THE APPLICATION OF BIG DATA
TECHNOLOGY IN THE PREDICTIVE
ANALYSIS OF ENTERPRISE CAPITAL
OPERATION RISK
Jian Wang
Sejong University, Gwangjin-gu, Seoul, 05049, South Korea
lingke1988@163.com
Yuzhen Wang*
Sejong University, Gwangjin-gu, Seoul, 05049, South Korea
Reception: 05/03/2023 Acceptance: 25/04/2023 Publication: 15/05/2023
Suggested citation:
Wang, J. and Wang, Y. (2023). The application of big data technology in the
predictive analysis of enterprise capital operation risk. 3C TIC. Cuadernos
de desarrollo aplicados a las TIC, 12(2), 227-242. https://doi.org/
10.17993/3ctic.2023.122.227-242
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
227
ABSTRACT
The background of the big data era makes enterprise tax management face many
opportunities and challenges, in order to improve the management of enterprise
capital operation risks and promote the enterprise to take the road of sustainable
development. This paper firstly indexes risk names with the help of web crawler
technology, establishes data sources, and then circulates the crawler to obtain the
required information. Secondly, a hashing algorithm is applied to compress the
massive data into a unique and extremely compact section of hash values by means
of constant mapping. Then association rules are used to determine the set of frequent
risk items, and the values of the two are continuously changed to derive the final
predictive analysis. Finally, a capital operation risk prediction and analysis platform is
built by combining the above processes. In this paper, the effectiveness of the
proposed platform is verified, and the practical results show that the accuracy of the
proposed platform for risk prediction discovery is as high as 97%, and the time spent
for risk discovery is controlled within 30 minutes. The relevant data results verify that
big data technology improves the accuracy of enterprise capital operation risk
prediction and analysis while accelerating the speed of risk discovery.
KEYWORDS
Web crawler; hashing algorithm; hash value; association rule; frequent risk item set.
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
228
INDEX
ABSTRACT
KEYWORDS
1. INTRODUCTION
2. CORPORATE CAPITAL OPERATION
2.1. Capital Operation
2.2. Capital operation risk
2.3. Characteristics of capital operation risk
2.3.1. Objectivity of capital operation risk
2.3.2. Variability of Capital Operating Risks
2.3.3. Predictability of Capital Operation Risk
3. Risk prediction under big data technology
3.1. Data Acquisition
3.2. Data Storage
3.3. Association rule prediction analysis
3.4. Predictive Analytics Building Platform
4. APPLICATION OF ENTERPRISE CAPITAL OPERATION RISK PREDICTION
ANALYSIS
4.1. Increased accuracy of predictive analysis
4.2. Increased speed of predictive analysis
5. CONCLUSION
REFERENCES
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
229
1. INTRODUCTION
Under the premise of continuous development of the economic market, enterprises
must continuously expand their own business scope as well as business scale and
management mode in order to gain a foothold in the market competition [1].
Therefore, enterprises often use capital operation to promote the efficient operation of
capital and lay the foundation for the improvement of enterprise efficiency. The
opportunities and risks faced by enterprises in capital operation are increased due to
the influence of internal and external environment [2-4]. Therefore, it is necessary to
study how to prevent the risk of enterprise capital operation and countermeasures.
Corporate capital operations have long been a hot topic of research in the industry
[5-6]. The literature [7] states that firms with high social capital exhibit higher levels of
risk-seeking behavior. Moreover, the relevant actions of firms lead to greater volatility
in stock returns and earnings. Thus, it is clear that firms should conduct capital
operations to generate returns while preventing risks from causing greater losses. The
literature [8] constructs a minimum risk versus capital and risk diversification strategy
for investment portfolios, taking into account the most frequent capital risks in various
industries today. Risky capital is placed separately from risk-free capital, so that the
benefit obtained is a weighted average of risk-free assets, while the risk is not a
weighted average of risky assets, spreading the capital risk. The literature [9] used
Bayesian network models in big data technology to calculate the risk of water pollution
and assess the impact of contaminants in water, which identified the critical causes
and thus the risk of adverse accidents. A new model for risk assessment was
proposed in the literature [10]. Preliminary estimates are made with the help of
reference scenario prediction methods and optimistic bias enhancement is performed.
