Popular financial service platform and method based on big data, artificial intelligence and RPA

Abstract

The invention relates to the technical field of intelligent financing, in particular to a general financial service platform and method based on big data, artificial intelligence and RPA, and the platform comprises a message push system, a data receiving and collection module, a data storage module, a pre-loan admission credit approval module, and a post-loan management module. The message pushing system is used for issuing credit product information and carrying out customer drainage and data interaction; the data receiving and collecting module is used for receiving a credit request of a customer, automatically inputting information of the customer and establishing a customer label; the data storage module is used for storing data information; the before-loan admission credit granting approval module is used for verifying credit materials provided by the customer, judging whether the customer meets a credit admission condition or not, and then performing risk assessment and limit calculation to complete customer credit granting approval; and the post-loan management module is used for performing post-loan management, and performing customer operation data tracking, customer interest settlement and repayment pre-reminding.

Description

Technical Field

[0001] The present invention relates to the fields of big data, artificial intelligence, RPA and smart financing technology, and in particular to an inclusive financial service platform and method based on big data, artificial intelligence and RPA. Background Art

[0002] In recent years, the government's financial support for small and micro enterprises has been increasing day by day, and various laws and policies on supporting the real economy have emerged one after another. The Central Economic Work Conference, the National Financial Work Conference and other conferences have clearly proposed to enhance the ability of financial services to serve the real economy and guide banking funds from virtual to real.

[0003] Small and micro enterprises have the characteristics of small scale and imperfect financial system. Traditional banks usually conduct in-depth investigations on loan enterprises through offline visits to obtain detailed basic information of enterprises, including business, finance, compliance, industry conditions and other information, in order to judge the repayment ability and willingness of enterprises. However, this due diligence method has high operating costs, long approval cycle, and it is difficult to guarantee the authenticity and validity of the information obtained. Compared with the loan amount of small and micro enterprises, it is not cost-effective. On the other hand, small and micro enterprises often lack traceable information and are difficult to effectively assess risks. Small and micro enterprises generally lack collateral, have imperfect management, irregular financial statements, and lack effective evaluation or proof of their credit level, making it difficult to cover the long tail group of small and micro enterprises.

[0004] In recent years, in order to solve the problem of information asymmetry of small and micro enterprises, my country has accelerated the construction of various credit reporting systems, but the overall development is still in the early stages. For example, the data is not unified, and the information is found in the industrial and commercial, tax and other systems. The credit reporting system lacks effective sharing and unification and is not dynamically updated. Most of the default risk assessments of small and micro enterprise customers by banks are still based on static data such as past credit data and transaction data. The biggest disadvantage of this method is the lack of foresight. Because the important factors affecting corporate defaults are not only the historical credit situation of the enterprise, but also the overall development status of the industry and the real-time operating conditions. The intervention of big data means makes credit risk assessment closer to the facts. The integration of internal and external data resources is the premise of big data credit risk assessment. Generally speaking, in the process of identifying customer needs, estimating customer value, judging customer pros and cons, and predicting the possibility of customer default, banks need to rely on both customer-related information already held within the bank and the credit information of the People's Bank of China, customer public evaluation information, business operation information, income and expenditure information, social connection information, etc. held by external institutions.

[0005] In addition, the bank's loan business is managed by the funding department of the financial shared center, which is usually participated by multiple departments such as branches, sub-branches, credit departments, loan review departments, risk assessment departments, and grant centers. The business involved mainly includes qualification rating, credit investigation, document review, and loan issuance, etc. Various audits and approvals are being carried out in many links, which is prone to duplication of links. The amount of material data involved is large, manual processing is inefficient, and it is very easy to make mistakes. Summary of the invention

