user behavior. This

knowledge service system mainly includes the

following modules:

2.1.1 Knowledge Retrieval

Knowledge retrieval is to provide targeted knowledge

services for users according to their needs, mainly

including rules, patterns, and other related

knowledge. Knowledge retrieval methods usually

include two kinds: keyword retrieval and classified

navigation (Guo et al., 2021).

2.1.2 Personalized Sorting

Personalized sorting is to sort the relevance of search

results according to the specific needs of users, and

quickly locate the target information by combining

engine algorithm, mathematical modeling and other

tools (Liu et al., 2021). This sorting method avoids

the uniformity of retrieval results, makes users

directly locate the target information in a large

number of retrieval information, and provides users

with the most targeted personalized knowledge

services. Among them, the extraction of user

personalized features mainly includes explicit and

implicit extraction methods. Explicit extraction refers

to the personal information that users manually

submit at the front end of the knowledge service

system; Implicit extraction mainly refers to the

background modeling according to the user's

browsing behavior and retrieval purpose, and

transmitting it to the database.

2.1.3 Knowledge Push

Knowledge push is also an important part of

personalized information service. Because the

information in the knowledge base is constantly

updated, users may not know the expansion of new

knowledge they need in time. Therefore, the

WRE 2021 - The International Conference on Water Resource and Environment

398

knowledge service system must include a knowledge

push mechanism to enable users to obtain the new

knowledge they need in time. The main push methods

include: knowledge push of new resource documents

and knowledge push associated with new knowledge.

2.1.4 Online Recommendation

Online recommendation is to recommend relevant

knowledge resources for users according to their

browsing behavior and retrieval objectives. This

method is to recommend relevant information for

real-time in the user's knowledge service system, and

return the result of the recommendation information

to the user. The intelligent analysis algorithm is

mainly used to predict and recommend the user's

upcoming retrieval goals and browsing behavior by

integrating the user's browsing history and retrieval

target results.

2.1.5 Knowledge Evaluation

Knowledge evaluation is a process of feedback,

summary, correction and supplement of existing

knowledge. It is mainly combined with modern

information analysis technology to mine potential

information in relevant fields, so as to promote

scientific knowledge guidance and evaluation, and

then promote the improvement of knowledge quality

and knowledge service level (Sun & Liao, 2019). The

knowledge framework is shown in Figure 2.

Figure 2: Composition of knowledge service.

Research on the Realization of Water-saving Social Service Application based on Knowledge Cloud Service

399

2.2 Construction Ideas of Knowledge

Cloud Service

Knowledge cloud service mainly provides users with

water-saving knowledge services, so it mainly

includes two aspects: knowledge recommendation

and knowledge evaluation. The combination of these

two processes can make users satisfied with the

knowledge services provided and complete the

updating and improvement of the knowledge base

(Zheng et al., 2021). With cloud service as the

intermediary, the knowledge base completes the

knowledge provision to users and the update iteration

of the knowledge base, which is of far-reaching

significance to promote the improvement of social

water-saving knowledge level and the awakening and

strengthening of water-saving consciousness, and the

water-saving effect is significantly improved. The

specific cloud service system diagram is shown in

Figure 3:

knowledge base

Knowledge

information

Knowledge of

quality

Knowledge service

application

Website

The phone

app

cloud services

Knowledge

recommend-

ation

Knowledge

evaluation

Information knowledge

Knowledge extraction

Knowledge update

User evaluationand other data

Figure 3: Diagram of knowledge cloud service system.

2.2.1 Mode of Knowledge Service

The effect of knowledge service depends on the mode

of knowledge service. There are three common user-

oriented knowledge service modes: passive service

mode, active service mode and mixed service mode.

(1) Passive service mode

Passive service, a knowledge service mode,

mainly completes knowledge service based on user

retrieval behavior. When users do not know their

knowledge needs, this service method can not achieve

accurate or fuzzy query through keyword retrieval to

get their desired results. Therefore, in the construction

of knowledge cloud service, we should not only

match the keywords based on the user's retrieval

information, but also use the recommendation

algorithm to analyze the keywords, so that the

recommended water-saving knowledge is close to the

user's knowledge needs, and the results are more

comprehensive. At the same time, the results are also

instructive for user screening.

