Loan limit management system, computer equipment and storage medium

Abstract

The invention discloses a loan limit management system, computer equipment and a storage medium, and the system comprises a dynamic questionnaire engine which is used for generating an extensible question tree and collecting user data; the rule decision cluster is connected with the output end of the questionnaire engine and is used for executing admission control, quota calculation and product matching; the risk pricing gateway is connected with the output end of the rule decision cluster and is used for implementing dynamic risk compensation; and the bank interface adapter is connected with the risk pricing gateway to realize multi-bank protocol conversion. The dynamic questionnaire engine improves the data integrity rate through question tree extension and real-time verification; and through three-level product filtering and four-dimensional risk scoring, the matching accuracy is improved. User experience is optimized; multi-channel instant feedback is realized; the scheme optimization unit provides operability suggestions, and the user conversion rate is improved. The protocol conversion unit is in unified butt joint with multiple banks, so that the development efficiency is improved; a real-time monitoring instrument panel realizes fault minute-level positioning, and the operation and maintenance response speed is improved.

Description

Technical Field

[0001] This invention relates to the field of financial lending technology, specifically to a loan limit management system, computer equipment, and storage medium. Background Technology

[0002] Current loan pre-approval systems generally suffer from the following problems:

[0003] Rigid questionnaires: Traditional questionnaires cannot dynamically adjust questions based on real-time user feedback, resulting in the omission or redundancy of key data;

[0004] High degree of rule coupling: Access control, quota calculation and product matching logic are scattered, making it difficult to maintain and expand in a unified manner;

[0005] Inefficient bank integration: Banks have different protocols (such as SOAP / REST / FTP), requiring customized interface development, which is costly.

[0006] Risk pricing is simplistic: it relies on static scoring models and fails to incorporate dynamic factors such as occupation and age to compensate for risk.

[0007] Weak feedback mechanism: lack of multi-channel real-time output and optimization suggestions, resulting in a poor user experience.

[0008] Therefore, there is an urgent need for an integrated loan limit management system that combines dynamic questionnaires, intelligent decision-making, and multi-protocol adaptation. Summary of the Invention

[0009] The purpose of this invention is to solve the problems mentioned above in the background art, and to propose a loan limit management system, computer equipment and storage medium.

[0010] The objective of this invention can be achieved through the following technical solutions:

[0011] A loan limit management system, comprising:

[0012] A dynamic questionnaire engine for generating scalable question trees and collecting user data;

[0013] The rule-based decision cluster connects to the questionnaire engine output and executes access control, quota calculation, and product matching.

[0014] The risk pricing gateway connects to the output of the rule-based decision cluster and implements dynamic risk compensation.

[0015] A bank interface adapter connects to a risk pricing gateway to enable multi-bank protocol conversion;

[0016] The decision feedback module connects to the bank's interface adapter and generates a multi-channel output scheme.

[0017] As a further aspect of the present invention: the dynamic questionnaire engine includes:

[0018] The problem tree storage unit stores the basic problem set and dynamically loaded rules;

[0019] The conditional injection unit adds an association question when a specific response is detected;

[0020] The response verification unit verifies the legality and logical consistency of input values ​​in real time.

[0021] As a further aspect of the present invention: the rule-based decision cluster includes:

[0022] The access control module implements a two-layer verification process, including both hard and flexible conditions.

[0023] The quota calculation module includes a weight configuration unit and a scoring mapping unit;

[0024] The product matching module implements a three-level filtering strategy.

[0025] As a further aspect of the present invention: the first-level filter of the product matching module classifies the product pool according to the purpose of funds, the second-level filter classifies the product type according to the asset holding status, and the third-level filter allocates specific bank products according to user ratings.

[0026] As a further aspect of the present invention: the decision feedback module includes:

[0027] Multi-channel output unit, supporting simultaneous output of API / SMS / WeChat / email;

[0028] The scheme optimization unit generates suggested strategies for increasing credit limits based on missing conditions.

