Credit card financial contract management method and system
Through biometric technology and blockchain and other means, the problems of identity verification and automated management in traditional credit card financial contract management have been solved, and higher security and efficiency have been achieved, adapting to complex market environments and user needs.
Patent Information
- Application Number
- CN202411139958.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-20
AI Technical Summary
The traditional credit card financial contract management method relies on static passwords and security issues in identity verification, and is vulnerable to phishing and data breaches, making it difficult to ensure the uniqueness and authenticity of user identities, and lacks automation and transparency, so it is impossible to effectively deal with complexity. Changeable market environment and user needs.
Verify user identity through biometric technology, collect user data to calculate credit scores, and generate smart financial contracts, use blockchain for secure storage and access control, combine multi-scale decomposition technology, image enhancement and GAN model for camouflage detection, and use Logistic regression The model dynamically adjusts the credit score, uses a linear planning model to optimize the financial contract combination strategy, and provides a visual interface to display financial contracts.
Improve the security of identity verification and the real-time and accuracy of credit scores, automate the financial contract management process, formulate the optimal financial contract combination strategy, improve the financial management efficiency of users, and effectively cope with complex and changing markets. Environment and user needs.
Smart Images

Figure CN118657609B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of contract management, and in particular to a credit card financial contract management method and system. Background Art
[0002] With the rapid development of information technology and Internet finance, the credit card market has ushered in unprecedented opportunities and challenges. Credit card financial contract management methods have become an indispensable part of the financial industry, providing users with convenient consumption methods and flexible means of fund management. In the past decade, the development of credit card technology has undergone a transformation from traditional offline services to full digitalization, involving the widespread application of emerging technologies such as online payment, cardless transactions, and blockchain-based smart contract management. At this stage, many financial institutions have gradually adopted biometric technology to enhance the security of user identity authentication, and more and more companies have begun to introduce big data and artificial intelligence technologies to more accurately conduct user credit scoring and risk assessment.
[0003] However, the existing technology still has many shortcomings. First, the traditional credit card financial contract management method relies on static passwords and security issues in identity authentication, which is vulnerable to phishing and data leakage, and it is difficult to ensure the uniqueness and authenticity of user identity. The existing financial contract management method also often relies on manual processing, lacks automation and transparency, and cannot effectively respond to the complex and changing market environment and user needs. Summary of the invention
[0004] In view of the problems existing in the above-mentioned existing credit card financial contract management method and system, the present invention is proposed.
[0005] Therefore, the problem to be solved by the present invention is that the traditional credit card financial contract management method relies on static passwords and security issues in identity authentication, is vulnerable to the threat of phishing and data leakage, and is difficult to ensure the uniqueness and authenticity of the user identity. The existing financial contract management method also often relies on manual processing, lacks automation and transparency, and cannot effectively respond to the complex and changing market environment and user needs.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a credit card financial contract management method, which includes: verifying the identity of registered users through biometric technology, collecting user data to calculate user credit scores and dynamically adjusting credit scores, generating preliminary financial contracts and deploying them on the blockchain to form smart financial contracts; creating smart financial contract accounts to set billing cycles and account types, performing combined analysis on multiple credit card financial contracts held by users, and formulating the optimal financial contract combination strategy; securely storing financial contracts and implementing access control, and displaying financial contracts through a visual interface.
[0007] As a preferred solution of the credit card financial contract management method of the present invention, wherein: the verification of the identity of the registered user by biometric recognition technology refers to collecting the user's facial image and iris image when the user registers for a credit card, and marking each user's data with a unique identifier;
[0008] After preprocessing the image, the multi-scale decomposition technology is used to decompose the iris and facial images into multiple sub-images of different scales and the sub-images are enhanced;
[0009] Perform recognition processing on the enhanced image.
[0010] As a preferred solution of the credit card financial contract management method of the present invention, the method comprises: using a multi-scale decomposition technique after preprocessing the image to decompose the iris and facial image into a plurality of sub-images of different scales and performing image enhancement on the sub-images, including:
[0011] By formula: , perform image enhancement on the sub-image;
[0012] in, , where is the final enhanced image, N is the number of scales of multi-scale decomposition, and w A (I) is the weight coefficient at the Ath scale, α and β are adjustment parameters, γ A is the smoothing coefficient associated with the Ath scale, δ is a constant, M(I A ) is the result of multi-scale transformation, F A (x,y) is the enhancement function at the Ath scale, H A (x,y) is the spatial filter of scale A, is the gradient of the image at the Ath scale, is the local contrast enhancement function, is the local mean of the A-th scale image at position (x, y), is the local standard deviation, is a constant.
