A Study on the Application of Big Data in Credit Risk Management

Shuning Cao

School of Financial Science and Technology, Shenzhen University, Shenzhen, Guangdong, 518000, China

Keywords:

Big Data, Credit Risk Management, Commercial Banks, MSMEs.

Abstract: Credit in the modern economy has a vital role. In today's increasingly complex financial markets, the problems

of traditional manual credit scoring in credit risk assessment have become more prominent. The emergence

of big data technology, on the other hand, has enabled some financial institutions to discover its potential

value. This paper analyzes the value of applying big data in credit risk management and the challenges that

need to be faced. Therefore, this paper aims to analyze the value of the application of big data in credit risk

management and the challenges that need to be faced. For current credit risk problems, big data technology

can produce corresponding countermeasures to effectively reduce credit risk. At the same time, the application

of big data in credit risk management is also characterized by the risks and challenges of privacy leakage, lack

of talent, and dominance of traditional credit thinking. Comprehensively speaking, there is great value in

applying big data to credit risk management. The technology can provide effective solutions to the current

problems, and it can also provide some reference for other risk management research.

1 INTRODUCTION

Credit is an important part of the modern economy,

providing a partial source of funding for individuals

as well as businesses. With the increasing complexity

of the global financial market and the gradual rise of

Internet finance, some credit risk management

challenges have arisen. Traditional credit risk

management methods often rely on manual credit

scoring, so it is difficult to capture changes in

information and prone to subjective bias accurately.

With the gradual transition of credit business to an

online model, financial institutions face new

challenges and opportunities.The institutional reform

of the State Council and the establishment of the

National Digital Agency have made digitalization a

central force driving economic growth. As the core

technology of this change, big data is gaining

attention for its potential value for credit risk

management. The use of big data for credit risk

management is of great practical significance. Big

data can provide richer data samples and more precise

model analysis, thus improving the accuracy and

reliability of risk assessment and prediction (Zhong,

2024). At the same time, big data can optimize the

https://orcid.org/0009-0002-3314-3772

credit decision-making process through the

comprehensive analysis of multi-dimensional

information on borrowers, providing financial

institutions with a more scientific and objective basis

for credit decision-making. Big data can also track

credit evaluation dynamically regularly to ensure the

continuity, accuracy, and timeliness of project

database updates (Lin, 2024).

Currently, relevant researchers have analyzed

the current status of the application of big data in

credit risk management. For example, Tingting Wang

(Wang, 2024) summarized the credit business and

risk management innovation of commercial banks in

the context of big data. In addition, Yu Hongfei (Yu,

2023) summarized the research on credit risk

assessment and control of micro, small and medium

enterprises (MSMEs) in the big data environment.

However, it is necessary to summarize and review the

existing studies again and look forward to the future

research direction due to the following three reasons.

First, the rapid development of big data technology

and the proliferation of related studies require

constant updating of the review. Second, most of the

existing reviews focus on specific areas or issues and

lack a comprehensive analysis of overall credit risk

management. Finally, with the advancement of digital

Cao, S.

A Study on the Application of Big Data in Credit Risk Management.

DOI: 10.5220/0013205300004568

In Proceedings of the 1st International Conference on E-commerce and Artiﬁcial Intelligence (ECAI 2024), pages 19-26

ISBN: 978-989-758-726-9

transformation, enterprises need to build a more

comprehensive digital system in credit risk

management.

This paper aims to analyze the application of big

data in credit risk management. The paper is

developed along the following lines. Firstly, this

paper will give an overview of big data technology.

Then it will sort out the current problems of big data

in credit risk management in the literature. Further,

the value of big data in credit risk management is

analyzed. In addition, it analyzes the main challenges

of big data in credit risk management, such as privacy

protection and other issues. Finally, summarize the

whole paper and look forward to the research

prospect and direction.

2 OVERVIEW OF BIG DATA

TECHNOLOGY

DEVELOPMENT

With the increasing complexity of global financial

markets and the deepening of China's digital

economy, financial institutions face a number of

credit risk management challenges. Traditional credit

risk assessment methods often rely on limited data

and expert experience. The data sources mainly rely

on limited information, such as financial statements

and credit reports provided by borrowers, so it is

difficult to accurately capture the impact of market

fluctuations, changes in the economic environment,

and changes in the behavior of individual borrowers

on the risk of credit defaults, and at the same time,

because it is an artificial score, it is susceptible to

subjective bias, which all increase the credit risk.

