Financial business process management system and method, storage medium and electronic equipment

Abstract

The invention discloses a financial business process management system and method, a storage medium and electronic equipment, and the method comprises the steps: building a product recommendation model based on a bank admission rule base; the dynamic allocation engine is used for allocating signing coordination personnel by adopting a multi-strategy rotation algorithm; the desensitization data reconciliation module supports fuzzy matching of various field combinations; the post-loan risk control platform integrates voucher auditing, channel risk grading and merchant blacklist linkage functions; the examination supervision module is used for realizing screen switching prevention monitoring and abnormal behavior detection; the activity management unit is used for completing closed-loop management of issuing and cancelling after verification of the immediate reduction money; the channel evaluation matrix is used for calculating a five-dimensional risk index including the expenditure rate deviation degree and reject ratio estimation; the input tracking system updates the bank approval progress in real time and synchronizes the bank approval progress to the business end; and the data center stores customer portraits, the approval knowledge base and third-party credit investigation data in a unified manner, and the financial business process management system greatly reduces account checking time consumption.

Description

TECHNICAL FIELD

[0001] The present application relates to the cross field of financial technology and risk management, in particular to a financial business process management system and method. BACKGROUND

[0002] There are three technical bottlenecks in traditional financial business management: low efficiency of manual matching; weak prevention and control of moral risk; and lagging response of risk control. Therefore, the present application provides a financial business process management system and method to solve the problems in the background art. SUMMARY

[0003] To solve the above technical problems, the present application provides a financial business process management system, comprising:

[0004] A product matching module based on a bank access rule library to construct a product recommendation model;

[0005] A dynamic allocation engine using a multi-strategy rotation algorithm to allocate signing coordination personnel;

[0006] A desensitization data reconciliation module supporting fuzzy matching of various field combinations;

[0007] A post-loan risk control platform integrating certificate review, channel risk grading, and merchant blacklist linkage functions;

[0008] An examination supervision module to realize anti-screen cutting monitoring and abnormal behavior detection;

[0009] An activity management unit to complete the closed-loop management of the issuance and cancellation of the reduction fund;

[0010] A channel evaluation matrix to calculate a multi-dimensional risk index including usage rate deviation and non-performing rate estimation;

[0011] An application tracking system to update the bank approval progress in real time and synchronize it to the business end;

[0012] A data center to store customer portraits, approval knowledge base, and third-party credit data uniformly.

[0013] Preferably, the dynamic allocation engine comprises:

[0014] A first allocation strategy: average allocation according to historical order quantity;

[0015] A second allocation strategy: peak load allocation based on time period;

[0016] A third allocation strategy: targeted allocation of key personnel for high-risk applications.

[0017] Fourth distribution strategy: random distribution when triggering distribution limit conditions.

[0018] Preferably, the limit conditions include:

[0019] The single-day distribution amount exceeds the threshold N;

[0020] The signing coordinator is on leave;

[0021] The same salesman has been continuously X single without changing the coordinator;

[0022] The feedback delay time exceeds the set threshold T.

[0023] Preferably, the AI scoring system includes a double-layer architecture:

[0024] Static rule layer: realizes product access preliminary screening based on if-then rules;

[0025] Dynamic learning layer: combines third-party data and bank feedback results through XGBoost model. Preferably, the model optimization of the dynamic learning layer adopts SHAP value analysis, specifically including: updating feature importance ranking every month;

[0026] When the bank approval result and the prediction difference ≥ 15%, trigger model retraining;

[0027] Set an upper limit on the weight of manual intervention for key features.

[0028] A business management method, comprising the steps of:

[0029] S1. Channel personnel scan the two-dimensional code to submit customer qualification data package;

[0030] S2. The product matching module outputs three optimal bank product schemes;

[0031] S3. The dynamic distribution engine selects the signing coordinator and encrypts the communication;

[0032] S4. Bank signing data is synchronized to the reconciliation module to complete automatic matching;

[0033] S5. The post-loan certificate audit result is fed back to the channel risk assessment model.

[0034] Preferably, the product matching in step S2 includes:

[0035] Primary matching: selects qualified products based on customer explicit qualifications;

[0036] Deep matching: combines historical approval pass rate to calculate recommendation priority.

[0037] Preferably, the channel risk assessment of step S5 includes:

[0038] The first dimension: suspicious credential proportion;

[0039] The second dimension: standard deviation of industry usage rate;

[0040] The third dimension: blacklisted merchant correlation degree;

[0041] The fourth dimension: bank bad data feedback rate;

[0042] The fifth dimension: customer qualification deviation index.

