Research on Enterprise Financial Management System Based on Big

Data Hadoop

Ying Liu

, Tingting Fan

and Sixuan He

Chongqing College of Architecture and Technology, Chongqing, China

Keywords: Big Data Technology, Enterprise Financial Management System, Hadoop, Web Technology.

Abstract: With the development of economy, enterprises pay more and more attention to the construction of financial

management system. However, as far as the current development of financial management system is

concerned, there are still some problems, such as unclear financial system planning, imperfect financial data

processing flow and low processing efficiency. Therefore, this paper constructs an enterprise financial

management system based on big data Hadoop. This system takes Linus system as the bottom operating

system, Java as the overall development environment of the system, and JDK version above 1.8 to keep the

system running normally. Hadoop architecture is installed and deployed in JVM virtual machine environment,

and components such as Flume, Sqoop, Hive, HDFS, MapReduce, SQL Server and Echarts are installed and

deployed in turn according to the processes of data capture, data cleaning, data analysis and processing, data

storage and data visualization, so as to realize the management and call of system data. The construction of

the system can effectively improve the data storage performance and overall security performance of the

financial management system, enrich the overall functions of the system, and make the working system of the

financial management system more perfect.

1 INTRODUCTION

In the era of big data with the development of science

and technology, the financial management system of

enterprises has become the focus of enterprise

modernization. Financial management lays the

capital foundation for the subsequent development of

the enterprise and determines the development

prospect and direction of the enterprise. Excellent

financial management system can improve the

economic effectiveness of enterprises, avoid the

financial risks of enterprises, and work out a long-

term plan suitable for the development of enterprises.

At present, China is in an important period of

realizing the great rejuvenation of the Chinese nation,

with the deepening of the reform of the economic

system, and the state's requirements for the financial

management of enterprises are more detailed. How to

build a scientific enterprise financial management

system is not only the focus of national development,

but also the subject of this paper. However, as far as

the current development form is concerned, there are

https://orcid.org/0000-0002-8144-1822

https://orcid.org/0000-0003-0294-3865

still some problems in the enterprise financial

management system. First of all, the financial

management system planning is not clear, and the

division of labor is scattered, which is difficult to

meet the actual needs of users. Secondly, the

processing flow of the financial management system

is not perfect enough, which is easy to cause data loss

and difficult to guarantee the accuracy of data results.

Finally, the processing efficiency of the financial

management system is low. The financial

management system needs to convert the data with

different formats into numbers and store them in the

library. The existing financial management system is

still using the old system, which leads to the poor

effectiveness, low efficiency and insufficient intuitive

data display of enterprise financial work. In view of

the above problems, enterprises should actively

cooperate with national policies, introduce advanced

science and technology into the financial

management system, and actively build an efficient

enterprise financial management system.

https://orcid.org/0000-0001-5603-227X

316

Liu, Y., Fan, T. and He, S.

Research on Enterpr ise Financial Management System Based on Big Data Hadoop.

DOI: 10.5220/0012030100003620

In Proceedings of the 4th International Conference on Economic Management and Model Engineering (ICEMME 2022), pages 316-321

ISBN: 978-989-758-636-1

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

The financial management system does not

simply organize and store unstructured data such as

pictures and words. It can proceed from these

unstructured data job scheduling and data parallel

processing, and then adopt parallel processing and

distributed storage according to the details and

complexity of these data, so as to realize reasonable

data management. The modern financial management

system has promoted the progress of enterprises to a

certain extent and improved the economic benefits of

enterprises. Therefore, this paper believes that using

big data technology, combining Hadoop with Web

technology to build a web-based enterprise financial

management system, relying on the advantages of big

data technology in data quality, data effectiveness,

data collection, data high-speed operation and

calculation, etc., can realize the network,

digitalization and intelligence of enterprise financial

management business, and provide substantial

reference for the successful construction of enterprise

financial management system.

2 INTRODUCTION OF KEY

TECHNOLOGIES

2.1 Big Data Technology

In the information age with the development of

science and technology, data has gradually penetrated

into all aspects of life. From work communication to

daily chores, these bits and pieces of information data

gradually form a huge data collection, that is, big data

resources. This collection of data resources has the

characteristics of large scale, fast circulation and

various types. Because of its huge number, traditional

data processing methods are difficult to process it in

real time, so big data processing technology comes

into being. Big data technology can explore the data

set as a whole, extract key information from it, and

then recommend activities according to users' daily

usage tendency, which provides convenience for

users' daily life. (Yang, 2018) With the improvement

of social demand, big data technology has also been

innovated and developed, so big data technology

stack came into being. Big data technology stack can

be divided into the following working layers in detail:

data acquisition layer, data storage layer, resource

management service coordination layer, calculation

engine layer and data analysis layer. The specific

structure is shown in Figure 1.

Figure 1: Structure diagram of big data technology stack (Source: https://blog.csdn.net/eternal _

188/article/details/104940054).