[0006] In view of the shortcomings of the above-mentioned prior art, this application provides an inclusive financial service platform based on big data, artificial intelligence and RPA, which takes small and micro enterprises, "agriculture, rural areas and farmers" and other key areas and weak links as the entry point, is driven by business scenarios, and uses intelligent means to empower all aspects of inclusive financial business, and build a "data + scenario + finance" trinity inclusive financial digital solution. Relying on big data and artificial intelligence, with the help of Internet channels, integrating tax, provident fund, real estate and other information, giving full play to the advantages of information sharing, legally and compliantly using internal and external data information, enriching the information dimensions of pre-loan access and post-loan risk control management, helping banks and other financial institutions to conduct panoramic portraits of corporate and personal credit, accurately focusing on high-quality target customers, and improving the comprehensiveness and accuracy of risk assessment of banks and other financial institutions, effectively alleviating the financing process difficulties, low financing efficiency, and no collateral guarantees of small and micro enterprises and downstream dealers of core enterprises. At the same time, RPA technology (robotic process automation technology) and OCR technology (optical character recognition technology) are used for operation-intensive work in inclusive services, such as corporate financial statement entry, invoice verification, etc., to reduce the desk work of account managers and improve work efficiency and accuracy. The specific plan is as follows: The inclusive financial service platform based on big data, artificial intelligence and RPA provided by the present invention is a full-process service platform. The inclusive financial service platform covers all aspects of inclusive services, and provides one-stop online services such as "customer flow, data interaction, risk control decision-making, credit application, online withdrawal, and post-loan management", achieving full coverage of solutions from pre-loan, mid-loan, and post-loan; the inclusive financial service platform includes a message push system, a data receiving and collection module, a data storage module, a pre-loan access credit approval module, and a post-loan management module; The message push system is used to release credit product information of banks and other financial institutions, conduct customer flow and data exchange, and release credit product information of banks and other financial institutions through self-built channels such as bank WeChat public accounts, H5 bank web pages, online banking systems, mobile banking, mini programs, third-party platforms, tax bureau tax payment platforms, and other third-party channels, so that enterprises or individuals can inquire about credit products that meet their needs, and customers with credit needs can be directed; The data receiving and collecting module is used to receive credit requests from enterprises or individuals, and automatically enter the basic information of customers (i.e., enterprises or individuals that have issued credit requests) through AI recognition, OCR and other technologies. After receiving full authorization from the customer, the module obtains the customer's recent data from the tax bureau, provident fund and other platforms. The customer's recent data to be obtained includes customer tax data, invoicing data, government big data, credit data of corporate legal persons and legal persons, stock business data including intra-bank settlement data, third-party anti-fraud data and other data information. The collected scattered customer recent data information is integrated, duplicate and invalid data is removed, and stored in the data storage module for centralized management and subsequent analysis; customer tags including customer business information and customer credit information are established based on the integrated customer recent data; The data storage module is used to store customer tags, basic information of corresponding customers, integrated recent customer data, and relevant analysis data generated during the credit approval process; The pre-loan access credit approval module verifies the credit materials provided by the customer based on OCR technology and RPA technology to determine whether the customer meets the credit access conditions; conducts risk assessment and credit limit calculation on the credit materials provided by the customer and the integrated recent customer data through the approval model, automatically completes the customer credit approval, obtains the approval result and credit limit of the corresponding customer, and uploads the approval result and credit limit of the corresponding customer to the data storage module; the approval model includes the access model, anti-fraud model, approval decision model, credit limit approval model, and risk pricing model; The post-loan management module is used for post-loan management. It conducts process-based and systematic tracking of customer withdrawals, interest payments, and repayments through real-time monitoring of tax data and monthly operating data; it uses multi-dimensional data to solve financing difficulties and conduct risk control analysis on customer operating data, and tracks customer operating data according to the post-loan management regulations of financial institutions, and provides advance reminders of customer interest payments and repayments.

[0007] Furthermore, the verification of credit materials based on OCR technology and RPA technology is specifically as follows: RPA technology is used to batch decompress the credit material compression packages provided by customers and pre-process the loan data to achieve automated pre-processing of loan materials, overcoming the problems of massive cross-domain data transmission and unknown data format processing; and OCR technology is used to identify the key information of the materials. After the identification is completed, the query interface of the financial institution's own management system is called to verify the identified key information, and the verification results and data integrity are summarized to generate an EXCEL file, which is notified to the business personnel of the financial institution by email and uploaded to the object storage module of the data storage module for subsequent inquiries by the business personnel.