[0029] As a further aspect of the present invention: the loan limit management system further includes:

[0030] Real-time monitoring dashboard, visually displaying TPS, latency, and accuracy metrics;

[0031] The automatic alarm unit triggers a multi-level alarm mechanism when the system malfunctions.

[0032] As a further aspect of the present invention: the risk pricing gateway includes:

[0033] The four-dimensional risk control calculator uses the formula RiskScore = 0.35 * Credit Score + 0.30 * Income + 0.25 * Assets + 0.10 * Education Level.

[0034] The dynamic compensation engine adjusts financial parameters based on occupation type and age range.

[0035] As a further aspect of the present invention: the bank interface adapter includes:

[0036] Protocol conversion unit supports multiple interface protocols such as SOAP / REST / FTP;

[0037] Asynchronous communication units achieve request-response decoupling through message queues;

[0038] The circuit breaker control unit switches to the local backup rule when the bank times out.

[0039] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement a component of the loan quota management system according to any one of claims 1 to 8.

[0040] A computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements a component of the loan quota management system as claimed in any one of claims 1 to 8.

[0041] The beneficial effects of this invention are:

[0042] Improve decision-making accuracy: The dynamic questionnaire engine improves data completeness through question tree expansion and real-time verification; three-level product filtering and four-dimensional risk scoring improve matching accuracy.

[0043] Optimize user experience: provide real-time feedback through multiple channels; the solution optimization unit offers actionable suggestions, improving user conversion rates.

[0044] Reduce operation and maintenance costs: The protocol conversion unit connects to multiple banks in a unified manner, improving development efficiency; the real-time monitoring dashboard enables fault location within minutes, improving operation and maintenance response speed. Attached Figure Description

[0045] The invention will now be further described with reference to the accompanying drawings.

[0046] Figure 1 It is a schematic diagram of the system framework of the present invention;

[0047] Figure 2 This is the three-level filtering framework in the product matching module of this invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Example 1, please refer to Figure 1-2As shown, the present invention is a loan limit management system, comprising:

[0050] A dynamic questionnaire engine for generating scalable question trees and collecting user data;

[0051] The rule-based decision cluster connects to the questionnaire engine output and executes access control, quota calculation, and product matching.

[0052] The risk pricing gateway connects to the output of the rule-based decision cluster and implements dynamic risk compensation.

[0053] A bank interface adapter connects to a risk pricing gateway to enable multi-bank protocol conversion;

[0054] The decision feedback module connects to the bank's interface adapter and generates a multi-channel output scheme.

[0055] Example 2, please refer to Figure 1-2 As shown, the present invention is a loan limit management system, comprising:

[0056] A dynamic questionnaire engine for generating scalable question trees and collecting user data;

[0057] The dynamic questionnaire engine includes: a question tree storage unit, which stores the basic question set and dynamic loading rules; a condition injection unit, which adds related questions when a specific response is detected; and a response verification unit, which verifies the legality and logical consistency of the input values ​​in real time.

[0058] The rule-based decision cluster connects to the questionnaire engine output and executes access control, quota calculation, and product matching.

[0059] The rule-based decision cluster includes: an admission control module that implements dual-layer verification of hard and flexible conditions; a quota calculation module that includes a weight configuration unit and a scoring mapping unit; and a product matching module that implements a three-level filtering strategy.

[0060] The risk pricing gateway connects to the output of the rule-based decision cluster and implements dynamic risk compensation.

[0061] The risk pricing gateway includes: a four-dimensional risk control calculator using the formula RiskScore = 0.35 * Credit Score + 0.30 * Income + 0.25 * Assets + 0.10 * Education Level; and a dynamic compensation engine that adjusts financial parameters based on occupation type and age range.

[0062] Credit limit calculation module:

[0063] Weighting configuration unit: Stores a weight matrix = {Income: 1.5, Education: 0.8, Real Estate: 2.2}

[0064] Rating mapping unit: realizes the conversion of rating to credit limit coefficient.

