Prepaid consumption supervision system and method based on big data and storage medium

Abstract

The invention discloses a prepaid consumption supervision system and method based on big data and a storage medium, and relates to the technical field of big data analysis, the system comprises the following components: a merchant information management module, a fund supervision module, a risk monitoring module, a consumption service module and a data analysis module; according to the invention, the comprehensive supervision system covering merchant information management, fund supervision, risk monitoring, consumption service and data analysis is constructed, so that the whole-process supervision of prepayment consumption from merchant settlement to consumption refund is realized, the safety of prepayment funds of consumers is ensured, the problems of'fee storage running 'and the like are effectively avoided, and the safety of the consumers is ensured. In addition, potential risks are identified and prevented in advance through a risk prediction model and a real-time early warning mechanism, the initiative and timeliness of supervision are remarkably improved, and meanwhile, through application of a data analysis module, deep market insight is provided for supervision departments.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and specifically provides a prepaid consumption supervision system, method and storage medium based on big data. Background Art

[0002] A prepaid consumption card refers to a voucher that a user purchases from a card-issuing merchant and can be used to redeem goods, which is divided into physical vouchers and virtual vouchers. Physical vouchers include magnetic stripe cards, chip cards, paper vouchers, etc., and virtual vouchers include passwords, serial numbers, graphics, biometric information, etc.

[0003] In recent years, with the continuous development of the economy, new consumption models have quietly emerged, and the fields covered by prepaid consumption have become more extensive. It is mainly applied to various aspects of economic life such as education and training, beauty and hair care, commercial retail, catering and entertainment, sports and fitness, etc., which has played a positive role in facilitating payment, promoting consumption, and prospering the market. At the same time, along with risks and problems such as "inducing card handling", "running away with deposited fees", "overbearing clauses", and "refusing to refund money when the operator changes the address", it has increased the difficulty of protecting consumers' rights and interests, and consumer disputes have occurred frequently. It is urgent to be regulated by legislative means. In view of the deficiencies of traditional prepaid consumption supervision methods, existing technical means seem inadequate in dealing with prepaid consumption supervision problems. Traditional supervision methods mainly rely on manual review and post-event processing, which are not only inefficient but also difficult to achieve real-time monitoring and early warning of the entire process of prepaid consumption. In addition, traditional technical means have limitations in data processing and analysis, are difficult to explore and utilize the potential value in big data, and cannot provide strong data support for prepaid consumption supervision. Therefore, a prepaid consumption supervision system, method and storage medium based on big data are proposed to solve the above problems. Summary of the Invention

[0004] The purpose of the present invention is to make up for the deficiencies of the existing technology and provide a prepaid consumption supervision system, method and storage medium based on big data. It can conduct online supervision of the entire process before, during and after the event, build data models for each merchant, and the data includes funds custody information, surety insurance information, tax information, employees' social security payment information, water, electricity, gas payment information, credit information, industry risks, illegal information, public events, etc. Each piece of data forms a feature. Through each data feature, a merchant risk prediction model is established to calculate the prediction results of each merchant in real time. At the same time, it also increases the publicity and support for operators, and better provides services and empowerment for operators.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: In the first aspect, a prepaid consumption supervision system based on big data, the system includes the following components: a merchant information management module, a funds supervision module, a risk monitoring module, a consumption service module, and a data analysis module;

[0006] The merchant information management module develops a QR code scanning and registration program, collects merchant registration information through front-end form verification, verifies the information through the government data interface and matching algorithm, sets up options for capital guarantee measures and verifies account information, and generates a prepayment code identifier and certificate associated with the filing information through the filing identifier generation technology;

[0007] The fund supervision module establishes a secure data interface with the bank to deposit the consumer recharge funds into the supervision account in real time. At the same time, it develops a scheduled task program. On the T+1 day after the consumer's consumption, it automatically releases the corresponding amount from the supervision account to the merchant settlement account according to the consumption record, and establishes a guarantee / insurance application process to interact with institutions to handle the unreleased funds, and processes refunds using a logical judgment program according to the card refund rules;