These challenges provide new opportunities for

applying big data technology in credit risk

management. Due to the massive, diverse, and fast

nature of big data can provide more comprehensive

and dynamic data support to help financial

institutions identify and manage credit risk more

effectively (Wang, 2024). The value density of big

data is low, and generally valuable data accounts for

no more than 10% of all data. Finally, big data

technology has efficient processing capabilities and

can quickly mine high-quality information from

massive amounts of data.

With the continuous development of big data

technology, financial institutions have begun to

realize the huge potential for big data in credit risk

management. Big data has features that make it

capable of integrating and processing data from

multiple channels, including but not limited to

comprehensively capturing all types of information

about borrowers, including transaction records, social

media behavior, consumption behaviour, credit

history and other multi-dimensional data (Chen,

2022). This data provides financial institutions with

unprecedented insights that enable more

comprehensive, in-depth and accurate credit risk

assessment.

In addition, with the rise of Internet finance, credit

business has gradually shifted from offline to online,

with a broader customer base and more diversified

credit needs. Financial institutions must have more

flexible and efficient credit default risk control

mechanisms to cope with the complex and changing

credit market environment. The application of big

data technology provides financial institutions with

such a possibility. Financial institutions can quickly

identify potential risks and realize precise prevention

and control of credit default risks by monitoring,

analyzing and adjusting massive data in real time.

Therefore, the study of big data in credit default risk

control is of great significance.

The comparative differences between manual

credit scoring and big data techniques are noted in

Table 1. The table focuses on the comparison in terms

of data sources, processing power, accuracy,

efficiency and timeliness.

Table 1: Comparison of manual credit scoring and big data techniques

Data sources Processing

capabilit

Accuracy Efficiency Timeliness

Manual

credit

scoring

Limited data

sources

Weak

processing

capacity

Accuracy is

subjective

Less efficient Limited ability

to monitor data

and lag in

adjustments

Big data

technology

Wide range of

data sources

High processing

power

High accuracy High

efficiency

Real-time

monitoring,

dynamic

ustment

ECAI 2024 - International Conference on E-commerce and Artiﬁcial Intelligence

3 PROBLEMS IN CREDIT RISK

MANAGEMENT

Credit risk refers to the possibility that the borrower

cannot repay the loan's principal and interest on time

due to various reasons in the course of the loan, which

leads to the creditor (usually the bank or other

financial institutions) to suffer losses. The objects

facing credit risk mainly include borrowers and

creditors. Borrowers include MSMEs, large

enterprises and individuals. Creditors are mainly

financial institutions. Large enterprises are generally

not exposed to credit risk because of their strong

economic strength and ability to repay debts.

Individuals contain diverse and complex groups that

need to be treated appropriately and will not be

discussed here. And MSMEs face credit risk because

of their small production scale, illiquid capital and

other problems. So, this paper will discuss the credit

risk problems faced by MSMEs as representatives of

borrowers. Meanwhile, the representative of financial

institutions is commercial banks, and this paper will

discuss commercial banks as the representative of

creditors.

3.1 Main Credit Risks of Commercial

Banks

Under the traditional credit risk management system,

commercial banks mainly face the problems of

insufficient credit system, difficulty obtaining

comprehensive, accurate and real customer

information and difficulty in post-loan risk control.

3.1.1 The Credit System Is not yet Perfect

In credit risk management, having a perfect credit risk

system is key to reducing credit risk. The credit

business development of China's commercial banks

started late, and the immaturity of the credit market

led to the difficulty of managers to adapt to market

changes, and the credit risk management system

needs to be improved (Du, 2021). As a result, some

problems have arisen. Due to the lack of exhaustive

customer information checking, the credit credit

system can not give full play to its role, which makes

it difficult for banks to assess customer risk

accurately (Yin, 2024). Compared with developed

countries with hundreds of years of management

experience, there are still obvious deficiencies in the

setup of credit market structure and the usefulness of

information management platform in China. This

affects the efficiency of credit risk management and

increases the possibility of credit default (Huang,

2023). At the same time, there are a series of problems

in the regulatory system, such as the regulatory

content not being detailed, the regulatory power not

being clear, the regulatory efforts not being strong

enough, and the regulatory duties not being strictly

fulfilled.