[0065] A bank interface adapter connects to a risk pricing gateway to enable multi-bank protocol conversion;

[0066] The bank interface adapter includes: a protocol conversion unit that supports multiple interface protocols such as SOAP / REST / FTP; an asynchronous communication unit that decouples requests and responses through a message queue; and a circuit breaker control unit that switches to local backup rules when the bank times out.

[0067] The decision feedback module connects to the bank interface adapter and generates multi-channel output solutions.

[0068] The decision feedback module includes: a multi-channel output unit that supports simultaneous output via API / SMS / WeChat / email; and a scheme optimization unit that generates credit limit increase suggestions based on missing conditions.

[0069] Example 3, please refer to Figure 1-2 As shown, the present invention is a loan limit management system, comprising:

[0070] A dynamic questionnaire engine for generating scalable question trees and collecting user data;

[0071] The dynamic questionnaire engine includes: a question tree storage unit, which stores the basic question set and dynamic loading rules; a condition injection unit, which adds related questions when a specific response is detected; and a response verification unit, which verifies the legality and logical consistency of the input values ​​in real time.

[0072] The rule-based decision cluster connects to the questionnaire engine output and executes access control, quota calculation, and product matching.

[0073] The rule-based decision cluster includes: an access control module, which implements a two-layer verification of both hard and flexible conditions; a credit limit calculation module, which includes a weight configuration unit and a scoring mapping unit; and a product matching module, which implements a three-level filtering strategy. The product matching module's first-level filter categorizes product pools according to the purpose of funds, the second-level filter classifies product types according to asset holdings, and the third-level filter allocates specific bank products based on user scores.

[0074] The risk pricing gateway connects to the output of the rule-based decision cluster and implements dynamic risk compensation.

[0075] The risk pricing gateway includes: a four-dimensional risk control calculator using the formula RiskScore = 0.35 * Credit Score + 0.30 * Income + 0.25 * Assets + 0.10 * Education Level; and a dynamic compensation engine that adjusts financial parameters based on occupation type and age range.

[0076] A bank interface adapter connects to a risk pricing gateway to enable multi-bank protocol conversion;

[0077] The bank interface adapter includes: a protocol conversion unit that supports multiple interface protocols such as SOAP / REST / FTP; an asynchronous communication unit that decouples requests and responses through a message queue; and a circuit breaker control unit that switches to local backup rules when the bank times out.

[0078] The decision feedback module connects to the bank interface adapter and generates multi-channel output solutions.

[0079] The decision feedback module includes: a multi-channel output unit that supports simultaneous output via API / SMS / WeChat / email; and a scheme optimization unit that generates credit limit increase suggestions based on missing conditions.

[0080] Example 4, please refer to Figure 1-2 As shown, the present invention is a loan limit management system, comprising:

[0081] A dynamic questionnaire engine for generating scalable question trees and collecting user data;

[0082] The dynamic questionnaire engine includes: a question tree storage unit, which stores the basic question set and dynamic loading rules; a condition injection unit, which adds related questions when a specific response is detected; and a response verification unit, which verifies the legality and logical consistency of the input values ​​in real time.

[0083] The rule-based decision cluster connects to the questionnaire engine output and executes access control, quota calculation, and product matching.

[0084] The rule-based decision cluster includes: an access control module, which implements a two-layer verification of both hard and flexible conditions; a credit limit calculation module, which includes a weight configuration unit and a scoring mapping unit; and a product matching module, which implements a three-level filtering strategy. The product matching module's first-level filter categorizes product pools according to the purpose of funds, the second-level filter classifies product types according to asset holdings, and the third-level filter allocates specific bank products based on user scores.

[0085] The risk pricing gateway connects to the output of the rule-based decision cluster and implements dynamic risk compensation.

[0086] The risk pricing gateway includes: a four-dimensional risk control calculator using the formula RiskScore = 0.35 * Credit Score + 0.30 * Income + 0.25 * Assets + 0.10 * Education Level; and a dynamic compensation engine that adjusts financial parameters based on occupation type and age range.