[0008] Furthermore, the EXCEL file includes the following information: the loan note number, whether the materials required for the corresponding loan note are complete, and the names of the missing materials.

[0009] Furthermore, the strategy of using multi-dimensional data to solve the financing dilemma is specifically: It integrates tax, provident fund and real estate information, and uses multi-faceted data including enterprise basic data, enterprise investor data, tax declaration data, tax collection data, enterprise upstream and downstream data, enterprise balance sheet, enterprise income statement, enterprise audit information, and enterprise change information for integrated analysis. The data is updated frequently, and the query content is comprehensive and accurate. By breaking through data barriers, it builds portraits for massive customers and conducts fine screening, conducts credit limit matching and post-loan management, and helps banks and other financial institutions to create a panoramic portrait of customer credit, accurately focus on customers with good credit and sufficient repayment ability, improve the comprehensiveness and accuracy of risk assessment of banks and other financial institutions, and effectively alleviate problems such as difficult financing processes, low financing efficiency, and lack of mortgage guarantees for small and micro enterprises and downstream distributors of core enterprises.

[0010] Compared with the prior art, the advantages of the present invention are: (1) Existing solutions mostly focus on optimizing the three main business links of pre-loan investigation, loan approval, and post-loan monitoring, but lack links such as customer flow and data interaction. This application provides one-stop online services such as "customer flow, data interaction, risk control decision-making, credit limit application, online withdrawal, and post-loan management."

[0011] (2) Introduce external data information such as taxation, provident fund, and real estate, and combine it with internal data of banks and other financial institutions to enrich the information dimensions of pre-loan access and post-loan risk management, and improve the comprehensiveness and accuracy of risk assessment of banks and other financial institutions.

[0012] (3) Use RPA (Robotic Process Automation) and OCR (Optical Character Recognition) technology to automatically query, identify, and compare image invoice information. Feedback the processing results directly to business personnel. The entire process of loan material preprocessing does not require human intervention, which greatly saves verification time, frees up the energy of branch business personnel, and improves the verification accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a structural block diagram of an inclusive financial service platform based on big data, artificial intelligence and RPA in an embodiment of the present invention. DETAILED DESCRIPTION

[0014] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention can also be implemented in other ways different from those described herein, and therefore, the present invention is not limited to the limitations of the specific embodiments disclosed below.

[0015] like Figure 1 As shown, the inclusive financial service platform based on big data, artificial intelligence and RPA provided by the present invention is a full-process service platform. The inclusive financial service platform covers all aspects of inclusive services, and provides one-stop online services such as "customer flow, data interaction, risk control decision-making, credit application, online withdrawal, and post-loan management", achieving full coverage of solutions from pre-loan, mid-loan, and post-loan; the inclusive financial service platform includes a message push system, a data receiving and collection module, a data storage module, a pre-loan access credit approval module, and a post-loan management module, which are as follows: Message push system: Used to publish credit product information, conduct customer flow and data interaction, and publish credit product information of banks and other financial institutions through self-built channels such as bank WeChat official accounts, H5 bank web pages (bank web pages developed based on HTML5 technology), online banking (online banking system), mobile banking, mini-programs, and third-party platforms (such as inclusive financial comprehensive service platforms led by local governments, etc.), so that enterprises or individuals can inquire about credit products that meet their needs and direct customers with credit needs.