[0008] The risk monitoring module collects multi-dimensional data of merchants from multiple data sources, integrates and stores them through ETL technology, and establishes a data model for each merchant. Each piece of data forms a feature. Through each data feature, a risk prediction model is constructed using the Bagging regression algorithm and optimization technology, the risk level is calculated according to the evaluation logic, and early warning information is pushed through the RocketMQ message queue;

[0009] The consumption service module develops an online contract signing mode to enable consumers to sign contracts online, associates and stores the contract information with the accounts of consumers and merchants, and establishes a complaint and reporting platform to process information through a classification and distribution and feedback mechanism. At the same time, it develops a data display interface to display the risk credit information of merchants to consumers. Through the consumption amount deduction program, when consumers consume normally, the corresponding amount is automatically deducted from the supervision account or performance guarantee;

[0010] The data analysis module uses data mining and machine learning algorithms to analyze market data, evaluate the potential risk level of the market, and uses visualization tools and real-time update mechanisms to display market dynamics.

[0011] Furthermore, the merchant information management module verifies the information through the government data interface and matching algorithm, and its algorithm is: Among them, V is the merchant information verification value, which is used to measure the difference between the information m submitted by the merchant i and the internal data g of the government regulatory agency i The smaller the V value, the higher the information consistency. n is the number of information items participating in the verification, including but not limited to the operator name, legal person name, and business address, and w i is the weight of the i-th information item, which reflects the importance of this information item in the overall verification, and m i is the content of the i-th information item submitted by the merchant, and g iIt is the content of the i-th information item corresponding to the internal data of the government regulatory agency.

[0012] Furthermore, the fund supervision module processes the refund using a logical judgment program according to the card refund rule, and the calculation formula for the refund amount is: where Refund is the amount that should be refunded to the consumer when the card is refunded, B is the balance in the prepaid card when the consumer refunds the card, which is calculated by the fund supervision module based on the recharge and consumption records, k is the number of times the consumer has enjoyed the preferential treatment or rights and interests, S j is the value corresponding to the j-th time of enjoying the preferential treatment or rights and interests, which is set by the merchant and recorded in the consumption record, r is the time value coefficient of the preferential treatment or rights and interests value, which is set by the platform according to factors such as the market interest rate, t j is the time of the j-th time of enjoying the preferential treatment or rights and interests, t 0 is the time of purchasing the card.

[0013] Furthermore, the risk monitoring module develops a data collection program to collect data on the merchant's fund deposit, guarantee insurance, tax payment, employee social security payment, water, electricity and gas payment, credit, industry risk, illegal information and public events from the bank, tax, social security and water, electricity and gas departments. It uses ETL technology to clean, transform and integrate the collected data, stores the data in the data warehouse, builds a risk prediction model for each merchant, converts each piece of data into a feature, uses feature engineering technology to extract and select features from the data, applies the Bagging regression algorithm to train the merchant risk prediction model based on historical data, and uses cross-validation and grid search technology to optimize the parameters of the model. At the same time, it formulates the risk warning index judgment logic, combines multi-dimensional factors such as the risk status factor, credit rating judgment logic and credit rating factor, calculates the risk assessment value of the merchant, and triggers the warning mechanism when the risk level reaches the threshold by setting the risk warning threshold. It uses the RocketMQ message queue technology to push the risk warning information to the regulatory agency in real time.

[0014] Furthermore, the risk monitoring module builds a risk prediction model for each merchant, and its model formula is: where Risk is the risk prediction value of the merchant, m is the number of different types of risk prediction models integrated, such as models based on financial data, models based on market data, etc., a l is the weight of the l-th model, reflecting its importance in the comprehensive risk prediction, b l0 is the constant term of the l-th model, b li is the coefficient of the i-th risk feature in the l-th model, n is the number of risk features in each model, x li is the value of the i-th risk feature in the l-th model, obtained from the data collection and integration module, including data provided by the bank, tax, credit institution, etc.