Uncertainties are introduced in the cost-benefit analysis. Thereafter, a quantitative risk
analysis is provided using Monte Carlo simulation. Although the above-mentioned
literature proposes a series of new methods for risk prediction, the proposed methods
do not fully take into account the large size of the risk data, the data storage system is
more settled, and the analysis of the data is not thorough enough, and the
conclusions obtained are not representative.
Therefore, this paper builds an enterprise capital operation risk prediction and
analysis platform based on big data technology. Firstly, with the help of the web
crawler technology in big data collection technology, the capital operation risk is
indexed and relevant data is obtained through continuous cyclic crawling. Secondly,
the hash algorithm in big data storage technology is used to compress the massive
data into unique and extremely compact hash values, and then realize the storage of
massive data. Finally, the set of frequently occurring risk items is determined by using
the confidence and support degrees in the association rules, and the values of both
are continuously adjusted to derive the final capital operation risk prediction analysis
data. In order to verify the effectiveness of the enterprise capital operation risk
prediction and analysis platform built based on big data technology, this paper
analyzes the accuracy and time required for enterprise capital operation risk
prediction and analysis in the simulation experiment and verifies that the enterprise
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
230
capital operation risk prediction and analysis can be achieved quickly and accurately
based on big data technology.
2. CORPORATE CAPITAL OPERATION
2.1. CAPITAL OPERATION
Capital operation as a business concept has a long history and has been
developed and perfected with the formation of a commodity economy and market
economy. Capital operation is becoming an important way for enterprises to enhance
economic efficiency and realize self-value appreciation. By capital operation, it means
that the enterprise operator takes all the tangible or intangible assets and production
factors owned by the enterprise, through flow, fission, combination, optimal allocation
and effective operation in various ways, to gather a large amount of capital in a short
period of time, and make the capital increase rapidly through capital expansion, in
order to achieve the maximum capital appreciation [11-12]. The process of capital
operation is also the concrete implementation process of capital management strategy
and capital movement.
The various aspects of capital operation are interlocked to form a closed loop.
Capital operation requires an all-around control of financing, investment and assets,
etc. The object of capital operation is property rights in the form of stock assets, or
physical capital that can be operated according to securitization and valuation, and is
a capital-oriented enterprise operation mechanism. The capital operation usually
leads to a transfer of ownership or a significant change in the original shareholder
structure. The core issue of capital operation is how to optimize the structure of
production factors to improve the efficiency of capital operation, which includes the
optimization of resource allocation structure, the optimization of industrial capital,
financial capital and property rights capital structure, the optimization of speculative
capital and incremental capital and the optimization of capital operation process.
However, capital operation is a risky economic activity, which can bring great risks
to the enterprise while bringing rapid development opportunities to the enterprise.
2.2. CAPITAL OPERATION RISK
Capital operation risk refers to the possibility of failure of capital operation or failure
of capital operation activities to achieve the expected goals and losses due to the
complexity and variability of the external environment and the limited cognitive ability
of the capital operation subject in the process of capital operation. In simple terms,
capital operation risk refers to the possible loss of the enterprise due to the
occurrence of unfavorable events in the process of capital operation, which is mainly
caused by the uncertainty of the environment.
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
231
According to the definition of capital operation risk, it can be seen that the direct
bearer of capital operation risk is the capital operation subject, i.e., the capital
operation enterprise rather than the owner of the capital, although it also brings losses
to the owner of the capital. Capital operation risk mainly comes from the complexity
and variability of the environment, i.e. the uncertainty of the environment. The
relatively limited cognitive ability of the capital operation subject to the environment is
also an important factor leading to the capital operation risk. There are two
consequences of capital operation risk: failure of capital operation and failure of
capital operation activities to achieve expected goals. Failure of capital operation
refers to the suspension of capital operation activities, while failure of capital operation
activities to achieve the expected goal means that the capital operation activities are
successful but do not achieve the desired efficiency. For example, the merging firm is
forced to terminate the merger due to the anti-merger resistance of the merged firm.
In general, enterprise capital operation risk mainly contains the following aspects,
namely, operational risk, information risk, management risk, legal and regulatory risk,
etc.
