=Paper=
{{Paper
|id=Vol-3734/invited5
|storemode=property
|title=Business management data analysis method based on big
data computing science
|pdfUrl=https://ceur-ws.org/Vol-3734/paper5.pdf
|volume=Vol-3734
|authors=Kai Wu
|dblpUrl=https://dblp.org/rec/conf/iccic/Wu24
}}
==Business management data analysis method based on big
data computing science==
https://ceur-ws.org/Vol-3734/paper5.pdf
Business management data analysis method based on big
data computing science
Kai Wu1, 2, ∗
1 Claro·M. Recto Academy·of Advanced Studies, Lyceum of the Philippines University, Manila.
2 Wuxi Institute of Technology, Wuxi, Jiangsu, China
Abstract
Under the development trend of economic globalization, big data computing and science have
been widely used in the management and data analysis of industrial and commercial
enterprises. Relevant technical theories can not only strengthen the comprehensive
management level of enterprises, but also provide a new impetus for business management
decision-making. Especially in the increasingly competitive market environment, if the
traditional decision-making management model continues to be applied, it is difficult to meet
the needs of the development of the new era, so it is necessary to scientifically conduct data
analysis and management decisions based on big data computing. After understanding the
operation mechanism of big data and its impact on the decision-making composition of business
management enterprises, this paper defines the enterprise-level data warehouse architecture
and performance optimization results according to the enterprise comprehensive data analysis
platform with big data as the core, and finally defines the enterprise management
countermeasures using big data computing science, so as to promote the better development of
enterprises.
Keywords
big data, computational science, business administration, enterprise, data analysis
1. Introduction
Under the development trend of economic globalization, big data is the basic strategic
resources for the construction of modern society development, in promoting business
management enterprise management, big data technology science has unique application
value, effective use can improve the industrial and commercial management system, deep
mining more data information, give full play to its potential value. In essence, big data is a
huge collection of data with various functions, which contains a very large amount of data
information and scale, far beyond people's knowledge and understanding of information
use [1]. In the process of big data collection, storage, management and analysis, the
reasonable use of professional software and technical equipment can better meet the
system application requirements. Nowadays, some scholars define big data as data
ICCIC 2024: International Conference on Computer and Intelligent Control, June 29–30, 2024, Kuala Lumpur,
Malaysia
∗ Corresponding author.
kai.wu@lpunetwork.edu.ph (K. Wu)
0009-0006-5896-1009 (K. Wu)
© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
information with large scale, fast transmission speed and relatively rich types, and big data
technology can obtain and store various technologies used by big data in a short time.
Once this concept was put forward, it has been highly valued by all sectors of society, and
has been widely used in many fields in practice and exploration [2]. From the perspective
of business administration enterprises, putting forward management decisions based on
big data computing science can provide more technical support in enterprise construction
and reform. Big data technology has the following advantages in business administration
and enterprise management: First, the storage and use of data information continue to rise,
which is more efficient than traditional data storage and analysis; Second, the application
speed of data flow is getting faster and faster, and the requirements of information
processing are getting higher and higher. Third, the overall data type is more abundant,
including structured and unstructured forms, the former is the use of basic behavior of
data information, in the process of processing in accordance with a specific mode of
operation, while the latter has a diversified composition, whether it is click, or
transmission can obtain a lot of information [3].
From the current perspective of enterprise business management decision-making, big
data computing science has the following impacts: First, decision-making environment.
With the continuous increase of data information, enterprises are gradually entering the
development stage of big data. How to effectively use corresponding technologies to obtain
economic laws and market characteristics, and ensure that management decisions are
targeted and effective is the core topic for enterprises to discuss at present. According to
the impact of big data environment on enterprise decision-making, information collection,
data application, etc., the actual management decision-making environment has changed
and has very significant data-driven characteristics, which lays a solid foundation for
enterprise reform and exploration in the new era. Second, decision data. The core of big
data is not data, but the potential value of data will have an impact on the development of
enterprises [4.5]. Therefore, enterprises should not only have the ability of information
collection, but also form the ability of data integration research in the practice and
development. With the faster and faster production speed of network information and
data, enterprise decision-making should put forward implementation requirements for
processing information, analyze the relationship between various data according to the
nature of big data, so as to grasp more valuable content and provide basis for business
management decision-making. Third, decision technology. In order to effectively process
more information and data, enterprises have developed and applied corresponding
technology platforms in practice and exploration, among which cloud computing platform
provides technical support for data management and facilitates the presentation of more
content to enterprise managers. In this process, the rational use of knowledge mining
technology to find data links can obtain more quality services in technological innovation.