2.2 Web Technology

Web is an application architecture based on the

Internet, and its core is to provide users with various

forms of information content and information

services. With the rapid development and application

of Web technology, the development and application

of its architecture presents a trend from simplicity to

Research on Enterprise Financial Management System Based on Big Data Hadoop

317

complexity. In the development process of Web

server-side technology, the core of Web back-end

development technology is to use a variety of object-

oriented programming languages to complete the

development of server-side functions. The content

involved in the development is rich, including

business logic, data storage and message queue

processing, as well as the design and definition of

several data interfaces, such as Web front-end

interface, third-party interface, server internal

interface, etc. The emergence of CGI technology has

changed the situation that Web servers can only

transmit static files. By executing external programs

through CGI, the external programs can generate

dynamic HTML pages according to the content of

Web requests, thus realizing the dynamic information

exchange between the client and the server. Since

then, Web server-side development technology has

entered the era of framework and template, and the

auxiliary development technology of Web has

increased the convenience for users to develop Web

programs and improved the overall efficiency of

program construction.

2.3 Hadoop

Hadoop is a software platform used to analyze and

process big data. With the help of Java language, open

source distributed system infrastructure is built. Users

can process distributed programs with complicated

data on this platform, and realize parallel high-speed

operations and complex calls to big data. Hadoop

architecture has become the most popular big data

analysis system because of its low development cost,

strong extensibility and high fault tolerance. As a

multi-component architecture, it mainly includes core

components such as HDFS, MapReduce, HBase, and

other plug-ins with data capture or transmission

functions such as Sqoop and Flume. (Wei, 2021) The

deployment of various components can provide many

functions for Hadoop architecture, and Hadoop

architecture is the core of the ecosystem. The overall

framework of Hadoop is shown in Figure 2.

Figure 2: Overall framework of Hadoop (Source: https://blog.csdn.net/w12345_ww/article/details/51910889)

2.3.1 HDFS

HDFS is composed of a Namenode node and several

Datanode. In the process of system operation, it

usually runs in the mode of manager and worker.

Namenode is the manager of the cluster, responsible

for managing the namespace of the file system and

maintaining the metadata of the file system. Datanode

is the storage node of data, and it is also the worker of

the cluster. The data will be divided into multiple file

blocks with the size of 64MB, distributed on multiple

machines, and its own storage information will be

sent to Namenode regularly.

2.3.2 MapReduce

When Mapreduce calculates files, it will read them

line by line and process them in batches. The

execution process of Mapreduce task can be divided

into map stage and reduce stage. The input data of the

Map stage are files stored on HDFS, which will be

divided into multiple file blocks, each file block

corresponding to a map process.

2.4 Development Environment

According to the above application requirements, we

can build and deploy the development environment.

Under the hardware environment, Linus system is

ICEMME 2022 - The International Conference on Economic Management and Model Engineering

318

adopted as the bottom operating system of the system,

Java environment is used as the basic development

environment, and JDK version 1.8 or above is

selected to facilitate the subsequent software

installation. Hadoop's overall architecture is installed

and deployed in the JVM virtual machine

environment, and components such as Flume, Sqoop,

Hive, HDFS, MapReduce, SQL Server and Echarts

are installed and deployed in turn according to the

processes of data capture, data cleaning, data analysis

and processing, data storage and data visualization.

With the help of the structure of the system, it is

layered one by one, so as to realize the analysis and

call of the data belonging to the business system,

financial system and other management systems by

the enterprise financial management system, and

ensure the overall efficiency of enterprise work. In the

overall development process of the system, the big

data technology represented by Hadoop architecture

is deployed on the Web server, and then Nginx is

selected as the Web server, so as to improve the

system's handling of static data content and control

the number of concurrent operations of the system as

a whole. SQL Server is selected as the database

server, Java language is selected as the system

development language, JSP technology is selected as

the Web client page, and then Java language and

HTML language are used to complete the basic

construction of the system page. Through the

introduction of the above key technical theories, the

overall environment of system development, the

configuration of related software and tools are

determined, and the technical feasibility of the overall

project of building enterprise financial management

system is also clarified.

3 SYSTEM REQUIREMENTS

ANALYSIS

At present, China's economic system reform is

gradually deepening, and the state has put forward

higher requirements for the development of

enterprises. Because financial management runs

through all the work of enterprise management,

financial management is the foundation of enterprise

management and development. Enterprises should

follow the trend, upgrade the original management

system with the help of new scientific and

technological means, and build a scientific financial

management system. How to manage enterprise

finance efficiently needs to grasp the enterprise

financial theory and capital structure, and make the

next planning according to the practical experience of

enterprise financial management. (Li, 2013) Based on

the big data technology, the enterprise financial

management system will meet the enterprise's

demand for refinement of its own management, and

achieve the whole process control of the enterprise's

daily production and operation management

activities, such as planning beforehand, being

cautious in the process and summarizing afterwards.