Business risk refers to the occurrence of business risk due to the lack of
comprehensive understanding of market information in the actual management
process and the lack of countermeasures for problems in internal management, which
creates problems in the operation process and thus affects the normal operation of
the enterprise [13-14]. Financial risk, on the other hand, refers to the fact that
enterprises do not have scientific planning for financial management work, the use of
funds is more arbitrary, and operational risks are increasing, which leads to financial
risk [15]. In the context of the information age, the processing of market information by
enterprises is not scientific and reasonable enough, which leads to the phenomenon
of information asymmetry and adversely affects the business decisions of enterprises.
As enterprises are subject to state regulation of capital operation, the phenomenon of
inefficient operation and unreasonable setting of capital structure still exists in the
actual operation of enterprises. In addition for the internal management of enterprises,
the deviation of management concept and the mistake of operation within the
enterprises can cause management risks. In addition, capital operations in other
countries can also spill over into the development of developing economies [16]. The
development of the market economy must be based on the relevant national laws and
regulations, and the macroeconomic regulation of the state will be adjusted, which will
have a certain impact on the M&A behavior of enterprises, thus making the operating
costs of enterprises higher.
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
232
2.3. CHARACTERISTICS OF CAPITAL OPERATION RISK
2.3.1. OBJECTIVITY OF CAPITAL OPERATION RISK
Objectivity is the essential characteristic of capital risk, as can be seen from the
definition of capital risk. Like all other risks, capital risks do not exist at the will of the
operator. It exists objectively regardless of whether the operator acknowledges it or
not, or whether he is aware of it or not. Capital operation risk exists not only in the
preparation stage and the operation stage of capital operation but also in the
commodity operation stage after capital operation.
2.3.2. VARIABILITY OF CAPITAL OPERATING RISKS
Capital operation risk can change under certain conditions. The probability of
occurrence, the degree of impact, and even the scope of impact of capital operation
risks are different in each period and each link of capital operation and under various
conditions. This requires the capital operating entity to make full use of various
methods and means to identify and prevent risks in the process of capital operation
risk prevention.
2.3.3. PREDICTABILITY OF CAPITAL OPERATION RISK
Although capital operating risks are variable highly contingent and uncertain, capital
operating risks can also be identified and predicted. While the occurrence of a single
risk may be contingent and uncertain, the occurrence of a large number of risks is
inevitable. In fact, the occurrence of risks before, during, and after the operation of
capital operation will have certain characteristics. As long as the capital operation
subject of the enterprise can capture such information, it is possible to detect the risks
in time and prevent and avoid them early through prediction and analysis. However, in
order to accurately anticipate risks and take effective preventive measures, it is
necessary for capital operators to have risk awareness and accumulate experience in
identifying and preventing risks.
Thus, it is very necessary to study and explore the risk of capital operation. In the
new development period, enterprises should establish capital operation risk early
warning mechanisms based on national and industry norms, based on laws and
regulations, based on their own capital operation needs, etc., and use the early
warning mechanism to timely discover, identify prevent, and control all kinds of risks in
the process of capital operation, so as to fundamentally improve the security of capital
operation. However, the current era is the information age, and big data technology
has become the core weapon of each enterprise, which will play an important role in
the transformation and development of enterprises and risk prevention and control.
Therefore, enterprises wanting to conduct risk prediction and analysis of capital
operations need to rely on big data technology.
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
233
3. RISK PREDICTION UNDER BIG DATA TECHNOLOGY
Big data technology refers to a new type of information processing technology in
which people and objects upload data between them through a third-party medium,
the computer, and the computer categorizes, fuses, and processes the data uploaded
into the network. The strategic significance of Big Data technology does not lie in the
mastery of huge data information but in the specialized processing of these data
containing meaning [17-19]. In other words, if big data is compared to industry, the
key to the profitability of this industry lies in improving the processing capability of the
data and realizing the value-added and prediction of the data through processing.