[0043] A computer-readable storage medium stores program code for implementing the business management method described above.

[0044] An electronic device includes a processor and the storage medium described above.

[0045] Technical effects and advantages of the present application:

[0046] The financial business process management system in the present application greatly reduces the account time, further improves the channel risk identification accuracy, and reduces the signing distribution complaint rate. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is a schematic diagram of the financial business process management system provided by the embodiments of the present application;

[0048] Figure 2 is a flowchart of the financial business process management method provided by the embodiments of the present application; DETAILED DESCRIPTION

[0049] The present application will be further described in detail below in conjunction with the drawings and specific embodiments. The embodiments of the present application are given for the purpose of illustration and description, and are not exhaustive or limit the present application to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present application, and to enable those of ordinary skill in the art to understand the present application so as to design various embodiments with various modifications suitable for specific purposes.

[0050] Please refer to Figure 1 In the present embodiment, a financial business process management system is provided, comprising:

[0051] A product matching module constructs a product recommendation model based on a bank access rule library;

[0052] A dynamic allocation engine allocates signing coordination personnel using a multi-strategy rotation algorithm;

[0053] A desensitization data reconciliation module supports fuzzy matching of various field combinations;

[0054] A post-loan risk control platform integrating certificate review, channel risk grading, and merchant blacklist linkage functions;

[0055] An examination supervision module that realizes anti-screen cutting monitoring and abnormal behavior detection;

[0056] An event management unit that completes the closed-loop management of the issuance and cancellation of the reduction fund;

[0057] A channel evaluation matrix that calculates a multi-dimensional risk index including usage rate deviation and bad rate estimation;

[0058] An application tracking system that updates the bank approval progress in real time and synchronizes it to the business end;

[0059] A data center that stores customer portraits, approval knowledge bases, and third-party credit data.

[0060] Further: the dynamic allocation engine includes:

[0061] First allocation strategy: average allocation according to historical order volume;

[0062] Second allocation strategy: peak load allocation based on time period;

[0063] Third allocation strategy: targeted allocation for high-risk application key personnel;

[0064] Fourth allocation strategy: random allocation when allocation limit conditions are triggered.

[0065] Further: the limit conditions include:

[0066] Daily allocation volume exceeds threshold N;

[0067] The signing coordinator is on leave;

[0068] The same salesperson has X consecutive orders without changing the coordinator;

[0069] The feedback delay time exceeds the set threshold T.

[0070] Further: the desensitization data reconciliation module uses an improved algorithm, which includes:

[0071] The name matching weight coefficient is set to 0.6;

[0072] The last four digits of the mobile phone number matching weight coefficient is set to 0.3;

[0073] The ID number birthday segment matching weight coefficient is set to 0.4.

[0074] Further: the model optimization of the dynamic learning layer uses SHAP value analysis, which includes: updating the feature importance ranking every month;

[0075] Model retraining is triggered when the bank approval result and the prediction differ by ≥15%;

[0076] Set an upper limit on the manual intervention weight for key features.

[0077] Please refer to Figure 2 A business management method, comprising the steps of:

[0078] S1. Channel personnel scan the two-dimensional code to submit customer qualification data package;

[0079] S2. The product matching module outputs three optimal bank product schemes;

[0080] S3. The dynamic allocation engine selects a signing coordinator and encrypts the communication;

[0081] S4. Bank signing data is synchronized to the reconciliation module to complete automated matching;

[0082] S5. The post-loan certificate audit result is fed back to the channel risk assessment model.

[0083] Further: In step S2, product matching includes:

[0084] Primary matching: Select qualified products based on customer explicit qualifications;

[0085] Deep matching: Calculate the recommendation priority in combination with the historical approval rate.

[0086] Further: The channel risk assessment in step S5 includes:

[0087] First dimension: Suspicious certificate proportion;

[0088] Second dimension: Branch industry standard deviation;

[0089] Third dimension: Blacklist merchant correlation;

[0090] Fourth dimension: Bank bad data feedback rate;

[0091] Fifth dimension: Customer qualification deviation index.

[0092] A computer-readable storage medium storing program code for implementing the above-mentioned business management method.

[0093] An electronic device comprising a processor and the above-mentioned storage medium.

[0094] In a specific embodiment, the financial business process management system, according to the business needs of independent research and development of customer relationship management system, through the system management mode to regulate business processes, provide business support, control business risk, strengthen the company management. The system has supported installment / consumption credit and housing loan business. The entire system function includes transaction management, product management, bank management, channel management, signing management, city and subject management, user rebate and file management, reconciliation management, post-loan certificate management, test management, etc. The main functions of each block are as follows:

[0095] I. Product management

[0096] Establish cooperative installment or loan products, according to the characteristics of different products into the corresponding category (car installment / home decoration installment / large installment / housing loan, etc.), and can decide whether to enable risk prompt function according to product.