[0015] Furthermore, the risk monitoring module calculates the risk level of the merchant by combining multi-dimensional factors such as risk status factors, credit rating evaluation logic, and credit rating factors. The formula for calculating its risk level is as follows: where RiskLevel is the risk assessment value of the merchant, p is the number of risk status factors, q is the number of credit rating factors, α i is the weight of the i-th risk status factor, β j is the weight of the j-th credit rating factor, R i is the score of the i-th risk status factor, C j is the score of the j-th credit rating factor, which are calculated based on the business operation data and credit data of the merchant respectively.

[0016] Furthermore, the data analysis module analyzes the market data using data mining and machine learning algorithms to evaluate the potential risk level of the market. The formula for calculating its potential risk level is as follows: where RiskPoint is the evaluation value of the potential risk point of the market, c is the number of market data indicators for analysis, λ i is the weight of the i-th market data indicator, reflecting its importance for the risk point evaluation, x i is the value of the i-th market data indicator, is the average value of the i-th market data indicator, σ 2 is the variance of the i-th market data indicator.

[0017] In the second aspect, a prepaid consumption supervision method based on big data is applicable to a prepaid consumption supervision system based on big data according to any one of claims 1-7. The method is characterized in that the system includes the following specific steps:

[0018] Merchant entry and filing: The card-issuing operator scans the QR code to complete code acquisition, filing registration, information authorization, selects the advance payment fund guarantee measure and binds the account information, and obtains the prepaid code identifier and certificate;

[0019] Fund supervision operation: After the consumer recharges, the funds enter the supervision account. When consuming, the funds are released to the merchant's settlement account according to the "consumption T+1 release" rule. The merchant can apply for the release of unconsumed funds, and the consumer's card refund is processed according to the platform rules;

[0020] Risk monitoring and early warning: The platform continuously collects multi-dimensional data of the merchant, conducts data modeling and risk prediction model calculation, evaluates the risk level, and pushes information to the regulatory agency when the risk exceeds the standard;

[0021] Consumer service guarantee: When consumers purchase a card, they sign an electronic contract. During consumption, the corresponding amount is deducted by the system. After consumption, consumers can file complaints and reports, and at the same time, they can view the risk credit information of merchants.

[0022] Data analysis and processing: Use machine learning algorithms to analyze market data, evaluate the potential risk level of the market, and use visualization tools and real-time update mechanisms to display market dynamics.

[0023] Thirdly, a computer-readable storage medium stores a computer program, and when the computer program is executed, it implements the prepaid consumption supervision system based on big data described above.

[0024] Compared with the prior art, the prepaid consumption supervision system, method and storage medium based on big data have the following beneficial effects:

[0025] First, by constructing an all-round supervision system covering merchant information management, fund supervision, risk monitoring, consumer service and data analysis, the present invention realizes the whole-process supervision of prepaid consumption from merchant entry to consumption refund. It not only ensures the safety of consumers' prepaid funds and effectively avoids problems such as "running away with deposits", but also through the risk prediction model and real-time warning mechanism, it can identify and prevent potential risks in advance, significantly improving the initiative and timeliness of supervision. At the same time, the application of the data analysis module provides in-depth market insights for the supervision department, helps to formulate more scientific and reasonable policy measures, further standardize the market order and protect consumers' rights and interests.

[0026] Second, through functions such as electronic contract signing, complaint and report acceptance, and risk credit information display, the present invention enhances consumers' awareness of protecting their rights and interests and their participation. Consumers can more conveniently understand the credit status and risk level of merchants and make more informed consumption decisions. In addition, by optimizing the fund supervision and release mechanism, it reduces the financial pressure on merchants while ensuring consumers' rights to refund cards and refunds, promoting trust and cooperation between merchants and consumers.