[0087] A bank interface adapter connects to a risk pricing gateway to enable multi-bank protocol conversion;

[0088] The bank interface adapter includes: a protocol conversion unit that supports multiple interface protocols such as SOAP / REST / FTP; an asynchronous communication unit that decouples requests and responses through a message queue; and a circuit breaker control unit that switches to local backup rules when the bank times out.

[0089] The decision feedback module connects to the bank interface adapter and generates multi-channel output solutions.

[0090] The decision feedback module includes: a multi-channel output unit that supports simultaneous output via API / SMS / WeChat / email; and a scheme optimization unit that generates credit limit increase suggestions based on missing conditions.

[0091] The loan limit management system also includes: a real-time monitoring dashboard that visually displays TPS, latency, and accuracy indicators; and an automatic alarm unit that triggers a multi-level alarm mechanism when the system malfunctions.

[0092] In another embodiment, a computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement a component of the loan limit management system of any one of claims 1 to 8.

[0093] In another embodiment, a computer-readable storage medium storing a computer program is characterized in that, when the computer program is executed by a processor, it implements a component of the loan limit management system as claimed in any one of claims 1 to 8.

[0094] This application should also include the following:

[0095] according to Figure 1 The specific steps are as follows:

[0096] Step 1: User Input and Questionnaire Generation

[0097] Client interface: Receives user loan application requests (such as input from a webpage / APP) and transmits the information to the dynamic questionnaire engine.

[0098] Call the user profile database (which stores historical data) to pre-fill basic information, and then transmit the information to the dynamic questionnaire engine;

[0099] A question tree is generated based on dynamically loaded rules from the rule database (e.g., if "owns property" is detected, an additional valuation question is added).

[0100] Step 2: Rule-based decision-making and product matching

[0101] Rule-based decision clusters:

[0102] Access control module: Verifies hard criteria (credit history) and flexible criteria (income-to-debt ratio);

[0103] Credit limit calculation module: Calculates a score based on a weight matrix (e.g., {income: 1.5, education: 0.8});

[0104] Product matching module: Uses product knowledge base to perform three-level filtering.

[0105] Step 3: Risk Optimization and Bank Integration

[0106] Risk pricing gateway:

[0107] The risk score is calculated using the formula: RiskScore = 0.35 * Credit Score + 0.30 * Income + ...

[0108] Dynamic compensation (such as increased interest rates for freelancers).

[0109] Bank interface adapter:

[0110] Protocol conversion (e.g., internal JSON → bank SOAP);

[0111] Decouple requests through asynchronous communication;

[0112] When the circuit breaker times out, switch to the local backup rule.

[0113] Step 4: Results Feedback and Output

[0114] Decision feedback module:

[0115] Synchronously return results by calling multi-channel output interfaces (API / SMS / WeChat);

[0116] Generate optimization suggestions (such as "Increase credit limit by supplementing transaction records").

[0117] Key logical closed loop: User input → Dynamic questionnaire collection → Rule-based decision-making (admission / credit limit / matching) → Risk pricing and compensation → Bank agreement conversion → Multi-channel feedback. External systems (user profile database, rule database, bank) provide real-time data support for each module.

[0118] according to Figure 2 As shown, the first-level filter categorizes product pools by the purpose of funding.

[0119] Input: User's intended use of funds (e.g., "renovation").

[0120] Logic: Filter product pools that match the intended use from the product knowledge base (e.g., "home improvement product pool").

[0121] Secondary filtering: Classification by asset holding type

[0122] Input: User's asset information (e.g., real estate holdings).

[0123] logic:

[0124] If you own property, you can apply for mortgage-backed products (such as mortgage loans).

[0125] If you do not own property, you can apply for credit-based products (such as unsecured personal loans).

[0126] Three-level filtering: Assigning specific products based on user ratings.

[0127] Input: User's overall score (from the credit limit calculation module).

[0128] logic:

[0129] Score > 80 → Allocate high-quality products (such as low interest rates and high credit limits);

[0130] Example: User with a score of 85 → Matched with "XX Bank Home Renovation Mortgage Loan (Interest Rate 4.5%)".