[0013] As a preferred solution of the credit card financial contract management method of the present invention, the recognition processing of the enhanced image refers to using the Sobel operator to perform preliminary edge detection on the enhanced image, identifying the edge features of the iris and pupil areas, and separating the iris and the background in the image by a dynamic circular convolution algorithm;
[0014] The pupil and iris boundaries are fitted using the Hough circle transform method, and the gradient of the iris image is calculated using the image gradient guided filter on the fitted iris image;
[0015] Use convolution operation to process the entire iris image to obtain the gradient value of each pixel, calculate the direction and magnitude of the gradient, and map the gradient magnitude and direction into a feature map. Each pixel corresponds to a set of gradient magnitude and direction values.
[0016] Use principal component analysis to encode and reduce the dimension of the extracted features, and input the reduced-dimensional features into a bidirectional long short-term memory network for time series analysis;
[0017] Use pre-trained convolutional neural network to extract facial deep features, encode and reduce the dimension of the extracted facial features, and perform weighted fusion of iris and facial features through multi-head self-attention mechanism;
[0018] Judging the authenticity of the sample based on the fused iris and facial features means building and training a GAN model, inputting the collected iris and facial features into the discriminator for real-time camouflage detection to obtain the authenticity probability value of the sample;
[0019] Set the judgment threshold Q and compare it with the authenticity probability value to judge the authenticity of the sample:
[0020] If the output authenticity probability value is greater than or equal to the judgment threshold Q, the sample is true, and the user is allowed to register a credit card through disguise detection;
[0021] If the output authenticity probability value is lower than the judgment threshold Q, the sample is disguised and cannot pass the disguise detection. A warning is issued and the user is notified to resample the image.
[0022] As a preferred solution of the credit card financial contract management method of the present invention, wherein: the collecting of user data, calculating the user credit score and dynamically adjusting the credit score, generating a preliminary financial contract and deploying it on the blockchain to form a smart financial contract refers to collecting multi-dimensional data of registered users and pre-processing the collected data, extracting key features from the data, including the user's income level, debt-to-income ratio, credit card utilization rate and credit history length, and normalizing different features;
[0023] Use the Logistic regression model as the credit scoring model, use historical user data as the training set to input the credit scoring model for iterative training, define the loss function to iteratively optimize the model parameters, and stop iterating when the loss of the credit scoring model no longer decreases significantly during continuous iterations, output the model parameters and update the credit scoring model;
[0024] Inputting user feature data into a credit scoring model to obtain a corresponding user credit score;
[0025] Set up to obtain new user data once a quarter, retrain the credit scoring model based on the latest user data and adjust the model parameters, and automatically adjust the credit score of each user according to the latest model;
[0026] Automatically generate financial contract terms based on the user's credit score, use Solidity to write smart financial contract code, embed the financial contract terms into the smart financial contract, select the Ethereum blockchain as the deployment platform, deploy the smart financial contract through the Ethereum mainnet and record the hash value of each financial contract on the chain.
[0027] As a preferred solution of the credit card financial contract management method described in the present invention, wherein: the creation of a smart financial contract account to set the billing cycle and account type refers to using a revolving credit account as the default account type, calculating the initial credit limit of each user based on the minimum credit limit set by the bank and the user's credit score, binding the user's initial credit limit and the user account and defining them as financial contract parameters, calling the API of the Ethereum platform to automatically deploy the smart financial contract to complete account creation and set the update cycle of the user's bill to monthly update.
[0028] As a preferred solution of the credit card financial contract management method of the present invention, wherein: the combination analysis of multiple credit card financial contracts held by the user and the formulation of the optimal financial contract combination strategy refers to collecting all credit card financial contract related data held by the user, including the credit limit, interest rate, annual fee, reward policy, usage frequency and historical payment record of each credit card, and integrating all the data into the central database after pre-processing;
[0029] Identify the features that influence the credit card portfolio, including interest rates, usage frequency, annual fees, and rewards policies, and quantify the features into feature vectors;
[0030] Use the linear programming model to optimize the financial contract portfolio and define the objective function with the financial contract costs and benefits as variables:
[0031] ,
[0032] In the formula, Z is the optimization target, the net income of the financial contract portfolio, z i is the annual expected return of the ith financial contract, u i is the interest rate of the ith financial contract, m i is the credit limit of the ith financial contract, f j is the annual fee of the jth financial contract, e represents the total number of financial contracts of the credit card held by the user, R i is the quantitative value of the reward policy, l i is the decision variable, l j is the selection variable of the jth financial contract, z0, F0 and R0 are standardization factors;
[0033] Define the constraints as: , where B is the upper limit of the total cost acceptable to the user, L is the minimum total benefit expected by the user, and l i is the decision variable, f i is the annual fee of the ith financial contract, C is the upper limit of the total credit limit acceptable to the user, N max The maximum number of contracts that the user is allowed to select; solve the model according to the linear programming algorithm and find the financial contract combination that maximizes the objective function as the optimal financial contract combination;
[0034] Select the credit card corresponding to the financial contract in the financial contract combination, combine the credit cards and recommend the best credit card combination to the user.