[0016] Data receiving and acquisition module: It is used to receive credit requests from enterprises or individuals, and automatically enter the basic information of customers (i.e., enterprises or individuals who make credit requests) through AI recognition, OCR and other technologies. After the customer fully authorizes, the bank can obtain the customer's recent data from the internal system of the bank (financial institution) or from the tax bureau, provident fund and other platforms through dedicated lines or the Internet. The customer's recent data to be obtained includes customer tax data, invoicing data, government big data, corporate and individual credit data, stock business data including intra-bank settlement data, third-party anti-fraud data and other data information; at the same time, the data receiving and collection module is used to integrate the collected scattered customer recent data information, remove duplicate and invalid data, and store it in the database of the unified data storage module for centralized management and subsequent analysis; based on the integrated customer recent data, use data analysis and mining technology to establish richer customer labels including customer business information, customer credit information, etc., to have a more real and in-depth understanding of target customers; There are two specific ways to obtain recent customer data from internal systems of banks (financial institutions), tax bureaus, provident funds and other platforms through dedicated lines or the Internet: (1) the inquirer calls the data call interface provided by the data storage provider to obtain data; (2) the inquirer provides the data call interface, and the data storage provider pushes the data to the inquirer by calling the data call interface or through files.

[0017] Data storage module: The data storage module includes a database and an object storage module, which are used to store customer tags, corresponding basic customer information, integrated recent customer data, and related analysis data generated during the credit approval process.

[0018] Pre-loan access credit approval module: Based on OCR technology and RPA technology, the credit materials provided by the customer are verified to determine whether the customer meets the credit access conditions; then the credit materials provided by the customer and the integrated recent customer data are used for risk assessment and credit limit calculation, and the customer credit approval is automatically completed, and the approval results and credit limits of the corresponding customers are obtained, and the approval results and credit limits of the corresponding customers are uploaded to the data storage module. The existing approval models involved include access model, anti-fraud model, approval decision model, credit limit approval model, risk pricing model, etc.

[0019] The admission model adopts a scoring card model, which evaluates admission qualifications by weighted scoring of multiple factors. For different products, first determine the factors related to admission, such as the customer's age, income, assets, credit record, etc., then assign a certain weight to each factor, and give corresponding scores based on the applicant's performance in each factor. Finally, add up the scores of all factors to get the total score; then judge based on the set score line, and applicants with a total score higher than the score line will be eligible for admission.

[0020] The anti-fraud model consists of four parts: risk feature extraction module, new account risk prevention and control module, dynamic quota control module and account verification and analysis platform. The risk feature extraction module extracts risk features through multiple dimensions such as customer view, account transaction, and device fingerprint. The new account risk prevention and control module strictly implements the identity review of the account holder and conducts risk scoring with the help of big data model capabilities. The dynamic quota control module builds a risk scoring model system and conducts regular rating and dynamic adjustment. The account verification and analysis platform forms a professional toolkit through multi-dimensional label analysis to quickly grasp risk features. The anti-fraud model monitors the risks of various transactions through in-memory computing and artificial intelligence algorithms, thereby identifying abnormal customers, abnormal accounts, and abnormal transactions, and accurately blocking them through stratified classification.

[0021] The approval decision model extracts the basic information of the customer and the integrated recent data of the customer from the data storage module according to the customer tag, including financial status, operating status, credit record, etc., and conducts a comprehensive risk assessment based on the basic information of the customer and the integrated recent data of the customer, including credit risk assessment, market risk assessment, operational risk assessment, etc. According to different scenarios, a clear approval flow chart is formulated, approval indicators are set, and approval time is managed, while ensuring compliance, efficiency and fairness of approval. Based on the above process, financial decisions such as whether to approve loans and credit are made.

[0022] The credit approval model is divided into data collection and preprocessing module, credit assessment module, risk assessment module, credit calculation module, and approval decision module. The credit approval model will preprocess and clean the collected basic information, financial data, and operating data of customers; and quantitatively evaluate the credit status of the enterprise based on the processed data, determine the risk level and risk control measures; and finally calculate the loan amount based on the credit assessment results, risk level, and repayment ability of the enterprise.

[0023] The risk pricing model will comprehensively consider factors such as the credit status, repayment ability, and risk level of the borrowing enterprise, and ensure that the risks and returns of the loan are matched by differentiated pricing for enterprises with different risk levels. In the pricing process, it will also comprehensively consider factors such as market competition and the overall business strategy of banks (financial institutions) to ensure the rationality and competitiveness of pricing.