3.1. DATA ACQUISITION
Data acquisition is data mining, i.e., extracting high-value data information from
inside massive data resources, and is an important method used to obtain association
rule attributes to filter data. Data acquisition belongs to a decision support process,
mainly based on artificial intelligence, machine learning, and pattern recognition, and
can also interact with users or knowledge bases. The mining object is also not limited
to a certain type of data source but can be a relational database, data warehouse,
text, multimedia data, and other data sources containing semi-structured data or even
heterogeneous data [20]. More common is the web crawler technique.
Web crawler technology is a technology based on the Internet that automatically
crawls a specific web page. Its implementation mechanism is similar to the human
click operation on web pages, and it can complete the interaction between the client
and the server without human intervention to achieve automatic, accurate, and large-
scale extraction of web data. According to the different crawling tasks, web crawlers
can be classified into various types such as general-purpose, focused, priority,
incremental, deep, etc. Meanwhile, users can also build custom web crawlers
according to their actual needs.
Web crawler technology is used to collect information related to enterprise capital
operation risk, the specific steps are as follows:
Firstly, collect data requirements according to predefined. Establish the data source
website with the name of enterprise capital operation risk as the index. Use web
crawler technology to crawl enterprise basic data and related information, such as
national enterprise credit system, judicial system, Tian-eye search, enterprise search,
Qixinbao, etc., focusing on asset data, trademark data, public litigation data, public
opinion data and deep mining and crawling of enterprise relationship.
Next, the seed initialization crawler is constructed. Using the name data of existing
corporate capital operation risks, we construct the initialization crawler based on the
characteristics of each website. Then, we obtain the source documents of the web
pages. The source document of the web page is parsed and the required text content
is stored in the database, or the required data is extracted and put into the queue to
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
234
be crawled and then entered into the cycle of crawling. Finally, the data obtained by
the crawler is stored in the database.
3.2. DATA STORAGE
At this stage, the most commonly used system for data storage is a distributed
electromechanical system. Distributed electromechanical systems can store massive
amounts of data on multiple spatial and temporal scales. However, as the service time
of the equipment becomes longer, the amount of data for remote monitoring of
distributed electromechanical systems grows exponentially. At this time, the use of
distributed storage systems for data storage may suffer from load-balancing
imbalance. The use of a hashing algorithm can achieve distributed data storage with
minimal and stable system changes, as shown in Figure 1.
Figure 1. Distributed storage under hash algorithm
As can be seen in Figure 1, using hashing algorithms to store enterprise capital
operation risk data, the entire risk data storage space can be abstracted as a ring of
fixed length, and then storage nodes are assigned to this ring. In this way, the nodes
on the ring all have a fixed hash value, and this ring is called a hash ring. The same
hashing algorithm is used to find out the hash value of the keys of the stored data and
they are mapped on the same hash ring as well. Finally, the storage node is found
clockwise from the position of the data mapping, and the data is stored on the first
found storage node. In this way, the enterprise capital operation data information
becomes a unique and extremely compact hash value of the data, which facilitates the
storage of information.
3.3. ASSOCIATION RULE PREDICTION ANALYSIS
In order to efficiently predict the risk of capital operations, we need to process and
analyze the large amount of data generated during capital operations. Eighty percent
of the data generated during capital operations is unstructured and grows
exponentially by 60% every year. This corresponds to the data processed by big data
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
235
technology. There are three main categories of data processed by Big Data
technology, namely structured data, semi-structured data, and unstructured data, and
unstructured data is becoming a major part of data.
In big data processing and analysis, correlation analysis is one of the simplest and
most practical analysis techniques, which can be used to better handle unstructured
data and generate frequent patterns and correlation models at the same time. The so-
called association reflects that an event is dependent or related to other events to
some extent and can be predicted according to the relevant rules. Association rules
are a widely used pattern recognition method, which can be applied to enterprise
capital operation risk prediction to effectively identify the risk factors involved.
Let the possible risks in the capital operation of an enterprise be set, and each risk
in it can be regarded as a subset. Each risk subset and the whole risk set are logically
implicitly related. If the probability of two risk subsets appearing simultaneously in the
whole association rule is small, it is proved that the relationship between the two risks
themselves is not significant. If the probability of the simultaneous occurrence of two
risk subsets is very frequent, it indicates that the two risk subsets are related to each
other, and this probability of simultaneous occurrence can also be called support. The
probability of two risk subsets occurring simultaneously is the confidence level, and
when the confidence level is 100%, then the two risk subsets are proved to be
relational and intimate. When one of the risk subsets appears, the other risk subset
also appears in a bundle.