3.1.2 Difficulty in Obtaining

Comprehensive, Accurate and

Truthful Customer Information

The main risk in credit business originates from the

repayment ability of customers. Therefore, in the

process of credit business, banks need to check and

record customer information in detail to reduce the

repayment risk (Yin, 2024). Chinese commercial

banks have a single channel for obtaining data and

information, the cost of obtaining them is high, and

the data are not updated in a timely manner. And

Chinese banks are easily affected by changes in the

external market environment in the process of credit

management (Hu, 2023). The backwardness of the

data information acquisition and processing mode has

led to the inability of commercial banks to accurately

and timely identify external market risks, and the

failure to take timely and reasonable interventions has

resulted in interest rate losses, affecting profitability.

At the same time, the indicators for obtaining

information under the traditional credit risk

management system are relatively single, including

credit, records of violations, economic disputes and

so on. For enterprises and individuals associated with

the financial / material / human data can not be

obtained, can not be cross-validation and correlation

analysis of all aspects of information, resulting in the

credit front-line staff and credit approving officers to

determine the strength of the customer's repayment

ability, it is difficult to pass the data and other

objective records, the determination is more

subjective (Huang, 2023). At the same time, the

relevant information is also easy to tamper with, so

the data that can not pass the approval can be

approved, which leads to the bank is difficult to grasp

the customer's real business and financial situation,

and can not accurately measure the credit risk of the

lender.

A Study on the Application of Big Data in Credit Risk Management

3.1.3 Difficulty in Post-Loan Risk Control

According to the field research, the research bank

after the loan carries out risk control mainly through

the phone, door to door, contact with customer units

and other ways to collect, at the same time to

encourage the borrower through the monthly,

quarterly, semi-annual and other ways to settle the

interest rate, the use of risk clues to monitor the

system of regular decentralization of suspicious loans

(Xue, 2023). Numerous commercial banks, however,

have shortcomings in financial data monitoring,

making it difficult to capture and follow up on key

data information quickly. This limits the ability of

commercial banks to predict risks in a timely manner,

take stop-loss measures, or develop reasonable

response strategies. Although China's commercial

banks have set up a post-loan risk warning

mechanism, by technology, data and other factors

constraints, only based on loan overdue, deposit and

loan ratios and other direct indicators to set early

warning rules, early warning matters are not

comprehensive enough (Huang, 2023), or failed to

take the results of post-loan risk control.

3.2 The Main Credit Risks of MSMEs

Under the traditional credit risk management system,

MSMEs are mainly faced with the problems of low

information transparency, high financial management

risk and high market competition credit risk. These

problems affect the quality of assets affecting

financial institutions, the development of MSMEs

and the stability of the overall financial market.

3.2.1 Low Information Transparency

Low information transparency is a significant

disadvantage for MSMEs in the financial market.

Specifically, due to their small size and limited

resources, MSMEs often find it difficult to establish

a sound financial management system and

information system, leading to deficiencies or lags in

the collection, organization and disclosure of

financial information and operational data. This

information opacity firstly affects financial

institutions' credit assessment of MSMEs. In the pre-

credit review stage, financial institutions need to

judge the repayment ability and operational stability

of MSMEs on the basis of sufficient and accurate

information. Still, the opacity of information makes

this process complicated and uncertain. Financial

institutions may make conservative credit decisions

because they are unable to fully understand the real

situation of MSMEs, thus limiting the financing space

for MSMEs. In addition, at the post-loan supervision

stage, information opacity also poses a challenge to

financial institutions. It is difficult for financial

institutions to monitor the operation status and capital

flow of MSMEs in real time, and discover potential

risk factors in time and take effective measures to

prevent and control them. This information opacity

makes credit institutions have blind spots in pre-credit

review and post-credit supervision, thus increasing

the riskiness of credit investment (Zhang, 2024).

3.2.2 High Risk of Financial Management

Cash flow stability is the key to the survival of

MSMEs, and its impact is extremely significant.