[0027] Other advantages, objectives and features of the present invention will be described to some extent in the subsequent specification, and to some extent, they will be obvious to those skilled in the art based on the study of the following text, or can be learned from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0029] Figure 1 It is a schematic structural diagram of a prepaid consumption supervision system based on big data;

[0030] Figure 2 It is a flowchart of a prepaid consumption supervision method based on big data. Detailed implementation manners

[0031] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in combination with the accompanying drawings and preferred embodiments, detail the specific implementation manners, structures, features and their effects of the present invention as follows.

[0032] Embodiment 1

[0033] In the education and training industry, an educational institution provides tutoring and interest cultivation courses for primary and middle school students, covering major subjects such as mathematics and English, and interest courses such as painting and music. In this scenario, the prepaid consumption supervision system, method and storage medium based on big data of the present invention play an important role.

[0034] The operator of the training institution scans the QR code specially provided by the supervision system and quickly enters the online filing platform. On the platform page, the operator truthfully fills in various registration information, including the name of the institution, the amount of registered capital, the name and identity information of the legal person, the contact information of the person in charge, the detailed business address, and the course service content provided by the prepaid consumption card, such as "Primary School English Synchronous Improvement Course", "Children's Piano Enlightenment Course", etc. After filling in, the system automatically compares and verifies with the internal data of the government supervision agency to ensure the authenticity and accuracy of the information. After the verification passes, the operator chooses to deposit the prepaid funds into the bank supervision account as a measure to ensure the safety of the prepaid funds and completes the necessary information authorization operation. After the filing is successful, the platform will issue a prepaid code identifier and a filing certificate to the training institution, granting it the qualification to legally carry out prepaid consumption business. After verification, the operator chooses to deposit the prepaid funds into the bank supervision account as a measure to ensure the safety of the prepaid funds and completes the necessary information authorization operation. After the filing is successful, the platform will issue a prepaid code identifier and a filing certificate to the training institution, granting it the qualification to legally carry out prepaid consumption business.

[0035] The training institution opens a special supervision account in the bank that cooperates with the platform. After completing the account opening, scan the QR code to submit the special account registration information. After receiving the registration information and verifying that it is correct, the platform will issue a registration mark and certificate for it. When the parents of students purchase courses for their children and recharge, the recharge amount will be deposited in full into the bank's supervision account. For example, a parent enrolls his child in the "Junior High School Mathematics Intensive Course" and recharges 1,500 yuan. After the recharge operation is completed, the platform will receive a notification in time and confirm that the funds have safely entered the supervision account. According to the "T+1 release of consumption amount" rule, when a student completes a course, the system will release the corresponding course fee from the supervision account to the settlement account of the training institution on the next day. If the training institution has business problems, If there are development needs, such as purchasing new teaching equipment, expanding teaching venues, etc., the platform can apply to the platform for the release of unspent pre-collected funds by providing guarantees or purchasing insurance. After receiving the application, the platform will jointly review it with the guarantee / insurance agency. After the review is passed, part of the unreleased funds will be transferred to the settlement account of the training institution to meet its reasonable financial needs. When a consumer initiates a refund application, for example, a student no longer participates in the "Children's Dance Basics Course" due to changes in personal interests, and the parents request a refund, the training institution will receive the refund notice and conduct an assessment based on the pre-established refund rules. If the conditions for the refund are met, the amount to be refunded will be calculated based on the consumption record and refund rules, and the funds will be returned to the consumer's payment account.