[0131] Output: Issuance of specific bank products.

[0132] In another embodiment, the system workflow is as follows:

[0133] Dynamic questionnaire engine launched:

[0134] The problem tree storage unit loads the basic problem set (such as income, debt, occupation);

[0135] When the condition injection unit detects that the user selects "owns property", it automatically adds related questions such as "property valuation" and "mortgage status".

[0136] The response verification unit verifies the input in real time (e.g., income ≥ 0, age 18-65) and intercepts logically conflicting data (e.g., "student" occupation but annual income 500,000).

[0137] Rule-based decision cluster processing:

[0138] Access control module:

[0139] Mandatory requirements: No negative credit history and age compliance;

[0140] Flexible conditions: The debt-to-income ratio of less than 70% can be relaxed to 75%.

[0141] Credit limit calculation module:

[0142] The weighted configuration unit calls the matrix {Income: 1.5, Education: 0.8, Real Estate: 2.2};

[0143] The rating mapping unit converts the comprehensive rating into a credit limit coefficient (e.g., rating 80 → coefficient 1.2).

[0144] Product matching module:

[0145] Primary filter: Classified by purpose of funds (e.g., consumer loans / business loans);

[0146] Secondary filter: Classified by asset holding (whether or not there is collateral);

[0147] Three-level filter: Specific products are assigned based on the score (score > 85 matches high-amount, low-interest-rate products).

[0148] Risk pricing gateway optimization:

[0149] Formula for the 4D Risk Control Calculator:

[0150] RiskScore = 0.35 * Credit Score + 0.30 * Income Score + 0.25 * Asset Score + 0.10 * Education Score;

[0151] Dynamic compensation engine: Increase interest rates by 0.5% for high-risk professions (such as freelancers), or reduce debt ratio requirements for users aged 55 and above.

[0152] Bank interface adapter integration:

[0153] The protocol conversion unit converts the internal JSON data into the SOAP / XML format required by the bank;

[0154] Asynchronous communication units decouple requests through Kafka queues to avoid system blockage due to bank response delays.

[0155] Circuit breaker control unit: When the bank interface times out (>3s), switch to the local backup rule to calculate the pre-approved credit limit.

[0156] Output of the decision feedback module:

[0157] Multi-channel output unit: AP I returns JSON to the front end, synchronizing the summary of SMS / WeChat push solutions;

[0158] Solution optimization unit: If a user is rejected due to "missing income proof", generate a suggestion to "upload tax bill to increase credit limit by 20%".

[0159] System monitoring and disaster recovery:

[0160] Real-time monitoring dashboard: tracks TPS (≥1000 / s), interface latency (<200ms), and rule accuracy (>98%);

[0161] Automatic alarm unit: If the error rate is greater than 5%, an alarm will be triggered via WeChat. If the error does not recover within 10 minutes, the maintenance department will be notified by phone.

[0162] Case 1: Mortgage Loan Application (Dynamic Questionnaire Engine + Product Matching)

[0163] User scenarios:

[0164] When Zhang applied for a business loan, he selected "owns property" on the initial questionnaire but did not fill in the valuation.

[0165] System execution process:

[0166] When the conditional injection unit detects a "owns property" response, it automatically adds a related question:

[0167] Property valuation range (>5 million / 2 million - 5 million / <2 million)

[0168] Mortgage status (mortgaged / unmortgaged)

[0169] The response verification unit discovered that Zhang filled in "estimated value of 6 million" but his occupation was "self-employed", triggering a logic verification (prompting "please provide property certificate").

[0170] Product matching module three-level filtering:

[0171] Level 1: Purpose of Funds = "Business Loan" → Select Product Pool A

[0172] Level 2: Asset Holding = "Unmortgaged Property" → Matching Mortgage-Backed Product B

[0173] Level 3: User rating 85 points → Assigned "XX Bank Mortgage Business Loan (Interest Rate 4.5%)"

[0174] result:

[0175] Traditional systems often result in incorrect product matching due to missing property valuations. This system improves data completeness by 40% and achieves a product matching accuracy of 98% by dynamically adding questions.