Especially after entering the new era, with the continuous improvement of economic and
technological level, the market has a variety of information transmission channels, and
enterprises use big data computing methods to obtain diversified information data, which
can not only put forward perfect management methods, but also create high-quality
conditions for catering to the market [6]. From the current situation of enterprise
construction management, big data computing and science have not been rationally used,
the corresponding technical advantages have not been fully demonstrated, and there are
still many problems during business management. Therefore, this paper starts from the
perspective of the impact of big data computing science on business management
enterprises, and after understanding the operation mechanism of big data and the
architecture of enterprise data warehouse, defines the method of enterprise management
data analysis based on big data computing science [7].
2. Method
2.1. Big data technology
Big data technology refers to the modern technology theory including language processing,
computer technology, artificial intelligence, data analysis, statistical analysis and many
other technologies. From the perspective of practical application, the types of technologies
and application tools included are shown in Table 1 below. Using big data to create a
comprehensive data analysis platform for enterprises can ensure the adaptability of
various business and management work [8].
Table 1
Types of Big Data Technologies and Application Tools
Types of big data technologies Big data technology and tools
Infrastructure Cloud computing platform, network technology,
cloud storage, virtualization technology, resource
monitoring technology
Data acquisition and basic ETL, tools, data intelligent perception technology,
processing web crawler.
Data storage The fusion of relational database and non-relational
database, main memory database
Calculation and analysis of data Data prediction and mining, data query and
analysis, BI business intelligence, map processing.
Presentation of data Visualization tools, graphs and reports, realistic
technology
2.2. Process of data mining
The process of data mining is shown in the figure below, which mainly includes the
following steps:
1. Determine the research problem domain: Clear mining needs and set goals are the
first steps to carry out mining operations.
2. Select the target data set: select some data from the data source as the mining target
according to the requirements.
3. Data preprocessing: targeted processing of the target data set, so that it can be
directly carried out high-quality mining data [9].
4. Implementation of data mining: according to the mining requirements for the actual
operation to generate patterns. In this stage, it is necessary to consider the algorithm
matching problem, which usually takes into account two aspects: data characteristics and
the requirements of the operating system [10].
5. Interpretation and evaluation model: the work that needs to be done before the final
result is presented to the user.
Figure 1：The general process of data mining
It is necessary to filter out a large number of patterns formed in the previous mining
phase, eliminating some useless or uninteresting patterns. If no useful patterns are found
among these patterns, it is necessary to return to the initial problem domain description
phase and re-mining.
6. Knowledge: apply the final patterns obtained using the above steps to practical
problems, guiding human behavior with useful knowledge [11].
2.3 Data Mining System
Figure 2 illustrates the mainstream architecture formed by the development of the system
to today's fourth-generation mining platform. Several issues need to be considered when
designing a data mining system: The solution to these issues forms the basis for the
system's ability to conduct mining. Currently, the most common data source systems are
databases and data warehouses. The integration of mining systems with data sources is
mainly considered in four solutions: decoupling, loose coupling, semi-tight coupling, and
tight coupling. Among these solutions, decoupling is the worst design while tight coupling
is the most ideal design. Tight coupling can effectively achieve mining objectives, improve
mining efficiency, and facilitate information exchange between different systems; however,
its implementation is very difficult. A compromise solution is semi-tight coupling [12].
Figure 2: Data mining classic system diagram
2.4. Enterprise comprehensive data analysis platform
In the context of the application of big data computing science, the enterprise
comprehensive data analysis platform architecture as shown in the following figure is
constructed, which includes the following levels: [13]
Figure 3: Platform architecture diagram
First, the platform service layer. This level design includes PaaS platform management
and IaaS platform management. The former mainly provides users with two kinds of
middleware services, mainly big data middleware and general middleware. The latter
provides network services, storage services, cloud services, and computing services.
Second, the data service layer. This level of design includes multiple links, among which
the integration and processing of professional data is to ensure the standardization and
standardization of relevant data management, and it is necessary to process the relevant
content in accordance with the business requirements of the enterprise while processing
the conventional data, so as to provide a reference for the project implementation. Data
exchange will use the database to complete data acquisition, batch collection and
encryption processing operations; Computing analysis will provide users with crawler
services, data mining, management services, etc., to improve the user's application
experience; Data support can ensure the standardized control of relevant data, ensure the
security and confidentiality of data information, and lay a solid foundation for the stable
development of enterprises [14].