[0036] The supervision platform will continuously and comprehensively collect various external data of the training institution. In terms of fund custody, it will closely monitor the flow of funds in the supervision account to ensure the safety of funds. In terms of tax information, it will pay attention to whether the training institution pays taxes on time and in full. The social security payment information of employees is used to determine whether the institution complies with labor laws and regulations and protects the rights and interests of employees. Credit information is obtained from professional credit rating agencies to comprehensively evaluate the reputation of the institution. By integrating these data, the platform builds a detailed data model for the training institution, using the risk warning indicator judgment logic, combined with risk status factors (such as basic information, business scale, credit record, years of operation, nature of operation, reward and punishment records, etc.) and credit rating factors (including market credit indicators, financial credit indicators, risk monitoring indicators, red line supervision indicators) and other multi-dimensional factors, and uses specific algorithms to build a risk prediction model. The model formula is: Once a training institution is found to have abnormal situations such as delayed tax returns or employee social security arrears, the risk model will calculate the institution's risk assessment value based on preset algorithms and weights. When the risk reaches the warning threshold, the platform will use standard risk warning prevention measures and advanced technologies such as cloud computing, the Internet of Things, and artificial intelligence to conduct all-round, real-time comprehensive risk monitoring of the institution's business operations, and promptly push risk information to the competent industry authorities so that the regulatory authorities can take effective regulatory measures in a timely manner.

[0037] When students consume during normal classes at the training institution, the system will automatically and accurately deduct the corresponding amount from the regulatory account. For example, the single-class fee for the "Junior High School English Special Tutoring Course" is 80 yuan. After the student finishes the class, the system will deduct 80 yuan from the regulatory account to the settlement account of the training institution on the T+1 day according to the rules, ensuring the safe flow of consumption funds and accurate records. If the training institution encounters business risks, such as tight capital chains, damaged reputation due to a large number of complaints caused by teaching quality problems, etc., the platform will immediately activate the response mechanism, strictly control the deposited prepaid funds, suspend the release of funds to prevent unreasonable loss of funds. At the same time, the platform will quickly push detailed risk information to the local education authorities and market supervision departments. After receiving the risk information, the relevant departments will take targeted measures according to the specific situation of the risk, such as requiring the institution to rectify, assisting in handling consumer refund matters, etc., to effectively protect the legitimate rights and interests of consumers.

[0038] The regulatory platform uses powerful big data analysis technology to deeply explore and analyze the prepaid consumption market of the entire education and training industry. Through the processing and analysis of a large amount of consumption data, potential risk points can be accurately discovered. The formula for calculating its potential risk level is: Among them, RiskPoint is the evaluation value of the potential risk point in the market, c is the number of market data indicators used for analysis, λ i is the weight of the i-th market data indicator, reflecting its importance to the risk point evaluation, x i is the value of the i-th market data indicator, is the average value of the i-th market data indicator, σ 2 is the variance of this indicator. For example, some training institutions have abnormal transaction behaviors such as concentrated large-scale recharges but subsequent services cannot be guaranteed, providing accurate early warning information for the regulatory department in a timely manner to assist the regulatory department in preventing industry systematic risks. At the same time, the platform deeply analyzes consumer behavior data through the method of traffic empowerment and provides personalized and accurate marketing suggestions for the training institution. For example, according to the age distribution and interest preferences of students in the surrounding communities, targeted course discount activities are pushed to parents to help the training institution attract more potential students and improve sales and business handling efficiency. In addition, the platform uses data visualization technology to comprehensively display the market dynamics of the education and training industry in the jurisdiction in the form of intuitive charts, reports, etc., including key information such as the total transaction volume, growth rate, and risk distribution of each institution. The regulatory department and the training institution can adjust regulatory strategies and business decisions in a timely manner based on these real-time and accurate data, effectively respond to potential risks, and promote the healthy and orderly development of the entire education and training industry.

[0039] Embodiment 2

[0040] In the beauty and hair industry, a merchant's business scope covers multiple fields such as haircuts, beauty treatments, and the sale of hair care products. The operator of this merchant can quickly enter the online filing platform by scanning the QR code set by the supervision system. On the platform interface, the operator needs to fill in various registration information in detail, including the merchant name, registered capital, legal person name, contact information of the person in charge, business address, and service items provided through prepaid consumption, such as "Fashion Haircut Customization Package" and "Deep Skin Care Card". The front-end form verification technology equipped in the system will check the format of the input information in real time to ensure the accuracy of the information. After the information is filled in, the system will automatically compare and verify it with the internal data of the government supervision agency. After passing the verification, the operator chooses to deposit the prepaid funds into the bank supervision account as a safeguard measure and completes the information authorization. The platform issues a prepaid code identification and a filing certificate for the merchant.