[0097] II. Bank management

[0098] Establish relevant business and product cooperation banks and branches to facilitate business staff to submit, back office summary and reconciliation, etc.

[0099] III. Channel management

[0100] Manage the cooperation channel, establish the basic information of the channel, and associate the cooperation bank / branch and the submission product of the channel, establish the rebate rules and coefficients of the channel, bind the channel rebate receiving personnel, and open the submission entry two-dimensional code for the channel personnel.

[0101] IV. Submission and signing management

[0102] Submit the intended customer by the business staff, enter the basic information and qualification of the installment / loan customer, and the system matches different signing coordination personnel according to certain rules (currently according to the signing product), the signing coordination personnel receives the submission and conducts preliminary examination on the customer's qualification, and after the preliminary examination is passed, the customer's basic information is sent to the designated bank's interface signing personnel, the bank personnel contacts the customer for signing. The signing coordination personnel follows up the signing and approval progress of the customer and updates the progress in the system, and the business staff can receive message prompts and real-time control the customer's approval progress to better serve the customer.

[0103] V. Reconciliation management

[0104] Manage the reconciliation data feedback by the bank, and bind it with the system to make it easier for the company and business staff to track the customer's progress and manage the customer.

[0105] Among them, the system provides the "account matching" function to solve the problem of difficult account reconciliation of desensitized data (such as customer name, mobile phone number or ID card number data) feedback by the bank. It can automatically match the data in the system according to the fields of desensitized data (provide only name, name + mobile phone number, name + ID card number, name + card number, name + mobile phone number + ID card number, name + mobile phone number + card number, name + source, etc. Matching method), which can effectively reduce the manual matching time and improve the reconciliation efficiency.

[0106] Six, post-loan management

[0107] It includes post-loan certificate uploading reminder, post-loan certificate uploading, post-loan certificate audit management, risk early warning and other functions, which can effectively control the customer, source and transaction risk. For the products that require loan after the company, the system will remind the business staff to upload the post-loan certificate according to the post-loan certificate uploading state. After the business staff uploads the post-loan certificate, the risk control management personnel of our company will receive the audit notice and audit the post-loan certificate, and give the audit conclusion of compliance, risk, non-compliance, etc. And regularly statistics and report to the company for further analysis of the risk of more merchants, business staff, etc.

[0108] Seven, examination management

[0109] The system establishes an examination system, including question bank management, test paper management and examination management. It is convenient to regularly examine the business staff and related personnel in product, compliance and other aspects. The system automatically scores and calculates the pass rate. The anti-screen cutting setting can effectively prevent the business staff from searching for data, looking up answers and communicating with others during the examination process.

[0110] Eight, activity management

[0111] According to the activities carried out by the company, the activity sign-in personnel management, activity participant management, and discount gold payment management are carried out to meet the needs of various business activities of the company.

[0112] In another embodiment,

[0113] Intelligent product matching and quota calculation: the loan access requirements of each bank are different. In the past, there were few banks and products, and personnel training was relied on word of mouth. However, with the increasing number of loan banks and products and the increasing number of business staff, manual training is not very reliable. Therefore, we establish a product recommendation model in the system to match products and calculate quotas by establishing the access requirements and quota calculation rules of each product. The product matching accuracy can reach about 75% (the construction of this model is based on bank access rules and continuously obtains internal approval rules to improve the model, which is relatively static).

[0114] Intelligent signing allocation function: One of the major management difficulties of loan docking is docking chaos. Businessmen are easy to collude with familiar client managers, causing a series of problems such as fraud, and the bank management institution has low control over the business below. Therefore, we introduce the signing coordination function to establish a "firewall" between the business staff and the bank client managers. The business staff submits the application, which is pre-audited by the system and then allocated to the signing coordination group (currently 5-6 people). The system intelligently allocates according to the rules and algorithms constructed, balancing the workload among the signing coordination personnel on the one hand, and making it difficult for business staff to collude with signing coordination personnel on the other hand, thus establishing a business security barrier.