[0176] Case 2: Microloans for Freelancers (Risk Pricing + Solution Optimization)

[0177] User scenarios:

[0178] Li, a freelancer, earns 20,000 yuan a month, but cannot provide tax receipts, so the initial tax credit limit is only 50,000 yuan.

[0179] System execution process:

[0180] The four-dimensional risk control calculator generates a risk score:

[0181] RiskScore = 0.35 * 70 (Good Credit) + 0.30 * 65 (No Proof of Income) + 0.25 * 40 (No Assets) + 0.10 * 60 (Bachelor's Degree) = 62.75 → High Risk Level

[0182] Dynamic compensation engine activated:

[0183] Occupation type compensation: Freelancers will have their interest rate increased by 0.5%.

[0184] Age-based compensation (30 years old): Debt ratio requirement relaxed from 70% to 75%.

[0185] The solution optimization unit analyzes missing conditions and generates suggestions:

[0186] "Adding 6 months of bank statements can increase your credit limit to 80,000; uploading asset proof can help you get access to products with lower interest rates."

[0187] result:

[0188] The risk pricing gateway expanded the coverage of eligible loan users by 15% through dynamic compensation, and the optimization suggestions enabled Mr. Li to finally obtain a loan of 80,000 yuan (conversion rate increased by 25%).

[0189] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the scope of the claims.

Claims

1. A loan limit management system, characterized in that, include: A dynamic questionnaire engine for generating scalable question trees and collecting user data; The rule-based decision cluster connects to the questionnaire engine output and executes access control, quota calculation, and product matching. The risk pricing gateway connects to the output of the rule-based decision cluster and implements dynamic risk compensation. A bank interface adapter connects to a risk pricing gateway to enable multi-bank protocol conversion; The decision feedback module connects to the bank's interface adapter and generates a multi-channel output scheme.

2. The loan limit management system according to claim 1, characterized in that, The dynamic questionnaire engine includes: The problem tree storage unit stores the basic problem set and dynamically loaded rules; The conditional injection unit adds an association question when a specific response is detected; The response verification unit verifies the legality and logical consistency of input values ​​in real time.

3. A loan limit management system according to claim 1, characterized in that, The rule-based decision cluster includes: The access control module implements a two-layer verification process, including both hard and flexible conditions. The quota calculation module includes a weight configuration unit and a scoring mapping unit; The product matching module implements a three-level filtering strategy.

4. A loan limit management system according to claim 3, characterized in that, The product matching module has a first-level filter that categorizes product pools according to the purpose of funds, a second-level filter that classifies product types according to asset holdings, and a third-level filter that assigns specific bank products based on user ratings.

5. A loan limit management system according to claim 1, characterized in that, The decision feedback module includes: Multi-channel output unit, supporting simultaneous output of API / SMS / WeChat / email; The scheme optimization unit generates suggested strategies for increasing credit limits based on missing conditions.

6. A loan limit management system according to claim 1, characterized in that, The loan limit management system also includes: Real-time monitoring dashboard, visually displaying TPS, latency, and accuracy metrics; The automatic alarm unit triggers a multi-level alarm mechanism when the system malfunctions.

7. A loan limit management system according to claim 1, characterized in that, The risk pricing gateway includes: The four-dimensional risk control calculator uses the formula RiskScore = 0.35 * Credit Score + 0.30 * Income + 0.25 * Assets + 0.10 * Education Level; The dynamic compensation engine adjusts financial parameters based on occupation type and age range.

8. A loan limit management system according to claim 1, characterized in that, The bank interface adapter includes: Protocol conversion unit supports multiple interface protocols such as SOAP / REST / FTP; Asynchronous communication units achieve request-response decoupling through message queues; The circuit breaker control unit switches to the local backup rule when the bank times out.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that... When the processor executes the computer program, it implements a step of the loan quota management system according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that... When the computer program is executed by the processor, it implements the steps of the loan quota management system as described in any one of claims 1 to 8.

Images