Third, the application service layer. In this level of design, according to the needs of the
enterprise's comprehensive data analysis platform, it will deeply understand the operation
of the enterprise, ensure that the follow-up work of the platform can be carried out
normally and stably, and lay the foundation for the standardization and high efficiency of
the platform.
2.5. Platform functions
In order to obtain a powerful and practical enterprise comprehensive data analysis
platform, the platform functions can be designed according to the following figure:
Figure 4: Structure diagram of platform functions
First, data quality management. This function mainly carries out the storage design and
development of the platform, including data filtering, data exchange and data storage, etc.,
which can help managers find and solve various problems, improve the operating
efficiency and supervision level of the platform; Second, update the comparison. This
function can use the exchange platform to complete data summary and data exchange,
comprehensive comparative analysis of the obtained data information, and ultimately
ensure the authenticity and effectiveness of the results, and transfer it to the designated
platform for storage; Third, data mining. This function will be organized according to the
information catalog, the result will be multiple links as a unified whole, make full use of big
data calculation, complete the design and implementation of information mining
scientifically, and finally complete the construction and application of reasoning model
with the help of computer technology in-depth analysis; Fourth, update monitoring. This
function will use the platform to analyze and mine different types of information data, and
on this basis, find the correlation between various data, find the problems existing in the
platform according to the dynamic monitoring, and automatically send an early warning
signal to notify the professional maintenance and treatment, so as to ensure the safety of
the platform application operation [15].
2.6. Data warehouse architecture
Data warehouse mainly provides analysis and decision-making services for enterprises.
The specific architecture is shown in Figure 5 below:
Figure 5: Structure diagram of platform functions
First, data quality management. This function mainly carries out the storage design and
development of the platform, including data filtering, data exchange and data storage, etc.,
which can help managers find and solve various problems, improve the operating
efficiency and supervision level of the platform; Second, update the comparison. This
function can use the exchange platform to complete data summary and data exchange,
comprehensive comparative analysis of the obtained data information, and ultimately
ensure the authenticity and effectiveness of the results, and transfer it to the designated
platform for storage; Third, data mining. This function will be organized according to the
information catalog, the result will be multiple links as a unified whole, make full use of big
data calculation, complete the design and implementation of information mining
scientifically, and finally complete the construction and application of reasoning model
with the help of computer technology in-depth analysis; Fourth, update monitoring. This
function will use the platform to analyze and mine different types of information data, and
on this basis, find the correlation between various data, find the problems existing in the
platform according to the dynamic monitoring, and automatically send an early warning
signal to notify the professional maintenance and treatment, so as to ensure the safety of
the platform application operation.
Table 2
Database for three sets of simulation experiments
Initial Performance Index of The First Round of Second Round Third Round
Vertica Database Testing Test Test
Concurrency 20 40 80
Number of errors 876（1） 48 688（2）
Average time consumption 00：01：29 00：03：28 05：42
Minimum time consumption 00：00：01 0：00：01 00：00：02
Maximum time consumption 00：47：45 1：00：01 00：49：27
Total time spent executing 138：57：40 325：31：18 535：24：00
query
Testing time-consuming 06：59：00 08：26：52 08：26：00
Number of successful queries 13.45 11.12 11.12
per minute
Table 3
Performance Results
Concurrency 24
Number of errors 43（1）
Average time consumption 00：00：00：38
Minimum time consumption 00：00：00：00
Maximum time consumption 00：19：56
Total time spent executing query 06：05：37
Testing time-consuming 02：18：17
Number of successful queries per minute 40.75
Based on the comparative analysis of Table 2 and Table 3above, it is found that the
database optimized by performance adjustment has been improved in terms of query,
which is embodied in five aspects: average time consumed, maximum time consumed, total
time consumed for executing queries, test time consumed, and the number of successful
queries per minute. According to the result summary analysis, compared with SeaQuest
database, the underlying database of EDW can execute most query requests faster; During
the test period, the resource pool of the system had realistic constraints on the query
requests to run. Without these constraints, some queries would consume more time. After
adjustment and optimization, various indicators of the database have been greatly
improved compared with the previous, such as the average time consumed per query has
been increased by about 5.5 times, the time consumed by drunk queries has been
increased by about 3 times, the number of successful queries per minute has been
increased by about 3.6 times, and the application efficiency of system resources has been
increased by about 2.8 times.