The risk dataset stored by the hashing algorithm is used as input data, and the
frequent item set is obtained by setting the minimum support, and then the next
process proceeds. According to the confidence threshold, the strong association rules
that meet the requirements are inferred from the results generated in the previous
step and are aggregated and verified, and the whole mining process is finished. In the
process, we can set different parameters to guide the mining process according to the
actual needs, and the final results of risk prediction analysis factors are derived by
continuously changing the values of both.
3.4. PREDICTIVE ANALYTICS BUILDING PLATFORM
Using the web crawler technology, hash algorithm, and association rule analysis in
big data technology, we can achieve optimization in the accuracy and analysis speed
of risk prediction, and better improve the process and results of capital operation risk
prediction analysis. Accordingly, this paper builds an enterprise capital operation risk
prediction and analysis platform based on the above algorithms, as shown in Figure 2.
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
236
Figure 2. Enterprise capital operation risk prediction and analysis platform
As can be seen from Figure 2, the built enterprise capital operation risk prediction
and analysis platform includes three modules: data interface layer, data integration
layer, and data distribution layer. The data collected in the major platforms by relying
on web crawler technology are input to the data interface layer and feature extraction
of the data to extract high-risk data. Through the interface with the data integration
layer, the high-risk data is connected to the data integration layer. In the data
integration layer, the hash value of the keys of the stored data is derived by the
hashing algorithm and mapped on the same hash ring. Finally, the storage node is
looked up clockwise from the location of the data mapping, and the data is stored on
the first found storage node. The data and hash ring are uniformly encoded to achieve
distributed high-speed storage and flexible management of data. Finally, the stored
data are analyzed by correlation rules to derive the most likely operational risks of the
final enterprise when conducting capital operations. The platform has powerful data
storage and processing capabilities in all aspects, which can effectively alleviate the
problem of information asymmetry.
4. APPLICATION OF ENTERPRISE CAPITAL
OPERATION RISK PREDICTION ANALYSIS
4.1. INCREASED ACCURACY OF PREDICTIVE ANALYSIS
The most fundamental purpose of applying big data technology to enterprise capital
operation risk prediction and analysis is to improve the accuracy of enterprise capital
operation risk prediction and analysis. Accordingly, in this paper, the platform was put
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
237
into use in 10 companies in a city, denoted by letters A, B, C, D, E, F, G, H, I and J, to
compare the predictive analysis accuracy of the 10 companies before and after using
the platform. The results obtained under the same conditions are shown in Figure 3.
Figure 3. Comparison of accuracy rates of sample companies
As can be seen in Figure 3, before the use of big data technology, the accuracy of
capital operation risk prediction analysis of Company A was only 53%, and after the
use of Company A, the accuracy of capital operation risk prediction of Company A has
increased by 40% to 93%. Company B's capital operation risk prediction accuracy
increased from 61% to 91%, an increase of 31%. Both Company C's risk forecast
accuracy and Company G's risk forecast accuracy peaked at 97%, an increase of
38% and 41%, respectively, from the previous levels. Before using the built platform in
this paper, the risk prediction analysis accuracy of Company D was only 64%, while
after using the built platform, the accuracy rate was 94%. Before applying the built
platform to the whole process of capital operation, Company E had the highest
accuracy rate of 70% in the risk prediction analysis of enterprise capital operation.
Company F's risk prediction accuracy rate also improved significantly, lower than
other companies, but also increased by 24% compared with that before. Company H
and Company I improved their risk forecasting accuracy by 30% and 24%,
respectively, compared to their pre-platform performance. Company J had the lowest
risk prediction accuracy of 55%, but after applying the built platform, it improved to
95%, ranking third.
It can be seen that the introduction of big data technology in enterprise capital
operation risk prediction and analysis has achieved a significant increase in the
accuracy of enterprise capital operation risk prediction and analysis, which can help
enterprises reduce unnecessary waste of resources and recover a lot of unnecessary
losses.