[0035] As a preferred solution of the credit card financial contract management method described in the present invention, wherein: the secure storage of financial contracts and implementation of access control refers to encrypting the financial contract combination, using the Advanced Encryption Standard for symmetric encryption and storing the hash value of the encrypted financial contract on the blockchain, using the RBAC mechanism, defining user roles and setting different permission levels according to business needs, allowing only authorized users to access the financial contract, logging each access and conducting regular audits.
[0036] As a preferred solution of the credit card financial contract management method described in the present invention, wherein: the display of the financial contract through a visual interface refers to using a front-end framework to develop a user-friendly visual interface, using a data visualization tool to display the key content and indicators of the financial contract, performing a financial contract decryption operation on a verified user request, and displaying the plain text content on the interface, and using charts and text boxes to display the financial contract change record in real time.
[0037] Another object of the present invention is to provide a credit card financial contract management method, which comprises:
[0038] The identity verification module is used to collect and process the user's facial and iris images to verify the authenticity of the user's identity;
[0039] The contract generation module is used to collect multi-dimensional data of verified users and calculate user scores, and automatically generate financial contract terms based on the user's credit score;
[0040] The contract combination module is used to analyze multiple credit card financial contracts held by the user and optimize the user's credit card financial contract combination based on the analysis results;
[0041] The storage and display module is used to securely store financial contract combinations and perform access control, and visually display the content of financial contracts to verified users.
[0042] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of a credit card financial contract management method when executing the computer program.
[0043] A computer-readable storage medium stores a computer program, which implements the steps of a credit card financial contract management method when executed by a processor.
[0044] The beneficial effects of the present invention are as follows: the present invention verifies the identity of registered users through biometric technology, collects user data, calculates user credit scores and dynamically adjusts credit scores, which not only improves the security of identity authentication, but also enhances the real-time and accuracy of credit scores, automates the management process of financial contracts, formulates the optimal financial contract combination strategy, improves the user's financial management efficiency, and can effectively respond to complex and changing market environments and user needs. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0046] Figure 1 The figure is a flowchart of a credit card financial contract management method.
[0047] Figure 2 This is a structural diagram of the credit card financial contract management system. DETAILED DESCRIPTION
[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.
[0049] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0050] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.
[0051] Example 1, reference Figure 1, which is the first embodiment of the present invention, and provides a credit card financial contract management method, the credit card financial contract management method comprising:
[0052] S1. Verify the identity of registered users through biometric technology, collect user data to calculate user credit scores and dynamically adjust credit scores, generate preliminary financial contracts and deploy them on the blockchain to form smart financial contracts;
[0053] Specifically, verifying the identity of a registered user through biometric technology means collecting the user's facial image and iris image when the user registers for a credit card, including using holographic projection technology to project a specific grating on the user's face to enhance the depth information of facial features, using multi-frame image synthesis technology to synthesize high-resolution facial images, using a holographic camera to shoot iris images at multiple angles to capture the 3D structural information of the iris, combining visible light and infrared light technology to reduce iris surface reflections, using a unique identifier to mark each user's data, automatically detecting the quality of the collected image through an image processing algorithm, and re-collecting biometric features if the image quality does not meet the standards, calculating the global illumination mean and standard deviation of the image, judging the suitability of illumination, and adjusting the initial parameters of the multi-scale Retinex algorithm based on the illumination evaluation results;
[0054] After preprocessing the image, multi-scale decomposition technology is used to decompose the iris and facial images into multiple sub-images of different scales and perform image enhancement on the sub-images:
[0055] , , , where is the final enhanced image, N is the scale number of multi-scale decomposition, which is determined by the number of scales to be decomposed. The optimal scale number is usually selected through experiments. This parameter determines the number of scales at which the image is decomposed. Through the decomposition of multiple scales, the system can capture information at different levels of detail in the image. Selecting an appropriate scale number N can balance the enhancement effect of the global information and local details of the image, and improve the accuracy of overall recognition. A (I) is the weight coefficient at the Ath scale, which is determined by the cross-validation method. By introducing the weight coefficient, the system can find the best balance between multiple scales, so that the enhanced image maintains consistency and clarity at each scale. α and β are adjustment parameters in the local contrast enhancement process, which control the intensity of