[0041] When a consumer recharges and applies for a membership card at this merchant, the recharge funds will be fully deposited into the bank supervision account. Suppose a consumer recharges 800 yuan. After the recharge operation is completed, the platform will receive a notice in a timely manner and confirm that the funds have safely entered the supervision account. According to the rule of "releasing the consumption amount on T+1", if the consumer consumes 100 yuan of hair care services, the system will release 100 yuan from the supervision account to the merchant's settlement account the next day. If the merchant needs funds for business operations, such as purchasing new hair care equipment or introducing new beauty products, it can apply to the platform for releasing the unconsumed prepaid funds by means of guarantee or purchasing insurance. After receiving the application, the platform will cooperate with the guarantee / insurance institution to conduct a review. After passing the review, the corresponding funds will be transferred to the merchant's settlement account. When a consumer initiates a membership card refund application, the merchant will evaluate according to the established refund rules after receiving the refund notice. If the consumer meets the refund conditions, the platform will, in accordance with the relevant procedures, return the funds to the consumer's original payment account.

[0042] The supervision platform continuously collects multi-dimensional external data of this merchant, including the fund deposit situation, tax payment records, employee social security payment information, credit status, etc. The platform sorts out and analyzes the collected data, constructs a data model for the merchant, and uses the risk warning index judgment logic. Combining risk status factors (such as basic information, business scale, credit records, etc.) and credit rating factors (market credit indicators, financial credit indicators, etc.), a risk prediction model is constructed. Once abnormal situations occur for the merchant, such as delayed tax declaration or overdue employee social security payments, the risk model will calculate the risk assessment value of the merchant according to the preset algorithms and weights. When the warning threshold is reached, the platform uses risk warning standard preventive measures and uses advanced technologies such as cloud computing, Internet of Things, and artificial intelligence to comprehensively monitor the business operations of the merchant for risks and push the risk information to the industry competent department in a timely manner so that the supervision department can take effective supervision measures in a timely manner.

[0043] When consumers make normal purchases at this merchant, the system will automatically deduct the corresponding amount from the regulatory account. For example, if a consumer spends 150 yuan on beauty services, the system will deduct 150 yuan from the regulatory account to the merchant's settlement account on T+1 day, ensuring the safe flow and accurate recording of consumer funds, and protecting the interests of both consumers and merchants. If the merchant encounters business risks, such as tight capital chains, damaged reputation due to a large number of complaints caused by service quality problems, etc., the platform will immediately activate the response mechanism, control the deposited advance funds, suspend the release of funds, and prevent unreasonable loss of funds. At the same time, the platform will quickly push detailed risk information to the local market supervision department and the commerce department. After receiving the risk information, the relevant departments will take targeted measures according to the specific situation of the risk, such as requiring the merchant to rectify and assisting in handling consumer refund matters, so as to effectively protect the legitimate rights and interests of consumers.

[0044] The regulatory platform uses big data analysis technology to deeply explore and analyze the prepaid consumption market of the entire beauty and hair industry. Through the processing of a large amount of consumption data, potential risk points can be accurately discovered. For example, abnormal transaction behaviors such as certain merchants having concentrated large-scale recharges but unable to guarantee subsequent services. The platform will promptly provide the information of these potential risk points to the regulatory department to assist the regulatory department in preventing industry systematic risks. At the same time, the platform analyzes consumer behavior data through the method of traffic empowerment and provides personalized and accurate marketing suggestions for this merchant. For example, according to the consumption preferences and habits of the surrounding community residents, the platform designs preferential activities for specific groups for the merchant to help the merchant increase sales and business handling efficiency. In addition, the platform uses data visualization technology to comprehensively display the market dynamics of the beauty and hair industry in the jurisdiction area in the form of intuitive charts, reports, etc., including key information such as the total transaction volume, growth rate, and risk distribution of each merchant. The regulatory department and merchants can adjust regulatory strategies and business decisions in a timely manner based on these real-time data, effectively respond to potential risks, and promote the healthy and orderly development of the entire beauty and hair industry.