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
238
4.2. INCREASED SPEED OF PREDICTIVE ANALYSIS
Enterprise capital operation risks often occur in a split second, while the losses
caused by the risks are infinite. The improvement of the speed of risk prediction and
analysis is also based on big data technology in building prediction and analysis
platforms needs to be considered. The speed of analysis should not decrease with the
increase in the amount of information such as data. To verify the analysis speed of the
platform built in this paper, 3,000 data, and 6,000 data were input into the platform to
derive the platform risk prediction time and compare it with the time required before
the enterprise uses the built platform, and the results are shown in Figure 4.
Figure 4. Time required to analyze the predicted risk of corporate capital operations
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
239
As you can see in Figure 4(a), before using the platform, it took 50 minutes for
companies to identify capital operation risks when analyzing data of up to 300.
However, after using the platform, it only takes 25 minutes to identify potential capital
operation risks, which is one-half of the time saved. When the data reached 600, the
built platform took only 24 minutes to identify risks, a reduction of 20 minutes from the
previous 44 minutes. When analyzing 900 pieces of data, it took 45 minutes to identify
risks before the enterprise used the platform, while after the application, risk discovery
took less than half the time. When the data was 1200, the time for risk discovery was
reduced by 21 minutes compared to the original. When the number of data is 1500,
the risk discovery time is reduced by 20 minutes. When the data reached 1,800, the
built platform took only 28 minutes to discover the risk, a reduction of 15 minutes from
the previous 43 minutes. As the data increased, the platform still took less time to
discover risks than it did before use. When the data was 3000, the time for risk
discovery was reduced by 15 minutes from the previous 42 minutes to only 27
minutes.
As can be seen in Figure 4(b), the risk analysis speed of the platform built in this
paper remains at a high level as the data volume increases. When the data volume
reaches 3300, the time taken by the built platform drops by 28 minutes. When the
data volume was 3600, the time for the platform to discover risks was 21 minutes,
which was 30 minutes less than before using the platform. Before using the platform,
it took 55 minutes to discover capital operation risks when analyzing 3900 data.
However, using the platform, companies can identify potential risks to capital
operations in just 20 minutes, saving nearly one-third of the time previously. When
analyzing 4,200 pieces of data, it took 57 minutes to identify the risk, but with the
platform, it took only half the time to identify the risk with 5 minutes remaining. When
the data reaches 4,500 to 5,700, the platform takes up to 27 minutes to identify
potential risks, compared to a minimum of 49 minutes before use, a reduction of 22
minutes. When the data volume is as high as 6,000, the platform takes only 22 hours
to discover risks, nearly a quarter of the time it took for the original enterprise.
By comparing the time spent on the same amount of data, it can be found that the
time required for the capital operation risk prediction and analysis platform based on
big data technology is controlled within 30 minutes, with a minimum of 20 minutes.
With the increase of data and other information, the prediction and analysis accuracy
rate can still be maintained at a high level. Thus, the application of big data
technology to the prediction and analysis of enterprise capital operation risk is helpful
to help enterprises quickly identify the existence of risks, and then take timely
measures before the occurrence of risks, so that losses can be controlled within
affordable limits.
5. CONCLUSION
In order to promote the benign development of enterprise capital operation, this
paper uses web crawling technology to index the enterprise capital operation risk and
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
240
then obtain the required data through cyclic crawling. Using a hashing algorithm, the
relevant numbers of the massive enterprise capital operation risks are compressed
through constant mapping, so that the massive data can be stored effectively. The
frequent risk items are identified through association rules to derive the final capital
operation risk prediction analysis data. Finally, an enterprise capital operation risk
prediction and analysis platform is built based on the above big data technology. The
accuracy of the built predictive analytics platform is 97%, and the risk prediction time
is as low as 20 minutes and does not decrease as the amount of information
increases. From the accuracy and speed of risk prediction analysis, it can be seen
that the enterprise capital operation risk prediction analysis can be well achieved by
relying on big data technology.
REFERENCES
(1) Прешкин, Г., Мехренцев, А
., et al. (2019). Formation of forest management in
conditions of digital market economy development. Forestry Engineering
Journal, 9(1), 211-223.