Insufficient monitoring of cash flow by MSMEs may

lead to the risk of capital chain breakage (Zhang,

2024). When the enterprise faces cash flow problems,

its daily operations will be seriously constrained,

resulting in a tense or even broken capital chain,

which not only puts the enterprise itself in a difficult

situation, making it difficult to fulfill its payment

obligations to suppliers and employee payroll but also

triggers credit defaults due to the inability to repay

bank loans on time, which in turn affects the banking

system that provides it with credit support, resulting

in a decline in the asset quality of the bank, which

may ultimately cause This could lead to a decline in

the quality of bank assets and ultimately lead to asset

losses, affecting financial stability. The credit review

and accurate risk assessment can effectively control

costs and correctly recognize the feasibility of return

on earnings, which is the basis of credit risk

management (Zhang, 2024). Financial information is

the core of credit assessment. However, the

asymmetry of financial information makes risk

assessment more complicated, and it is difficult for

credit institutions to grasp the true and accurate

condition of the enterprise accurately, thus increasing

the risk of default. In addition to financial

management risks, MSMEs face unique challenges in

market competition, which also have an impact on

credit risk.

3.2.3 High Credit Risk of Market

Competition

MSMEs themselves are small in production scale and

not strong enough in production strength (Yu, 2023).

In the fierce market competition, MSMEs often have

limited resources, which causes it to be more difficult

ECAI 2024 - International Conference on E-commerce and Artiﬁcial Intelligence

to obtain credit compared with large enterprises

(Zhang, 2024). Banks and other financial institutions

tend to be more cautious about the repayment ability

of MSMEs when assessing the risk of loans.However,

the financial needs of MSMEs also continue to

increase (Feng, 2023).The existence of these reasons

has led to the persistence of the problem of difficult

and expensive financing for MSMEs. Moreover,

MSMEs are often at the end of the supply chain, with

a high degree of dependence on upstream suppliers

and downstream buyers. Once a problem occurs in

one part of the supply chain, such as price increase by

upstream suppliers or default by downstream buyers,

it will significantly impact the production and

operation of MSMEs. This transmission effect of

supply chain risks makes MSMEs more vulnerable in

the face of market changes, and their credit risks are

further amplified.

4 THE VALUE OF BIG DATA FOR

CREDIT RISK MANAGEMENT

Big data used in credit risk management has certain

value. For example, big data can provide richer data

samples and more accurate modelling analysis,

making it easier and more comprehensive to obtain

data, and more scientific and objective to process the

data results. At the same time, big data has high

efficiency, which can quickly process data processing

data and efficiently find valuable data from massive

data. Big data also allows for regular dynamic

tracking of credit evaluations, ensuring that problems

can be identified and interventions taken promptly.

The paper will expand and analyze the value of big

data for commercial banks and MSMEs.

4.1 The Value of Big Data Used in

Credit Risk Management of

Commercial Banks

Applying big data to commercial banks' credit risk

management system has certain value. Firstly, it can

build a complete credit system and enhance the effect

of risk management. Secondly, it can effectively

mine, process and follow up the data information.

Finally, it can solve the problem of post-credit

management.

4.1.1 Constructing a Complete Credit

System and Enhancing the Effect of

Risk Management

For commercial banks, credit risk is the main risk

factor (Feng, 2023).Applying big data technology in

credit credit system is more conducive to building a

complete credit system and enhancing the effect of

credit risk management (Yin, 2024). Big data helps

build a credit credit system by analyzing a large

amount of data, identifying customer information and

repayment ability, to accurately judge their borrowing

qualifications and reduce credit risk. Big data can also

innovate the bank risk decision management mode. In

the actual development of risk management, banks

should first identify the risk, the impact of the risk,

and the risk level, which should be scientifically

divided, and do a good job of risk decision

management (Wang, 2024). At the same time, big

data can establish a more complete risk monitoring

mechanism, which has a positive role in risk

management innovation and internal risk

management of banks, is an important initiative for

the development and innovation of commercial bank

credit business (Yin, 2024).