(2) Active service

The active service method is to timely push the

user's personalized knowledge by obtaining the user's

browsing behavior, characteristics, hobbies, age,

occupation and other information, combined with the

knowledge recommendation algorithm. This service

method does not complete knowledge push through

the user's search keywords. By collecting data on

users' interest in water-saving subjects and browsing

information of water-saving knowledge, we can

achieve real-time, push users' needs and interest in

water-saving knowledge by means of e-mail and

official account subscription.

(3) Mixed service

When users cannot retrieve relevant knowledge or

have special needs for knowledge, knowledge cloud

service can quickly customize according to users'

knowledge needs, return the results to users, and store

the customized knowledge in the knowledge base for

subsequent use by other users. The comprehensive

integration platform and visual knowledge map are

adopted to quickly respond to the new knowledge

needs of users, quickly draw the knowledge map and

push the knowledge in time. This method combines

knowledge retrieval, knowledge map customization

and knowledge push to form a hybrid service mode,

enriching the traditional knowledge service mode and

improving the quality and effect of knowledge service.

2.2.2 Knowledge Recommendation

Knowledge recommendation is to complete

knowledge recommendation services for users,

WRE 2021 - The International Conference on Water Resource and Environment

400

according to users' characteristics, interests, age,

occupation and other information, combined with

models and algorithms. Knowledge recommendation

is mainly used in knowledge push, personalized

knowledge ranking and online push.

(1) Knowledge recommendation system model

Recommendation system modeling mainly

includes three parts: system modeling, project

modeling and recommendation algorithm. User

modeling is to express the user's browsing

information, evaluation information, purchase

records and other information in data, and store the

data in the user's preference model, so that the

recommendation system can master the user's

characteristics, interests and preferences. Project

modeling is mainly to quantify and express the

information of all user related projects, and form a

project data system. Recommendation algorithm is a

recommendation model that recommends items to

users. The specific process of knowledge

recommendation system is shown in Figure 4:

Figure 4: Recommended system flow chart.

At present, the commonly used recommendation

algorithms include rule-based, demographic, content-

based recommendation methods and collaborative

filtering recommendation methods. Among the four

methods, collaborative filtering recommendation

algorithm is the most widely used. In the process of

use, it does not need knowledge in related fields, and

has strong pertinence. Therefore, the collaborative

filtering recommendation algorithm is mainly

introduced:

Collaborative filtering mainly obtains the

similarity between users or items according to the

user's love for items, predicts the user's love for

untouched items based on these similarity, and

obtains a score. Complete personalized

recommendation to users according to different

scores. At the same time, collaborative filtering

algorithms can be divided into two categories: user

based collaborative filtering recommendation

algorithm and item based collaborative filtering

recommendation algorithm. The implementation

steps of the two algorithms are similar. This paper

mainly introduces the user-oriented collaborative

filtering recommendation algorithm.

(2) Steps of user based collaborative filtering

recommendation algorithm

a. Information collection

The user's preference for the viewed items is

expressed by score, and then put into an M × Users of

N - in the item scoring matrix, M represents the

number of users and N represents the number of

items. Through matrix representation, the collected

information is processed to realize the management

of user information and determine user preferences.

b. Nearest neighbour search

Nearest neighbour search mainly determines the

similarity between target users and calculates the

similarity value of each user in the target group. The

calculated similarity value can provide decision

support for recommended content. Common

similarity calculation methods include: cosine

similarity, adjusted cosine similarity, Pearson

similarity and Jaccard similarity. These four methods

are implemented in the constructed knowledge cloud

service system, and the more accurate and simple

similarity algorithm will also be incorporated into the

knowledge recommendation system.