(2)
Abramov, R. A. (2016). To the problem of evaluation of the role of state in the
process of market economy development. (páginas 59-63).
(3) Zheng, W., Ma, K., & Wang, X. (2015). TE-Shave: Reducing Data Center Capital
and Operating Expenses with Thermal Energy Storage. IEEE Transactions on
Computers, 64(11), 3278-3292.
(4)
Xiao, J. H., Lee, S. J., Liu, B. L., et al. (2018). Capital Operations in China's
Logistics Industry. En Contemporary Logistics in China (Capítulo 8, páginas
161-180). doi: 10.1007/978-981-13-0071-4
(5)
Passador, M. L. (2015). Crowdfunding: among Suitability and Appropriateness
Rules and its Applicability to Capital Increase Operations. Banca Impresa
Società, (2), 287-332.
(6)
Woodroof, E. A., Turner, W. C., Heffington, W., et al. (2021). Energy
Conservation Also Yields: Capital, Operations, Recognition and Environmental
Benefits (páginas 169-187).
(7) Ferris, S. P., Javakhadze, D., & Rajkovic, T. (2017). CEO social capital, risk-
taking and corporate policies. Journal of Corporate Finance. doi: 10.1016/
j.jcorpfin.2017.09.004
(8)
Cesarone, F., Colucci, et al. (2018). Minimum risk versus capital and risk
diversification strategies for portfolio construction. Journal of the Operational
Research Society, 1-18.
(9)
Tang, C., Yi, Y., Yang, Z., et al. (2016). Risk forecasting of pollution accidents
based on an integrated Bayesian Network and water quality model for the South
to North Water Transfer Project. Ecological Engineering, 96, 109-116.
(10)
Salling, K. B., & Leleur, S. (2015). Transport Project Evaluation: Feasibility Risk
Assessment and Scenario Forecasting. Transport, 32(2), 180-191.
(11)
Dotan, A. (2018). Empathy And Co-Creation In Capital Markets Operations –
Insights From The Field. Journal of Financial Transformation, 48, 1-12.
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
241
(12)
Iwamura, K., & Liu, B. (2017). Dependent-chance integer programming applied
to capital budgeting. Journal of the Operations Research Society of Japan,
42(2), 117-127.
(13)
Zhan, J. (2021). A financing model with rebate contract in a capital-constrained
supply chain. RAIRO - Operations Research, 55(1), S811-S836.
(14)
Shenoy, R., & Haley, K. (2020). Improving Operations Within Capital Project
Execution with Case Examples. In Abu Dhabi International Petroleum Exhibition
& Conference, 202034.
(15)
Zhuang, X., Tian, Y., & Zheng, Y. (2017). Dual-channel supply chain operations
with working capital constraint: constant vs increasing marginal costs. In
International Conference on Economic & Business Management, 215-225.
(16)
Ikechukwu, N. C., & Anyanwokoro, M. (2019). Investigating the Impact of the
Capital Market Operations on a Developing Economy: The Nigerian Experience
(1983 – 2016). South Asian Journal of Social Studies and Economics, 1-13.
(17)
Weil-Dubuc, P.-L. (2018). Big Data Individualisation des risques Individualisation
of risks Médecine predictive Predictive medicine Solidarity Solidarité. Revue
d'epidemiologie et de sante publique, PP S19-S23.
(18)
Rogers, M. A., Elena, et al. (2018). Cardiovascular calcification: artificial
intelligence and big data accelerate mechanistic discovery. Nature reviews.
Cardiology, 261-274.
(19)
Li, F., & Abo Keir Mohammed Yousuf. (2021). Mathematical model of back
propagation for stock price forecasting. Applied Mathematics and Nonlinear
Sciences. doi:10.2478/AMNS.2021.1.00089.
(20)
Rumsfeld, J. S., Joynt, K. E., & Maddox, T. M. (2016). Big data analytics to
improve cardiovascular care: promise and challenges. Nature Reviews
Cardiology, 13(6), 350-359.
https://doi.org/10.17993/3ctic.2023.122.227-242
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed.43 | Iss.12 | N.2 April - June 2023
242