[0024] Post-loan management module: Used for post-loan management, through real-time monitoring of tax data and taking corresponding post-loan management measures, the cooperating third-party platform will timely update the monthly operating data and push it to the bank (financial institution) based on the customer's authorization, so as to facilitate the bank (financial institution) to conduct process-based and systematic tracking of customer withdrawals, interest settlements, repayments, etc.; use multi-dimensional data to solve the financing difficulties strategy, so that the bank (financial institution) can conduct risk control analysis on the customer's operating data every month, track the customer's operating data according to the post-loan management regulations, and pre-remind the customer of interest settlement and repayment.

[0025] Among them, the specific content of post-loan management includes: regularly calling relevant data calling interfaces to obtain the latest tax data of customers, and using data analysis technology, machine learning, deep learning technology, combined with other customer data information, to build a risk assessment model, comprehensively considering factors such as customer credit history, income stability, debt level, operating conditions, industry risks, etc., to quantitatively assess the customer's default risk and provide a scientific basis for approval decisions. Real-time risk monitoring of issued financial products, by tracking customer repayments, changes in operating conditions, market environment fluctuations and other information, timely discover potential risk signals. Once the risk index exceeds the preset threshold, the system automatically issues an early warning to remind bank (financial institution) business personnel to take corresponding risk disposal measures, such as early collection, adjustment of quotas, additional guarantees, etc. According to the risk warning information and risk assessment results, formulate corresponding risk disposal plans, including negotiating repayment plans with customers, taking legal means to recover arrears, and writing off bad debts. At the same time, analyze and summarize risk data, continuously optimize risk assessment models and management strategies, and improve the risk prevention capabilities of inclusive financial service platforms.

[0026] The strategy of using multi-dimensional data to solve the financing dilemma is as follows: Make full use of data analysis and mining technology, machine learning, and deep learning technology, rely on Internet channels, integrate tax, provident fund, real estate and other information, give play to the advantages of information sharing, use internal and external data information legally and compliantly, and use data including enterprise basic data, enterprise investor data, tax declaration data, tax collection data, enterprise upstream and downstream data, enterprise balance sheet, enterprise income statement, enterprise audit information, enterprise change information and other aspects for integrated analysis. The inclusive financial service platform of the present invention has a high data update frequency, and the query content is comprehensive and accurate. By breaking through data barriers, we build portraits for massive SMEs and conduct detailed screening, screen out enterprises with good credit and sufficient repayment ability as potential borrowers, and provide them with funds. In addition, in order to reduce the default risk of banks (financial institutions) and avoid the loss of profits caused by default, it is also necessary to conduct overdue analysis and build portraits of overdue users, so as to facilitate the improvement of the risk prevention capabilities of the inclusive financial service platform, so as to more efficiently match the quota and manage post-loans, and help banks and other financial institutions to conduct panoramic portraits of corporate and personal credit, accurately focus on high-quality target customers (customers with good credit and sufficient repayment ability), improve the comprehensiveness and accuracy of bank (financial institution) risk assessment, and effectively alleviate the problems of difficult financing processes, low financing efficiency, and lack of mortgage guarantees for small and micro enterprises and downstream distributors of core enterprises.

[0027] The verification of credit materials based on OCR technology and RPA technology is specifically as follows: Through RPA technology, the credit material compression package provided by the customer is decompressed in batches and the loan data is preprocessed to realize the automatic preprocessing of loan materials, overcoming the problems of cross-domain transmission of huge amounts of data (TB level) and processing of unknown data formats. And through OCR technology to identify the key information of the material, taking invoice information as an example, the key information to be identified includes the name of the purchaser of the invoice, invoice code, invoice number, invoice date, machine-printed number, machine-printed code, invoice amount, total price in lowercase, value-added tax amount, etc. After the identification is completed, the query interface of the bank invoice management system (the financial institution's own internal management system) is called to verify the invoice information, complete the invoice verification results and other information, realize the processing of single loan information in seconds, and automatically summarize the verification results and data integrity to generate an EXCEL file, notify the business personnel of the bank (financial institution) by email and upload it to the object storage module of the data storage module for subsequent inquiries by business personnel. The generated EXCEL file includes the following information: loan number, whether the materials required for the corresponding loan are complete, and the name of the missing materials; among them, the specific material details will vary according to different credit products.