[0045] The above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Although the present invention has been disclosed above with the preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to the above-disclosed technical content within the scope of the technical solution of the present invention to obtain equivalent embodiments with equivalent changes. However, as long as it does not depart from the content of the technical solution of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A prepaid consumption supervision system based on big data, characterized in that: The system includes the following components: merchant information management module, fund supervision module, risk monitoring module, consumer service module and data analysis module; The merchant information management module develops a QR code scanning registration program, collects merchant registration information with the help of front-end form verification, verifies information through government data interface and matching algorithm, sets fund security measure options and verifies account information, and generates a prepaid code identifier and certificate associated with the filing information through filing identifier generation technology; The fund supervision module establishes a secure data interface with the bank to allow consumers to deposit their recharged funds into the supervision account in real time. At the same time, a timed task program is developed to automatically release the corresponding amount from the supervision account to the merchant settlement account based on the consumption record on T+1 day after the consumer's consumption, and establish a guarantee / insurance application process to interact with the institution to handle the unreleased funds, and use a logical judgment program to handle the refund according to the card return rules; The risk monitoring module collects multi-dimensional data of merchants from multiple data sources, integrates and stores it through ETL technology, and establishes a data model for each merchant. Each piece of data forms a feature. Through each data feature, the risk prediction model is constructed using the Bagging regression algorithm and optimization technology, and the risk level is calculated according to the judgment logic, and the warning information is pushed through the RocketMQ message queue; The consumer service module develops an online contract signing model to enable consumers to sign contracts online, associates and stores contract information with the accounts of consumers and merchants, and establishes a complaint reporting platform to process information through classification allocation and feedback mechanisms. At the same time, a data display interface is developed to display the risk credit information of merchants to consumers. Through the consumption limit deduction procedure, when consumers consume normally, the corresponding amount is automatically deducted from the supervision account or performance guarantee; The data analysis module uses machine learning algorithms to analyze market data, assess the potential risk level of the market, and use visualization tools and real-time update mechanisms to display market dynamics.

2. A prepaid consumption supervision system, method and storage medium based on big data according to claim 1, characterized in that: The merchant information management module verifies information through the government data interface and matching algorithm, and the algorithm is: Among them, V is the merchant information verification value, which is used to measure the merchant's submitted information m i Internal data with government regulators i The smaller the V value, the higher the information consistency. n is the number of information items involved in the verification, including but not limited to the name of the operator, the name of the legal person and the business address. i is the weight of the i-th information item, reflecting the importance of the information item in the overall verification, m i is the content of the i-th information item submitted by the merchant, g i It is the corresponding content of the i-th information item in the internal data of the government regulatory agency.

3. A prepaid consumption supervision system, method and storage medium based on big data according to claim 1, characterized in that: The fund supervision module processes refunds using a logic judgment program according to the card refund rules, and the refund amount calculation formula is: Where Refund is the amount that should be returned to the consumer when the card is returned, B is the balance in the prepaid card when the consumer returns the card, k is the number of discounts or benefits that the consumer has enjoyed, and S j is the value corresponding to the jth enjoyment of preferential treatment or rights, r is the time value coefficient of preferential treatment or rights, t j is the time when the jth discount or benefit is enjoyed, and t0 is the time of card purchase.