Calculation formula of cosine similarity:

Sim



i,j



∑











∑











∑









(1)

Where:

𝑥



, 𝑥



is user’s ( 𝑥



、 𝑥



)score of the k-th

knowledge;

Research on the Realization of Water-saving Social Service Application based on Knowledge Cloud Service

401

K represents the sequence number of knowledge,

k = 1,2,3,..., m;

The value of cosine similarity algorithm is [0,1].

The closer the value is to 1, the higher the similarity

is. The closer it is to 0, the lower the similarity is.

Cosine similarity algorithm is suitable for qualitative

judgment, which judges the similarity of abstract

items such as knowledge and interest. Figure 5 shows

the Java programming implementation of cosine

similarity algorithm.

Figure 5: Implementation of cosine similarity algorithm

Adjust the cosine similarity calculation formula:

sim



i,j



∑

R

,

−R





R

,

−R







∈



∑

R

,

−R



∈







∑

R

,

−R



∈





(2)

Where:

𝑅

,

refers to the i user's rating of knowledge;

𝑅





refers to the average score of all knowledge

evaluated by user i;

Adjusting the cosine similarity recommendation

algorithm can effectively solve the problem of

different users' preference for items, that is, the

scoring of items is different. The problem of scoring

preference can be effectively solved by subtracting

the average of all item scores from the user's score for

each item. Figure 6 shows the Java programming

implementation of adjusting cosine similarity

algorithm.

Pearson similarity calculation formula:

sim



X,Y





X,Y







E



X−





Y−















−E















−E







(3)

Where:

𝑐𝑜𝑣



𝑋,𝑌



refers to the covariance of X, Y;





refers to the standard deviation product of

index X and Y;

Pearson similarity algorithm is mainly used to

calculate the linear correlation between the two

variables. Its role in knowledge recommendation is

mainly used to calculate the sum of squares of the two

users' scores when both users have evaluated the same

knowledge, and calculate the product sum of each

score. The value range of Pearson similarity

algorithm is [0,1]. The closer the value is to 1, the

higher the similarity between the two users. The

closer the value is to 0, the lower the similarity

between the two users. The result set obtained by

Pearson algorithm is larger and the recommendation

range is wider. Therefore, compared with Jaccard

similarity algorithm, the accuracy is lower, and it is

WRE 2021 - The International Conference on Water Resource and Environment

402

suitable for scenes with low requirements for the

accuracy of recommendation knowledge. Figure 7

shows the Java programming implementation of

Pearson similarity algorithm.

Figure 6: Implementation of adjusted cosine similarity algorithm.

Figure 7: Implementation of Pearson similarity algorithm.

Jaccard similarity calculation formula:

Sim



X,Y



X∩Y

∪

(4)

Jaccard similarity calculation is to find the ratio

between the number of intersection elements and the

number of union elements in two sets X and Y. the

Research on the Realization of Water-saving Social Service Application based on Knowledge Cloud Service

403

value range is [0,1]. The closer the value is to 1, the

stronger the commonality of the two sets is, and the

closer it is to 0, the weaker the commonality is. The

Jaccard similarity coefficient can quantitatively

describe the common characteristics and interests of

two users. In the case of high similarity, it can

recommend the knowledge that the other user is

interested in but not exposed to two users at the same

time. This method is applicable when Boolean value

is used to measure the similarity between individuals,

and can not give the size of difference value. In the

process of knowledge recommendation, the similarity

can be calculated according to the user's interest to

obtain whether the user is interested in a knowledge

topic, so as to recommend according to the correlation

points between interests and related knowledge.

Using the Jaccard similarity calculation method, the

result set is small and accurate, which is suitable for

scenes with high requirements for knowledge

recommendation accuracy. Figure 8 shows the Java

programming implementation of the Jaccard

similarity algorithm.

Figure 8: Implementation of Jaccard similarity algorithm.

Figure 9: Implementation of prediction scoring algorithm and adjusted prediction scoring algorithm.