[0028] Based on the above inclusive financial service platform, the present invention also provides a credit approval method for inclusive financial services based on big data, artificial intelligence and RPA, and the specific steps are as follows: Step S1: When receiving a credit request from an enterprise or individual, a bank or other financial institution uses AI recognition and OCR technology to enter and store the basic information of the customer (i.e., the enterprise or individual that issued the credit request), obtains the customer's recent data from the tax bureau, provident fund and other platforms with the customer's authorization, integrates the collected scattered customer recent data information, removes duplicate and invalid data, and obtains integrated customer recent data; based on the integrated customer recent data, establishes customer tags; The recent customer data is established based on customer tax data, invoicing data, government big data, credit data of corporate and individual legal persons, stock business data including intra-bank settlement data, third-party anti-fraud data and other data information; Step S2: Based on OCR technology and RPA technology, the credit materials provided by the customer are verified to determine whether the customer meets the credit access conditions. If the customer does not meet the credit access conditions, the customer is notified by email that the credit service cannot be provided. If the customer meets the credit access conditions, the credit materials provided by the customer and the integrated recent customer data are subjected to risk assessment and credit limit calculation through the access model, anti-fraud model, approval decision model, credit limit approval model, and risk pricing model, and the customer credit approval is automatically completed to obtain the approval result and credit limit of the corresponding customer. Among them, risk assessment includes account holder identity review, credit risk assessment, market risk assessment, operational risk assessment, asset assessment, repayment ability assessment, risk level assessment, etc. Step S3, determine whether the customer accepts the credit plan and whether the credit business has been processed. If the customer has processed the credit business, post-loan management is required. Through real-time monitoring of the corresponding customer's tax data and monthly operating data, the customer's withdrawal, interest payment, repayment and other process-based and systematic tracking is carried out; the strategy of using multi-dimensional data to solve the financing difficulties is used to conduct risk control analysis on the customer's updated operating data every month, and the customer's operating data is tracked according to the post-loan management regulations of the bank (i.e., financial institution), and the customer's interest payment and repayment are pre-reminders.

Claims

1. An inclusive financial service platform based on big data, artificial intelligence and RPA, characterized by: It includes message push system, data receiving and collecting module, data storage module, pre-loan access and credit approval module, and post-loan management module; The message push system is used to publish credit product information of financial institutions, conduct customer flow and data exchange; The data receiving and collecting module is used to receive the customer's credit request and automatically enter the customer's basic information through AI recognition and OCR technology. After receiving the customer's full authorization, it obtains the customer's recent data, integrates the collected customer's recent data information, removes duplicate and invalid data, and stores it in the data storage module; Based on the integrated recent customer data, establish customer labels including customer business information and customer credit information; The data storage module is used to store customer tags, basic information of corresponding customers, integrated recent customer data, and relevant analysis data generated during the credit approval process; The pre-loan access credit approval module verifies the credit materials provided by the customer based on OCR technology and RPA technology to determine whether the customer meets the credit access conditions; conducts risk assessment and credit limit calculation on the credit materials provided by the customer and the integrated recent customer data through the approval model, automatically completes the customer credit approval, obtains the approval result and credit limit of the corresponding customer, and uploads the approval result and credit limit of the corresponding customer to the data storage module; The post-loan management module is used for post-loan management, tracking customer business data, and providing advance reminders for customer interest settlement and repayment.

2. The inclusive financial service platform based on big data, artificial intelligence and RPA according to claim 1, characterized in that: The recent customer data that needs to be obtained include customer tax data, invoicing data, government big data, credit data of corporate and individual legal persons, existing business data including intra-bank settlement data, and third-party anti-fraud data; the sources of recent customer data include the tax bureau data platform and provident fund data platform.