2.7. Application countermeasures
In view of the application and construction platform of big data computing, science in
business administration, enterprise management and data analysis, in order to fully
demonstrate the application value of big data, we should start from the following aspects:
First, strengthen information construction. Enterprises should put forward strategic plans
that are consistent with their own business nature, start from a macro perspective, deal
with information construction at different levels, gradually improve the management
countermeasures and configuration processes of the decision-making system, and
strengthen the internal information management level of enterprises. Secondly, determine
the value orientation. Whether it is internal business optimization or strategic plan
determination, in-depth analysis should be made according to customer needs to ensure a
good communication bridge between the enterprise and customers. Department
employees can use data analysis to sort out customer information, so as to facilitate the
adjustment of business direction, and determine infrastructure and management
mechanism. Finally, improve the decision-making mechanism. In order to demonstrate the
application value of big data computing science during the decision-making of business
administration enterprises, staff should propose a sound management mechanism based
on enterprise decision-making, pay attention to refining data management, data analysis,
performance appraisal and other related content, and ensure that data collection and
application can be implemented in all aspects, so as to improve the effectiveness of
business administration decision-making. At the same time, it is necessary to regularly
organize employees to participate in training activities with big data as the theme, actively
learn new computational science knowledge, rationally use basic means of big data to
solve business problems, and propose visual means in line with business management
decision-making in technological innovation, so that information analysis and information
extraction become more simple and convenient, and ensure the scientific management
decisions of enterprises.
3. Conclusion
To sum up, business administration enterprises use big data computing science to build a
comprehensive data analysis platform, complete data analysis and data application in an
orderly manner, which can facilitate managers to grasp the actual situation of enterprise
operation faster, reasonably control enterprise cost expenditure, improve the economic
and social benefits of enterprise operation, and provide technical support for enterprise
development.
Reference
[1] L. Wang, A. Chen, H. Wu, “Analysis of enterprise management mode based on big data,”
Chlor-alkali Industry, pp. 39-45, 2023.
[2] N. Tan, “Research on digital operation method of Big Data Platform,” Computer
Knowledge and Technology: Academic Exchange, 2022.
[3] L. Zhong, Y. Chang, F. Wang, “Distributed dynamic industrial process based on analysis
of characteristics of slow running state evaluation,” Journal of Automation, pp. 1-13,
2024, doi: 10.16383 / j.a as c230154.
[4] L. Liu, “Analysis and key points of big data on business management decision,”
Modern Business, pp. 100-103, 2022.
[5] Z. Han, “Based on big data era under the background of the innovation of the
enterprise management mode to explore,” Journal of Economics and Management
Research, pp. 59-60, 2022, doi: 10.12346 / emr v4i1. 5296.
[6] F. Song, J. Li, Y. Wei, “Application research of intelligent electric drive bridge foresight
Technology based on DMBOK,” Mechanical and Electrical Engineering Technology, pp.
159-162, 2023.
[7] M. Tang, “Exploration and Practice of Using data analysis to promote enterprise
management Improvement,” Science and Information Technology, pp. 190-192, 2022.
[8] Q. Xing, “Analysis on the innovation path of enterprise business administration under
the new normal of economy,” Chinese Science and Technology Journal Database
(full-text Edition) Economic Management, 2022.
[9] A. Zhang, A. Li, “Research on financing problems of Small and medium-sized
enterprises,” Chinese Science and Technology Journal Database (full-text Edition)
Economic Management, 2022.
[10] J. Gu, Q. Li, Y. Ban, “Whether digital transformation can promote import expansion of
Chinese enterprises: Empirical analysis from listed companies,” The Southern
Economy, pp. 123-141, 2023. doi: 10.19592 / j.carol carroll nki scje. 410643.
[11] K. Jia, “Economic structure transformation of enterprise business management
analysis,” Journal of Economics and Management Research, pp. 7-9, 2023. doi:
10.12346 / emr v5i1. 8238.
[12] L. Luo, Q. Ye, S. Wang, “Study on the influence of party organization on enterprise
management efficiency: an empirical analysis based on the survey data of
enterprise-labor matching in China,” Photocopying Newspaper Materials: Labor
Economy and Labor Relations, 2022.
[13] Z. LIU, B. Wang, F. Liu, “Application of data analysis in textile and garment enterprise
management,” Woolen Textile Technology, 2022.
[14] J. Shi, “Analysis of enterprise business administration innovation path under the new
normal of economy,” Chinese Science and Technology Journal Database (full-text
Edition) Economic Management, 2023.
[15] D. Deng, “Analysis and key points of big data on enterprise business management
decision,” Financial Management: Collection, 2023.