4.1.2 Effectively Mining, Processing and

Following up Data Information

Big data can establish a data sharing platform and

break data silos. Banks can strengthen cooperation

among themselves to eliminate data barriers and

realize data sharing. The government can take the

lead in establishing a data sharing platform,

integrating credit data from government departments

and various credit bureaus, and granting and

regulating access to financial institutions (Xue,

2023). The sharing platform can improve the

efficiency of data mining, while making data

collection more comprehensive. And the

synchronization of data to regulatory agencies

strengthens the credibility and accuracy of data, thus

enhancing the accuracy of risk assessment.

Meanwhile, the correlation between the data can

enhance the power of risk prediction. Commercial

banks should make full use of the large amount of

customer information they have accumulated, and

classify and process it with the specific situation to

further explore the correlation between the sample

data and customer behavior, such as the connection

between the non-performing loan rate and the

borrower's investment behavior, the type of work, and

A Study on the Application of Big Data in Credit Risk Management

other factors, in order to comprehensively improve

the level of controlling credit risk (Xue, 2023).

Furthermore, the timeliness characteristics of big data

technology can deliver customer information to the

bank in real time, so that if the latest data show that

the customer's credit risk is too high, it can stop the

loss in time and reduce the credit risk.

4.1.3 Solve the Problem of Post-Loan

Management

China's banking industry in the loan supervision often

focusing on pre-loan and loan strict review, but the

post-loan management attention is insufficient,

mainly due to information asymmetry. After the

customer loan, the actual use of funds is difficult to

control, often deviating from the original direction,

such as business loans for investment, increasing the

risk of repayment, affecting the bank's wind control.

Big data can solve the problems of post-loan

management (Huang, 2023). It can track the flow of

funds in real time, provide timely warning of illegal

use, and reduce loan risks; at the same time, record

information about customers and their affiliates to

help approve new loans; and improve the efficiency

of interdepartmental communication to solve the time

gap and collaboration problems in traditional post-

loan management. In addition, big data technology

accumulates rich post-loan management data through

data traces, providing strong support for risk analysis

and customer resource mining.

4.2 The Value of Big Data Used in the

Credit Risk Management of

MSMEs

Applying big data to MSMEs' credit risk management

system has certain value. First, it can solve the

limitations brought by low information transparency.

Secondly, it can reduce the risks brought by

insufficient financial management. Finally, it can

enhance market competitiveness.

4.2.1 Addressing Constraints Brought

About by Low Information

Transparency

Restricted by their size and financial transparency,

these MSMEs often face heavy obstacles in obtaining

credit funds, which largely binds their development

potential. As a result, how to scientifically and

efficiently assess the credit risk of MSMEs has

become a common issue for both financial institutions

and MSMEs. In this context, the application of big

data technology shows irreplaceable value (Zhang,

2024). Traditional credit assessment favors financial

reports and ignores unstructured information such as

online activities, social media reputation, and

geographic location. Big data technologies, on the

other hand, excel at mining these pieces of Internet

information and applying advanced algorithms to

identify credit risk signals. For example, analyzing

online purchases to predict cash flow stability, or

social media public opinion to assess a company's

public image and potential risks. This provides

financial institutions with a comprehensive and multi-

dimensional view of credit assessment, helping them

identify risks and accurately make more informed

lending decisions. Even though information

transparency is low, big data can still mine

information from other sources to effectively address

its limitations.

4.2.2 Reducing the Risk of Inadequate

Financial Management

On the one hand, big data can solve the risks brought

about by financial information asymmetry. The big

data platform can promote data sharing and

integration among the government, financial

institutions, enterprises and other parties. By

integrating data from different sources, credit

institutions can gain a more comprehensive

understanding of the real situation of MSMEs and

reduce the risks brought by information asymmetry.

In addition to traditional financial data, big data

technology can also mine and analyze non-financial

information of MSMEs, such as online behavior and

social media reputation. This information helps credit

institutions to more comprehensively assess the

creditworthiness and operational capacity of

enterprises, thus making more informed lending

decisions. On the other hand, big data also monitors

cash flow in real time to prevent businesses from

getting into trouble and thus avoiding property

damage. Big data technology monitors key financial

data such as online transactions, bank account

changes, supply chain payments, etc. of MSMEs in

real time, and analyzes and predicts cash flow trends

through algorithms. This real-time monitoring helps

enterprises identify potential cash flow shortage risks

in a timely manner and take measures to address them

in advance. This can effectively solve the problems

caused by insufficient cash flow monitoring.