Compared with the conventional financial

management system, the financial management

system combined with modern technology

constructed in this paper can rely on network

information technology and big data analysis and

processing technology to collect and analyze all the

data related to the development of enterprise

production experience management. It can

thoroughly solve the problem of data disconnection

between various business departments and financial

departments of an enterprise, improve the situation

that enterprise managers or decision-makers only rely

on financial and accounting reports to obtain the

business development status of the enterprise,

effectively realize the deeper integration of business

and financial information, and provide timely,

accurate, comprehensive and personalized

information or reports to all responsible units,

departments and management within the enterprise

by orderly processing and transmitting relevant data.

4 FUNCTION REALIZATION

4.1 Financial Side

4.1.1 Voucher Management Module

When users use the system for the first time, they first

need to complete the user registration according to the

relevant guidance given by the page, and then they

can log in to the system for subsequent operations.

The login code is shown in Figure 3. Under this

module, accounting vouchers are divided into paper

vouchers and electronic certificate vouchers. After

the user enters the paper voucher, click the

proofreading button, and the system will proofread

the entered information in the manuscript accurately,

and any errors will be marked with red letters. This

module also has two modules: voucher posting and

voucher auditing. Users can generate corresponding

bookkeeping vouchers for the entered and approved

vouchers, and then collect them in the library for

classified storage. When querying, enter the date or

voucher number in the search field to query.

Research on Enterprise Financial Management System Based on Big Data Hadoop

319

Figure 3: System accession code (Original).

4.1.2 Financial Revenue and Expenditure

Module

Users can use this module to record the daily

transactions of enterprises, which can be divided into

two parts: financial revenue and financial

expenditure. In financial revenue, users can use the

module scanning function to convert paper

documents into electronic data and store them in the

library, which is easy to find. The financial database

will divide the income data according to the time and

type, generate the income report in the fixed node,

and submit it to the report management module for

the enterprise leaders to consult. In financial

expenditure, users can query the capital flow

according to expenditure records. (Gui, 2013) With

the function of this module, users can generate

revenue and expenditure comparison reports

according to the needs of enterprises, and then upload

them to the report management module for

subsequent comparative review. The upload code is

shown in Figure 4.

Figure 4: Report upload code (Original).

4.2 Leadership Side

4.2.1 Report Management Module

Under this module, the system will automatically

generate reports according to the entered trade

information, which can be divided into daily reports,

profit distribution statements, cash flow statements,

etc. These statements can intuitively reflect the

business status and financial revenue and expenditure

of an enterprise, and users can choose the generation

of data charts according to their personal needs, such

as line charts, pie charts or bar charts. (Zhang, 2014)

This system omits the process of data operation, and

turns the data results into charts and graphs, which

can be displayed to users more intuitively, so as to

compare with the previous data and make a more

detailed development plan for the subsequent

development of the enterprise.

4.2.2 Foreground Building Module

Under this module, users can use the query function

in the system to obtain the required financial data of

internal and external departments of the enterprise.

The function of this module is to use Flume to capture

the data of each system's log files, then use Sqoop to

transmit the data of each business system, financial

system, management system, accounting system and

external database, and then use the web crawler in

Scrapy framework to capture and collect the

unstructured data. When the whole data is extracted,

it is stored in the distributed file system, waiting for

the system to analyze and process the data in more

detail. Finally, through the analysis and calculation of

MapReduce or Spark and other computing engines,

the corresponding data analysis results are formed,

and the report contents are presented in visual data

charts. According to the report content formed by the

system, the enterprise can make a detailed analysis of

the future development strategy and formulate a long-

term business strategy to ensure the stable

development of the enterprise.

ICEMME 2022 - The International Conference on Economic Management and Model Engineering

320

5 CONCLUSIONS

In the Internet era of big data development,

developing new system programs based on various

data has become the focus of modern enterprise

information construction. The enterprise financial

management system based on big data Hadoop

constructed in this paper integrates big data

technology with enterprise financial management,

which can effectively improve the data storage

performance and overall security performance of the

financial management system, enrich the overall

functions of the system, and make the working system

of the financial management system more perfect. At

the same time, it also improves the business

processing efficiency of employees, makes the

financial management system of enterprises run more

efficiently, and promotes the further development of

enterprises. In the follow-up research, we will further

expand the extensibility and applicability of the

system, and make the system function more perfect.

REFERENCES

Gui Jing. (2013) Design and Implementation of Financial

Management System for Small and Medium-sized

Enterprises[D]. Xiamen University.(05).

Li Dan. (2013) Design and Implementation of Financial

Management System for Small and Medium-sized

Enterprises[D]. University of Electronic Science and

Technology of China.(01).

Wei Pengjuan. (2021) Key Technologies of Web Front-end

Development[J]. Electronic Technology & Software

Engineering.03.

Yang Yi. (2018) Introduction to the Basis and Application

of Big Data Technology[M]. Beijing: Publishing House

of Electronics Industry.07.

Zhang Zhe. (2014) Design and Implementation of

Enterprise Financial Management System[D]. Jilin

University.

Research on Enterprise Financial Management System Based on Big Data Hadoop

321