3. The inclusive financial service platform based on big data, artificial intelligence and RPA according to claim 1 is characterized in that: The approval model includes an access model, an anti-fraud model, an approval decision model, a quota approval model, and a risk pricing model.

4. The inclusive financial service platform based on big data, artificial intelligence and RPA according to claim 1, characterized in that: The verification of credit materials based on OCR technology and RPA technology is specifically as follows: RPA technology is used to batch decompress the credit material compression packages provided by customers and pre-process the loan data to achieve automated pre-processing of loan materials, overcoming the problems of massive cross-domain data transmission and unknown data format processing; and OCR technology is used to identify the key information of the materials. After the identification is completed, the query interface of the financial institution's own management system is called to verify the identified key information, and the verification results and data integrity are summarized to generate an EXCEL file, which is notified to the business personnel of the financial institution by email and uploaded to the object storage module of the data storage module for subsequent inquiries by the business personnel.

5. The inclusive financial service platform based on big data, artificial intelligence and RPA according to claim 4 is characterized in that: The EXCEL file includes the following information: the loan note number, whether the materials required for the corresponding loan note are complete, and the names of the missing materials.

6. The inclusive financial service platform based on big data, artificial intelligence and RPA according to claim 1, characterized in that: The post-loan management includes real-time monitoring of tax data and monthly operating data, and procedural and systematic tracking of customer withdrawals, interest payments, and repayments; using multi-dimensional data to solve financing difficulties and conduct risk control analysis on customer operating data, and tracking customer operating data in accordance with the post-loan management regulations of financial institutions, and pre-reminders of customer interest payments and repayments.

7. The inclusive financial service platform based on big data, artificial intelligence and RPA according to claim 6 is characterized in that: The strategy of using multi-dimensional data to solve the financing dilemma is specifically: It integrates tax, provident fund and real estate information, and uses multi-faceted data including enterprise basic data, enterprise investor data, tax declaration data, tax collection data, enterprise upstream and downstream data, enterprise balance sheet, enterprise income statement, enterprise audit information, and enterprise change information for integrated analysis. The data is updated frequently, and the query content is comprehensive and accurate. By breaking through data barriers, it builds portraits for massive customers and conducts fine screening, conducts credit limit matching and post-loan management, and helps financial institutions to create a panoramic portrait of customer credit, accurately focus on customers with good credit and sufficient repayment ability, and improve the comprehensiveness and accuracy of financial institutions' risk assessment.

8. A credit approval method for inclusive financial services based on big data, artificial intelligence and RPA, applied to the inclusive financial service platform described in any one of claims 1 to 7, and the specific steps are as follows: Step S1: When a financial institution receives a credit request from a customer, it uses AI recognition and OCR technology to enter and store the basic information of the customer who issued the credit request, obtains the customer's recent data through the customer's authorization, integrates the collected customer's recent data information, removes duplicate and invalid data, and obtains integrated customer's recent data; and establishes customer tags based on the integrated customer's recent data; Step S2: Based on OCR technology and RPA technology, the credit materials provided by the customer are verified to determine whether the customer meets the credit access conditions. If the customer does not meet the credit access conditions, the customer is notified that the credit service cannot be provided; if the customer meets the credit access conditions, the credit materials provided by the customer and the integrated recent customer data are subjected to risk assessment and credit limit calculation through the access model, anti-fraud model, approval decision model, credit limit approval model, and risk pricing model, and the customer's credit approval is automatically completed, the approval result and credit limit of the corresponding customer are obtained, and the customer is notified; Step S3, determine whether the customer accepts the credit plan and whether the customer has applied for a credit business. If the customer has applied for a credit business, post-loan management is required. Through real-time monitoring of the corresponding customer's tax data and monthly operating data, the customer's withdrawal, interest payment, and repayment are tracked in a procedural and systematic manner; the strategy of using multi-dimensional data to solve the financing difficulties is used to conduct risk control analysis on the customer's updated operating data every month, and the customer's operating data is tracked according to the post-loan management regulations of the financial institution, and the customer is pre-reminded of interest payment and repayment.

Images