[0115] The main allocation and algorithm strategy currently used: According to the available signing coordination personnel for the specified product (1) average rotation allocation; (2) time period allocation strategy; (3) key personnel allocation strategy, but when the allocation limit condition is triggered (including the number of single allocation in a certain period reaching the limit, personnel on leave, personnel accepting orders and feedback time limit extension, the signing coordination personnel for the business staff submitted application has not changed for xx consecutive times, etc.), secondary personnel, there are many, can continue to use (1) (3) (4) allocation strategy; (4) random allocation strategy (the system randomly selects available signing coordination personnel within the limited personnel workload).

[0116] Intelligent post-loan risk control platform: Post-loan management and fund security are the top priority for banks. Currently, the company has an independent internal control and compliance department responsible for post-loan management. The system has a built-in post-loan certificate uploading system. Business staff need to continuously track customer use and upload POS single, payment proof and other related certificates. Through manual review, the authenticity and reasonableness of the use are determined, and normal, suspicious, and illegal use are judged. The system will automatically analyze the risk index of the application channel based on the submitted data and divide it into four grades:

low

medium

high

very high

medium

very high

[0117] (1) The suspicious and illegal proportion of channel application post-loan certificates;

[0118] (2) The change trend of channel application, loan amount, and use rate, and the comparison with the same industry merchants;

[0119] (3) Channel application customer use merchant distribution and merchant risk analysis (merchant blacklist system);

[0120] (4) Estimate the bad rate of the channel through bank feedback bad data;

[0121] (5) Channel entry customer profile, qualifications, income and compared with the same industry deviation.

[0122] Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art and related fields without creative labor shall belong to the scope of protection of the present application. The structures, devices and operation methods not specifically described and explained in the present application are implemented according to the conventional means in the art, unless otherwise specified and limited.

Claims

1. A financial services process management system, characterized by, Comprising, a product matching module that builds a product recommendation model based on a bank access rule library; a dynamic allocation engine that allocates signing coordinators using a multi-strategy rotation algorithm; a desensitization data reconciliation module that supports fuzzy matching of various field combinations; a post-loan risk control platform that integrates certificate review, channel risk grading, and merchant blacklist linkage functions; an examination supervision module that realizes anti-screen cutting monitoring and abnormal behavior detection; an event management unit that completes the closed-loop management of the issuance and cancellation of the reduction fund; a channel evaluation matrix that calculates a multi-dimensional risk index including a deviation degree of use rate and an estimated bad rate; an application tracking system that updates the bank approval progress in real time and synchronizes it to the business end; a data middle platform that stores customer portraits, approval knowledge bases, and third-party credit data.

2. The financial process management system of claim 1, wherein The dynamic allocation engine includes: First allocation strategy: allocate according to the average historical order volume; Second allocation strategy: allocate based on peak load in time periods; Third allocation strategy: targeted allocation for key personnel of high-risk applications; Fourth allocation strategy: random allocation when triggering allocation limit conditions.

3. The financial process management system of claim 1, wherein The limit conditions include: Daily allocation volume exceeds threshold N; The signing coordinator is on leave; The same salesperson has not changed the coordinator for X orders in a row; Feedback delay time exceeds the set threshold T.

4. The financial process management system of claim 1, wherein The AI scoring system includes a double-layer architecture: Static rule layer: implements product access preliminary screening based on if-then rules; Dynamic learning layer: integrates third-party data and bank feedback results through an XGBoost model.

5. The financial process management system of claim 1, wherein The model optimization of the dynamic learning layer uses SHAP value analysis, which includes: Update feature importance ranking every month; Trigger model retraining when the bank approval result and prediction difference ≥ 15%; Set an upper limit on the weight of key features for manual intervention.

6. A service management method characterized by, The steps include: S1. Channel personnel scan the two-dimensional code to submit customer qualification data package; S2. The product matching module outputs three optimal bank product schemes; S3. The dynamic allocation engine selects the signing coordinator and encrypts the communication; S4. Bank signing data is synchronized to the reconciliation module to complete automatic matching; S5. The post-loan certificate review results are fed back to the channel risk assessment model.

7. The service management method according to claim 6, characterized by, The product matching in step S2 includes: Primary matching: filter eligible products based on customer explicit qualifications; Deep matching: calculate recommendation priority based on historical approval pass rate.

8. The service management method according to claim 6, characterized by, The channel risk assessment in step S5 includes: First dimension: suspicious certificate proportion; Second dimension: industry standard deviation of use rate; Third dimension: blacklisted merchant correlation degree; Fourth dimension: bank bad data feedback rate; Fifth dimension: customer qualification deviation index.

9. A computer-readable storage medium, characterized in that, The storage medium stores program code for implementing the business management method of any one of claims 7-9.

10. An electronic device comprising a processor and the storage medium of claim 9.

Images