WRE 2021 - The International Conference on Water Resource and Environment

404

(3) Result generation

The set of users with high similarity can be

obtained through nearest neighbor search. According

to the user's score on the knowledge in the user set,

the knowledge not scored by the target user is

predicted and scored, the prediction scores are

arranged from high to low, and the first n prediction

scores are selected to generate recommendation

results and complete the knowledge recommendation

process.

Calculation formula of prediction score:

|

∑

sim





×R

∈

∑|

sim



|

∈

(5)

|





∑

sim









−R







∈

∑|

sim



|



∈



(6)

Where:

𝑃

|

refers to the prediction score of user u on the

ith untouched knowledge;

𝑠𝑖𝑚



𝑢,𝑣



refers to the similarity calculation

results of user u and user v;

𝑅



refers to the user v's rating of the ith

knowledge;

𝑅





refers to the average score of user v on the

knowledge they have browsed;

𝑅





refers to the average score of user u on the

knowledge they have browsed

Figure 9 shows the Java implementation of the

user's prediction score calculation of an unpriced

knowledge.

2.2.3 Knowledge Evaluation

Water saving knowledge evaluation is based on users’

behavior such as browsing information, downloading

data and scoring data, through the algorithms, to

evaluate knowledge, and process the knowledge

according to the evaluation results. The water-saving

knowledge evaluation process is visible, and users

can participate in the knowledge evaluation process.

The knowledge evaluation standard is generated

based on the user's behavior. The specific knowledge

evaluation is as follows:

(1) Knowledge clicks

1 point will be added every time knowledge is

browsed. The higher the score, the better the quality

of knowledge. The cloud service statistics module is

displayed according to the score from high to bottom.

Note that for the same user, browsing multiple times

on the computer can only be counted as one time.

(2) Number of knowledge Downloads

Add 1 point every time the knowledge is

downloaded. The higher the score, the better. The

cloud service statistics module displays the

knowledge according to the score from high to

bottom. Note that for the same user, downloading

multiple times on the computer can only be counted

as one time.

(3) Label selection times

This standard is to count the times of label

selection, and add 1 point for each label used. The

cloud service statistics module is displayed according

to the score from high to bottom. Note that for the

same user, multiple selections on the computer can

only be counted as one.

(4) Number of knowledge references

Add 1 point every time the knowledge is quoted.

The higher the score, the better. The cloud service

statistics module is displayed according to the score

from high to bottom.

(5) Knowledge recommendation times

Add 1 point for each recommended knowledge.

The higher the score, the better. The cloud service

statistics module is displayed according to the score

from high to bottom.

(6) Knowledge scoring times

Every time knowledge is scored, 1 point will be

added. The higher the score, the better. The cloud

service statistics module is displayed according to the

score from high to bottom.

(7) Number of knowledge reviews

Add 1 point every time knowledge is commented.

The higher the score, the better. The cloud service

statistics module is displayed according to the score

from high to bottom. Note that multiple comments by

the same user on the computer can only be counted as

one.

Cloud services can sort the knowledge in the

knowledge base from high score to low score

according to different indicators by making statistics

on users' evaluation data. The sorting results can be

seen by users and managers, and the knowledge with

high score can be retained in the knowledge base; For

the knowledge with low score, re-enter the

knowledge representation process, save the updated

knowledge in the knowledge base, and recommend

the knowledge again.

In short, the water-saving knowledge cloud

service provides scientific knowledge

recommendation service and knowledge evaluation

service for the realization form of water-saving

knowledge. The specific implementation process of

knowledge cloud service is reflected in the process of

meeting users' knowledge needs through the water-

saving website.

Research on the Realization of Water-saving Social Service Application based on Knowledge Cloud Service

405

3 ESTABLISHMENT OF

WATER-SAVING SOCIALIZED

SERVICE WEBSITE BASED ON

KNOWLEDGE CLOUD

SERVICE

Based on the construction of knowledge cloud service,

combined with information technology, a water-

saving socialization website is developed on the

Internet. According to different service objects and

audience groups, different knowledge service

methods are adopted to carry out water-saving

knowledge service.