ECAI 2024 - International Conference on E-commerce and Artiﬁcial Intelligence

4.2.3 Enhancing Market Competitiveness

MSMEs can actively utilize the big data platform to

accumulate their own credit data. By timely and

accurately recording the enterprise's transaction

behavior, repayment records and other information, it

can gradually build up a good credit record and

improve its credit rating among financial institutions.

At the same time, MSMEs can establish a data

sharing mechanism with other enterprises in the

supply chain, and use big data technology to monitor

and analyze data changes in each supply chain link.

This helps enterprises identify potential risk points in

a timely manner and take appropriate measures to

prevent and respond to them. The big data-based risk

early warning system is also able to monitor

anomalies in the supply chain in real time, such as

price increases by upstream suppliers and defaults by

downstream buyers. When the system detects a

potential risk, it will immediately issue an early

warning signal to remind the enterprise to pay

attention and take corresponding measures to reduce

the impact of supply chain risks on the enterprise's

production and operation, thus increasing market

competitiveness.

5 CHALLENGES OF BIG DATA

IN CREDIT RISK

MANAGEMENT

In this era of rapid development of information

technology, the application of big data to credit risk

management system also has the risks and challenges

of privacy leakage, lack of talent traditional credit

thinking dominates.

Currently, in the traditional credit field, risky

cases of selling personal credit reports are common.

However, with the arrival of the Internet of Things

(IoT) era, any behavior may be collected and utilized

by big data (Luo, 2019)]. Many other associated data

will be involved once part of the data is leaked.

Moreover, big data is characterized by sea

quantization and huge amount of data, so it is

extremely difficult to preserve and protect it. For

these reasons, in the field of big data credit, it is easier

to obtain information through the Internet, and the

cost of crime is lower, which is more likely to

increase the chance of the risk of information leakage

(Luo, 2019)]. Privacy security is not negligible and

difficult to solve, which is undoubtedly a great

challenge for big data in credit risk management.

Applying big data to credit risk management also

has the problem of lack of talents. This is because

integrating credit risk control with big data requires a

group of composite talents familiar with credit

business, data mining technology and system

programming ability at the same time. However, at

present, talents with experience in both fields are

extremely scarce, resulting in financial institutions

facing technical bottlenecks and talent gaps in the

implementation of big data strategies. According to

the current view, there is a great lack of talents that

meet the needs in this area, resulting in commercial

banks and MSMEs being unable to scientifically and

effectively utilize big data technology, which is not

conducive to their development. On the other hand,

commercial banks and MSMEs also lack the

corresponding talent training mechanism. As a result,

the ability and quality of the employees of

commercial banks and MSMEs have not been

sufficiently improved, and it is difficult to effectively

integrate big data technology into credit risk

management, thus limiting the further development

and expansion in the field of credit business.

In the traditional credit approval mode of

commercial banks, human subjective judgment still

occupies a dominant position (Xue, 2023). The use of

big data technology is mostly limited to providing

auxiliary information for manual decision-making,

failing to give full play to its potential for risk

identification, so that the approval process lacks

fundamental changes compared with the past. In

addition, although the concept of comprehensive risk

management has been put forward, the

implementation of the bank's internal implementation

is not satisfactory, often on the surface, failed to

penetrate into the various levels and business

processes, resulting in the risk management of

grassroots employees with a weak sense of risk

management, and insufficient practical experience. In

terms of specific operation, when approving loans,

banks often tend to select key industries with high

maturity and development potential as well as

customers with rich experience in production and

operation in order to minimize credit risks. However,

this preference inevitably ignores those in the early

stages of growth or small-scale "long-tail customers",

whose financing needs are difficult to fully satisfy,

which limits the expansion of the bank's revenue scale

and the enhancement of credit risk control

capabilities. Therefore, realising more

A Study on the Application of Big Data in Credit Risk Management

comprehensive and accurate risk assessment with the

support of big data technology, while considering the

financing needs of long-tail customers, has become

an important issue for commercial banks to solve.