4. A prepaid consumption supervision system, method and storage medium based on big data according to claim 1, characterized in that: The risk monitoring module develops a data collection program to collect data on merchants' fund custody, guarantee insurance, tax payment, employee social security payment, water, electricity and gas fee payment, credit, industry risk, illegal information and public events from banks, taxation, social security and water, electricity and gas departments, and uses ETL technology to clean, convert and integrate the collected data, and store the data in a data warehouse. By establishing a risk prediction model for each merchant, each piece of data is converted into a feature, and feature engineering technology is used to extract and select features from the data. The Bagging regression algorithm is used to train the merchant risk prediction model based on historical data, and cross-validation and grid search technology are used to optimize the parameters of the model. At the same time, a risk warning indicator evaluation logic is formulated, and the risk assessment value of the merchant is calculated by combining risk status factors, credit rating evaluation logic and credit rating factor multi-dimensional factors. By setting a risk warning threshold, when the risk level reaches the threshold, the warning mechanism is triggered, and the RocketMQ message queue technology is used to push the risk warning information to the regulatory agency in real time.

5. A prepaid consumption supervision system, method and storage medium based on big data according to claim 4, characterized in that: The risk monitoring module establishes a risk prediction model for each merchant, and the model formula is: Among them, Risk is the risk prediction value of the merchant, m is the number of different types of risk prediction models integrated, and a l is the weight of the lth model, reflecting its importance in comprehensive risk prediction, b l0 is the constant term of the lth model, b li is the coefficient of the ith risk feature in the lth model, n is the number of risk features in each model, x li is the value of the ith risk feature in the lth model.

6. A prepaid consumption supervision system, method and storage medium based on big data according to claim 4, characterized in that: The risk monitoring module combines the risk status factor, the credit rating judgment logic and the multi-dimensional factors of the credit rating factor to calculate the risk level of the merchant. The risk level calculation formula is: Among them, RiskLevel is the risk assessment value of the merchant, p is the number of risk status factors, q is the number of credit rating factors, α i is the weight of the ith risk profile factor, β j is the weight of the jth credit rating factor, R i is the score of the ith risk profile factor, C j It is the score of the jth credit rating factor, which is calculated based on the merchant’s operating data and credit data.

7. A prepaid consumption supervision system based on big data according to claim 1, characterized in that: The data analysis module uses data mining and machine learning algorithms to analyze market data and assess the potential risk level of the market. The potential risk level calculation formula is: Among them, RiskPoint is the evaluation value of the potential risk point in the market, c is the number of market data indicators used for analysis, and λ i is the weight of the ith market data indicator, reflecting its importance to the risk point assessment, x i is the value of the ith market data indicator, is the average value of the ith market data indicator, σ 2 is the variance of the ith market data indicator.

8. A prepaid consumption supervision method based on big data, the method is applicable to a prepaid consumption supervision system based on big data as described in any one of claims 1 to 7, characterized in that: The system comprises the following specific steps: Merchant registration: Card issuers scan the QR code, complete code collection, registration, information authorization, select pre-collected funds protection measures and bind account information, and obtain prepaid code identification and certificate; Fund supervision operation: After consumers top up, the funds enter the supervision account. When making a purchase, the funds are released to the merchant's settlement account according to the "T+1 release after consumption" rule. Merchants can apply to release unspent funds. Consumer card withdrawal is handled according to platform rules. Risk monitoring and early warning: The platform continuously collects multi-dimensional data from merchants, performs data modeling and risk prediction model calculations, assesses risk levels, and pushes information to regulatory agencies when risks exceed the limit; Consumer service guarantee: Consumers sign electronic contracts when purchasing cards, and the system deducts the corresponding amount when making purchases. After making purchases, they can file complaints and report, and can also view the merchant's risk credit information; Data analysis and processing: Use machine learning algorithms to analyze market data, assess the potential risk level of the market, and use visualization tools and real-time update mechanisms to display market dynamics.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed, a prepaid consumption supervision system based on big data as described in any one of claims 1 to 7 is implemented.

Images