Due to the large number of audience groups, the

user groups are classified and subdivided according

to the characteristics of age, occupation and education

level, and the users are divided into groups with some

common or similar characteristics and individuals

with basically different characteristics. Combined

with information technology such as web and

integrated platform, carry out group oriented water-

saving knowledge service, that is, water-saving

socialized service.

3.1 Construction Ideas of Knowledge

Cloud Service

The water-saving socialized service website created

in this paper is not to overthrow the previous

traditional water-saving websites, but to give better

play to the utility of water-saving websites and solve

the problem of existing websites (Wang, 2016). First,

the starting point of building water-saving websites is

the same. The development mode of building water-

saving socialized service websites is basically similar,

and the development mode of door type websites is

adopted; Second, in order to make the service more

targeted, it is necessary to divide the user object

image; Third, we should improve the way of

knowledge service (Wang & Jeong, 2018).

3.1.1 Realize Cloud Service and Provide

Website Content

The content displayed by the traditional water-saving

website is messy and illogical. On the one hand, the

developer of the website is computer related

personnel, on the other hand, the operator of the

website is management personnel. The personnel of

these two organizations are not professional relative

to the water-saving content. The water-saving service

website proposed in this paper is built on the basis of

knowledge cloud service. The water-saving content

of the website does not need developers and managers

to provide. In a certain sense, it realizes the

professionalism, and pertinence of water-saving

content.

3.1.2 Realize Personalized Knowledge

Service

Traditional water-saving websites do not distinguish

user groups. The website content is large and

extensive, and all the content is presented to users at

one time. Users can only find what they need from a

large number of messy content, which is not simple.

Therefore, the water-saving socialization website

should first divide the user groups according to the

relevant characteristics, provide the knowledge

content that users are interested in, realize

personalized knowledge service and improve the

quality of water-saving service.

3.2 Technical Support of Water-saving

Socialized Service Website

The operation of water-saving socialized website

needs continuous maintenance. With the continuous

updating of water-saving knowledge, the service form

of the website needs to be deepened and improved,

and the module content of the website needs to be

updated in time (Zhou et al., 2007). Microservice

technology is used to build the website. When the

website module is upgraded, some microservices

remain running and the other part stops running. After

the upgrade, the two parts are exchanged to complete

the upgrade of the website system and improve the

sustainable operation ability of the website. Spring

boot, Spring cloud and Docker technologies are

required to complete this exchange operation.

Section, subsection and sub subsection first

paragraph should not have the first line indent, other

paragraphs should have a first line indent of 0,5-

centimeter.

3.2.1 Spring Boot

Spring boot framework can be used to develop a

single micro service. Spring boot can realize

automatic configuration, simplify users' work on

project configuration, and improve the efficiency of

project construction. Moreover, Spring boot

integrates embedded web container, which can

quickly complete the construction of applications.

During deployment, the web container can also run

normally to complete the hot deployment.

WRE 2021 - The International Conference on Water Resource and Environment

406

3.2.2 Spring Cloud

Spring cloud is a tool set for rapidly building a

microservice system based on spring boot, and has

lightweight microservice components with perfect

functions. The microservice cluster built by spring

cloud provides a guarantee for the smooth operation

of microservices. Spring boot builds a single

microservice; Spring cloud organizes these micro

services to form a micro service group, and then

constitutes the whole business system.

3.2.3 Docker

Docker is a lightweight virtual engine. Compared

with previous virtual machines, docker has a very low

share of hardware resources and improves the

utilization of system resources. You can deploy the

image files generated by the running environment and

application software into the docker container,

instead of completing the environment configuration

of the application software every time, so as to

improve the efficiency of operation and maintenance.

3.3 Functional Modules of

Water-Saving Socialized Service

Website

The main functional modules of the water-saving

socialization website constructed in this paper are

divided into five parts: industry news, laws and

regulations, water-saving science popularization,

water situation map and water-saving forum. Water

saving knowledge service is the most important

module of website construction, so the water saving

knowledge service module is listed separately and

detailed separately. The main function modules of the

website are shown in Figure 10:

Figure 10: functional module diagram of water saving socialized service website.