6 CONCLUSIONS

Credit is an important part of the modern economy

and is widely concerned. This paper analyzes the

study of big data in credit risk management. From it,

the paper can conclude that there are many problems

in traditional credit risk management, and the

application of big data in it has a chance to solve these

problems. However, applying big data also brings

some risks and challenges, and people urgently need

a corresponding policy system, technology and

relevant professional talents.

Existing studies have comprehensively and

deeply analyzed the opportunities and challenges of

big data in credit risk management, etc., and have

achieved rich research results, but there are still some

niche areas that need more exploration.

First, improving the accuracy of the big data

model is subject to more in-depth research. Although

the current credit risk assessment model has achieved

certain results, there is still room for improvement in

terms of prediction accuracy and stability. How to

further optimize the model algorithm to improve the

accuracy and timeliness of risk identification is an

area that requires in-depth study of big data in credit

risk management.

Secondly, with the continuous development of big

data technology, relevant laws and regulations are

also constantly improving. However, in the field of

credit risk management, balancing the contradiction

between data security and data utilization and

ensuring the effective implementation of laws and

regulations is still a topic that needs to be studied in

depth.

Third, with the development of financial

technology, new credit business scenarios continue to

emerge, such as Internet finance and supply chain

finance. These emerging business scenarios are

characterized by complexity and uncertainty, putting

forward higher requirements for credit risk

management. Utilizing big data technology to

effectively identify and manage risks in these

emerging business scenarios is another gap area that

requires in-depth research.

REFERENCES

Dai, Lei Y. H., Peng C., et al. Research on credit default

prediction method based on LightGBM and SHAP[J].

Journal of Shaoyang College (Natural Science Edition),

2024, 21(01):84-91.

Chen Y. J., Chen Y. M. Forecasting corporate credit ratings

using big data from social media[J]. Expert Systems

with Applications, 207(2022)118042.

Dong X. T. Defining characteristics of big data and its

application analysis[J]. Silicon Valley, 2013,

6(11):120+60.

Du G. S., Liu Z. X., Lu H. F. Application of innovative risk

early warning mode under big data technology in

Internet credit financial risk assessment[J]. Journal of

Computational and Applied Mathematics,

386(2021)113260.

Feng Q. Y. Supply chain financial credit evaluation

mechanism under the background of big data[J]. Optik,

272(2023)170266.

Huang J. Research on the application of big data in

commercial bank credit risk management[J]. Chinese

Science and Technology Journal Database (Full Text

Edition) Economic Management, 2023(1):0116-0119.

Hu Y. Z., Chen H. J. Research on Credit Risk of

Commercial Banks under the Background of Big

Data[J]. Journal of Hengyang Normal College, 2023,

44(02):99-107. DOI:10.13914/j.cnki.cn43-

1453/z.2023.02.021.

Lin Y. X. Building a dynamic risk evaluation system for

SME credit based on big data perspective[J]. Business

Economics, 2024, (01):188-192.

DOI:10.19905/j.cnki.syjj1982.2024.01.005.

Luo J. X., Feng Y. L. Thinking about the development and

security of China’s credit collection industry in the era

of big data[J]. Credit Collection, 2019, 37(06):27-33.

Wang T. T. Innovation of Credit Business and Risk

Management of Commercial Banks under the

Background of Big Data[J]. Old Brand Marketing,

2024, (07):85-87.

Yin S. T. Risk management innovation of commercial bank

credit business under the background of big data[J].

Market Outlook, 2024, (02):114-116.

Yu H. F. Research on Credit Risk Assessment and Control

of Micro, Small and Medium Enterprises under Big

Data Environment[J]. China Management

Informatization, 2023, 26(12):65-68.

Zhang H. Q. Research on Credit Risk Assessment and

Control of Micro and Small Enterprises under the

Background of Big Data[J]. China E-commerce,

2024(5):32-35.

Zhong M. Q. Application of big data in risk control

management of bank financial business[J]. China

Business, 2024, (03):170-171.

Xue K. Z., Tang Q., Zhu X. Y. Research on risk control of

commercial bank credit business under the perspective

of big data[J]. China Business Journal, 2023, (04):106-

108. DOI:10.19699/j.cnki.issn2096-0298.2023.04.106.

ECAI 2024 - International Conference on E-commerce and Artiﬁcial Intelligence