3.4 Construction of Water-Saving

Social Service Website

3.4.1 Home Page

The home page of the website includes the entrance

of each module, mainly including the brief

information of other modules and relevant water

resources news information. By clicking on different

functional modules, you can view industry news,

popular science knowledge, laws and regulations,

water regime map, water-saving services, water and

health and other knowledge information by category,

as shown in Figure 11:

Figure 11: Homepage of water saving socialized service website.

Research on the Realization of Water-saving Social Service Application based on Knowledge Cloud Service

407

Figure 12: Main interface of industry news.

3.4.2 Industry News

The industry news module includes three parts,

including national policies, detailed reading of news

and water-saving highlights. It is mainly about the

interpretation of the national policies and policies on

water resources management, the interpretations of

the latest local water use news, and the report of

relevant water-saving results. Figure 12 is the cross-

sectional view of the industry news module.

3.4.3 Laws and Regulations

This module mainly introduces national policies, laws

and regulations, mechanisms and systems, as well as

the interpretation of relevant policies and regulations.

It can be divided into five parts: the latest

introduction, water-saving laws, water-saving

regulations, industry quota and policy interpretation.

The public and technicians can click the

corresponding interface to access the knowledge they

are interested in, which is of great significance to

domestic water saving and economic production

water saving. Figure 13 is a cross-sectional view of

laws and regulations.

Figure 13: Main interface of laws and regulations.

WRE 2021 - The International Conference on Water Resource and Environment

408

3.4.4 Water Saving Science Popularization

This module mainly serves two objects: one is the

popularization of domestic water-saving skills and

water-saving technologies and tools for the public;

The second is the popularization of relevant

knowledge about industrial water saving and

agricultural water saving for relevant personnel of

water-saving technology. Figure 14 shows the

interface of water saving Science Popularization.

Figure 14: Main interface of water saving science popularization.

3.4.5 Hydrological Map

This module mainly establishes a two-dimensional

plane map, and intuitively reflects the sewage source

information, watershed water regime information,

quoted water source information and other relevant

water regime information in the map. By obtaining

the location of the user, the user can know the water

environment of the area. Figure 15 shows the

interface of water regime map:

Figure 15: Main interface of water regime map.

Research on the Realization of Water-saving Social Service Application based on Knowledge Cloud Service

409

3.4.6 Water Saving Forum

This module mainly provides the public with a

platform for water-saving exchange. In the water-

saving forum, the public can exchange water-saving

experience, report polluted rivers, and answer

questions from experts. Specific discussion topics

include: domestic water-saving exchange, water

environment supervision, expert Q & A and water-

saving technology exchange. Figure 16 shows the

interface of water saving Forum:

Figure 16: Interface of water saving Forum.

3.5 Service Process of Water Saving

Socialized Service Website

3.5.1 Website Service Process Design

The application of cloud services realized through

websites mainly completes the knowledge service

process through keyword retrieval (Deng, 2015).

Users send knowledge requests to the cloud service

through keywords, and the cloud service recommends

water-saving knowledge to users according to

keyword matching and related algorithms; The web

page feeds back the user's browsing history,

download times, knowledge map evaluation scoring

and other data to the cloud service(Du et al., 2021);

Cloud service evaluates the knowledge content

according to the knowledge evaluation algorithm, and

the high-quality knowledge content will be pushed

again to complete the push process; Low quality

knowledge needs to be improved and then enter the

next round of knowledge recommendation process to

complete knowledge push (Zhang et al., 2017). The

specific flow chart is shown in Figure 17.

3.5.2 Implementation of Website Service

Process

Social service websites not only passively provide

users with required knowledge services, but also

actively complete knowledge services for users. Push

water-saving knowledge to users through

subscription, knowledge space and personalized

push, constantly attract users' attention, stimulate

users' interest, and make water-saving knowledge

service more convenient and humanized (Zhang,

2011). In order to make the provision of water-saving

knowledge more targeted, user groups need to be

divided. According to the role played by user groups

in water conservation, water conservation knowledge

services are divided into three service modes:

managers, professionals and water users (Yang,

2015). The manager group is defined as the

government manager. The focus of this group on

water-saving knowledge is mainly in the construction

of water-saving society, water-saving policies and

regulations, water use efficiency of enterprises, etc;

Professionals mainly refer to scholars and engineers

who study water-saving systems, technologies, laws

WRE 2021 - The International Conference on Water Resource and Environment

410

and regulations, etc. Their focus is on the evaluation

of water-saving effect and the promotion of water-

saving system; Water users refer to the actual water

users, including residential water users and non

residential water users. Non residential water users

mainly refer to enterprise water users, irrigation area

water users, etc. This group mainly focuses on the

improvement of their water use efficiency and water-

saving potential (Si, 2009). The specific classification

of knowledge service types is shown in Figure 18.

Figure 17: Flow chart of knowledge provision based on

retrieval service.

Figure 18: Division of knowledge service types.

3.5.3 Take the Implementation of

Knowledge Retrieval Service for

Managers as An Example

This section mainly introduces the case realization of

water-saving knowledge service in the service mode

of knowledge retrieval with managers as the user

group. First, the service modules of the water saving

service include retrieval service, knowledge

recommendation, concerned topics and management,

WeChat official account and so on.

Enter industrial water saving in the keyword

search box, and return the search results of the

knowledge map related to industrial water saving

knowledge to the user through the retrieval of water

saving cloud services and knowledge

recommendation services based on topics and

personal concerns. Click the industrial water saving

assessment icon to enter the main interface of

industrial water saving assessment.

Figure 19 is an introduction to the knowledge map

of industrial assessment, which can be downloaded

locally:

Figure 19: Functions of industrial water saving assessment interface.

After downloading, the web page will transfer the

real-time download times of the knowledge map to

the cloud service; After browsing the knowledge

map, you can score the evaluation of the knowledge

map. The cloud service evaluates the knowledge

through the knowledge evaluation algorithm by

downloading times and scoring. And the web page

will recommend knowledge maps with similar topics

to users.

Research on the Realization of Water-saving Social Service Application based on Knowledge Cloud Service

411

4 CONCLUSION

This paper completes the iteration of providing and

updating water-saving knowledge based on cloud

services, and provides targeted and timely water-

saving knowledge services for users according to

different user groups. In terms of website

construction technology, Spring boot, Spring cloud

and Docker technologies are used to realize the

construction of web pages by micro services, and the

functional modules of the website are divided.

Through the knowledge cloud service, it connects

the application of water-saving knowledge base and

water-saving knowledge service, makes the

knowledge recommendation dynamic and

continuously updated, and improves the quality of

knowledge content and the efficiency of knowledge

service application. The service website and cloud

service are interrelated. According to the

characteristics of water-saving knowledge service

objects, the user groups are divided, and the design

and implementation of knowledge service process for

different groups and different knowledge service

modes are completed. The development of water-

saving socialized service application not only

improves and supplements the application system of

water-saving knowledge service, but also has a

certain reference significance for the development of

water-saving work in the future.

Based on knowledge cloud service, according to

the concept of water-saving socialized service, a web

portal of water-saving socialized service is developed

based on Java language, using microservice

architecture, Spring boot, Spring cloud, Docker and

other framework technologies.

ACKNOWLEDGEMENTS

This work was supported by Natural Science Basic

Research Program of Shaanxi Province (2019JLZ-

16), Science and Technology Program of Shaanxi

Province (2019slkj-13, 2020slkj-16) and Research

Fund of the State Key Laboratory of Eco-hydraulics

in Northwest Arid Region, Xi’an University of

Technology (Grant No. 2019KJCXTD-5). The

authors thank the editor for their comments and

suggestions.

REFERENCES

Deng, G. Q. (2015). On the Current Siutation and

Countermeasures of Service Oriented Government

Website Construction. Wireless Internet Technology,

16, 39-40+59.

Du, Y., Jiang, W., Ji, S. J, & Li, Z. F. (2021).

Communication impact and operation strategy of

typical WeChat official accounts for publishing centres

of university journals. Chinese Journal of Scientific and

Technical Periodicals, 32(01), 75-82.

Gu, Z. Z., Li, K., Chen, H., Gu, D. X., & Xu, J. (2019).

Investigating Big Data-driven Cloud Service Platform

and Knowledge Service Mechanism for Dynamic

Health Demand. Information Science, 37(11), 106-111.

Guo, J., Zhou, Y., Zhang, P., Song, B., & Chen, C. (2021).

Trust-aware recommendation based on heterogeneous

multi-relational graphs fusion. Information Fusion, 74,

87-95.

Kang, S. Z. (2019). National water conservation initiative

for promoting water-adapted and green agriculture and

highly-efficient water use. China Water Resources, 13,

1-6.

Liu, Z., Wang, L. and Li, X., & Pang, S. (2021). A multi-

attribute personalized recommendation method for

manufacturing service composition with combining

collaborative filtering and genetic algorithm. Journal of

Manufacturing Systems, 58, 348-364.

Qiao, D., Lu, Q., & Xu, T. (2017). Social Networks,

Information Acquisition and Water － Saving Irrigation

Technology Adoption:An Empirical Analysis from

Minqin County, Gansu Province. Journal of Nanjing

Agricultural University (Social Sciences Edition),

17(04), 147-155+160.

Si, L. (2009). Realization of knowledge service based on

knowledge architecture. Library Tribune, 29(06), 216-

219+88.

Sun, J. H., Wang, S. R., Zhu, Q. D., Li, M., & Chen, J.

(2018). Development and enlightenment of water issues

and its governance. Advances in Water Science, 29(05),

607-613.

Sun, Y. S., & Liao, P. (2019) Research Development on

Core Technology of Knowledge Service Evaluation in

China. Computer & Digital Engineering, 47(12), 3045-

3052+3088.

Wang, C. R., & Jeong, M. (2018). What makes you choose

Airbnb again? An examination of users’ perceptions

toward the website and their stay. International Journal

of Hospitality Management, 74, 162-170.

Wang, X. (2016) Why have some water-saving websites

become "furnishings". China Water Resources News,

China 06 September (008).

http://cds.chinawater.com.cn:8088/wasdemo/detail?rec

ord=23&channelid=113513&searchword=bzlm%3D%

272016-09-06%27&keyword=bzlm%3D%272016-09-

06%27

Yang, B., & Yang, M. (2021). Research on enterprise

knowledge service based on semantic reasoning and

WRE 2021 - The International Conference on Water Resource and Environment

412

data fusion. Neural Computing and Applications,

https://doi.org/10.1007/s00521-021-06382-z

Yang, M. (2015) To Strengthen the Construction of

Informatization Construction and Enhance the Level of

Water-saving Management. Modern Industrial

Economy and Informationization, 5(24), 101-103+111.

Zhang, C. W. (2011) The Development of Community

Grids Service Integrated Platform. Library and

Information Service, 55(11), 56-61.

Zhang, S., Xu, J., Gou, H., & Tan, J. (2017). A Research

Review on the Key Technologies of Intelligent Design

for Customized Products. Engineering, 3(5), 631-640.

Zheng, W., Yin, L., Chen, X., Ma, Z., Liu, S., & Yang, B.

(2021). Knowledge base graph embedding module

design for Visual question answering model. Pattern

Recognition, 120, 108153.

Zhou, N., Cheng, H. L., & Chen, H. Q. (2007) Website

Information Architecture and Information

Visualization. Library and Information Service, 9, 58-

61.

Research on the Realization of Water-saving Social Service Application based on Knowledge Cloud Service

413