the enhancement. By adjusting α and β, the contrast of the image can be finely controlled to avoid over-enhancement or under-enhancement, ensure that the details of the image are clearer, and adapt to the recognition needs under different lighting conditions. γ Ais a smoothing coefficient related to the Ath scale, which is used to control the impact of multi-scale transformation on the result, especially the smooth transition between different scales, which helps to reduce the discontinuity caused by scale changes, thereby improving the visual effect and recognition accuracy of the final image. δ is a small constant to prevent the denominator from being zero. In order to avoid the situation where the denominator is zero during the calculation of the formula, thereby ensuring the stability of the formula and the reliability of the calculation, M(I A ) is the result of multi-scale transformation, which is the weighted sum of images at different scales. It integrates the image information at each scale, so that in the image enhancement process, information at different scales can be used for comprehensive processing, thereby obtaining a clearer and more detailed enhanced image. A (x,y) is the enhancement function at the Ath scale, H A (x,y) is the spatial filter of the Ath scale, which is used to extract image details at different scales, so that the enhancement process can better capture the structural features of the image, especially the detail changes of the image at different scales. It is the gradient of the image at the Ath scale, which is used to measure the rate of change of the image at a certain position. In the enhancement process, the introduction of gradient information helps to strengthen the edges and contours in the image, making the enhanced image sharper and clearer. It is a local contrast enhancement function. By calculating the contrast of the image in the local area and enhancing it, the enhanced image is clearer and more distinct in details. It can significantly improve the visual quality of the image, enhance the contrast, and make the details in the image more prominent. is the local mean of the A-th scale image at the (x, y) position, which is used to calculate the average brightness of the local area, thus providing a benchmark for the calculation of local contrast. It is the local standard deviation, which indicates the brightness distribution of the image in the local area. Through this parameter, the local contrast of the image can be judged, thus determining the strength of the enhancement. is a small constant that prevents the denominator from being zero;
[0056] Perform recognition processing on the enhanced image.
[0057] In the process of user registration for credit cards, traditional authentication methods such as passwords and security questions can no longer meet modern security needs. Through biometric technology to verify user identity, the system collects user's facial and iris images, and enhances the image's depth information and resolution through holographic projection and multi-frame image synthesis technology. This multi-level and multi-angle biometric feature collection method significantly improves recognition accuracy and security, eliminates the possibility of identity theft, and thus enhances user trust and user experience. In practical applications, high-quality image data is the basis for accurate recognition and analysis.
[0058] Each sub-image corresponds to a different spatial resolution and can capture different details in the image. By enhancing the sub-images, the system can enhance local details while maintaining global structural information. This combination of multi-scale decomposition and enhancement helps improve the stability and accuracy of the recognition algorithm when processing complex images, especially in the case of low-quality images or partial occlusion, it can still ensure a high recognition success rate.
[0059] Furthermore, the recognition processing of the enhanced image refers to using the Sobel operator to perform preliminary edge detection on the enhanced image to identify edge features of the iris and pupil regions: ; Where G(x,y) is the image gradient amplitude, Y x is the horizontal gradient of the image, Y y is the vertical gradient of the image, and x, y represent the pixel coordinates of the image. The image is further processed by the dynamic circular convolution algorithm to separate the iris from the background: Where M(x,y) is the local gradient change, It is The pixel brightness value of the input image of channels, is the local average brightness value, J is the number of channels, and x, y represent the pixel coordinates of the image; large gradient changes usually indicate the presence of iris boundaries, and gradient changes M(x, y) are used to identify and separate the iris from the background;
[0060] Fit pupil and iris boundaries using the Hough circle transform method:
[0061] Set the parameter range of the Hough space, including the possible center position (a, b) and radius range r. The (a, b) range is within the image size. The range of r can be based on the expected size of the pupil and iris. For each edge point (x, y), assume it is a point on the circumference, traverse the possible radius r, and calculate the corresponding center position (a, b). Increase the count of the calculated (a, b, r) combination in the accumulator:
[0062] ; Where (a, b) is the coordinate of the center of the circle, r is the radius, and x1 and x2 are the pixel points on the circumference;
[0063] Find the local maximum in the accumulator, which represents the most likely combination of center and radius. Use the detected (a, b, r) combination to draw a fitted circle in the original image to verify the accuracy of the fitting. Adjust the parameters according to the fitting results to ensure that the detection results are consistent with the actual pupil and iris boundaries.
[0064] Use the image gradient guided filter to calculate the gradient of the iris image after fitting:
[0065] ; In the formula, Gradient vector, indicating the image at point The rate of change, is the horizontal gradient component of the image, It is the gradient component in the vertical direction of the image; the whole iris image is processed by convolution operation to obtain the gradient value of each pixel, the direction and magnitude of the gradient are calculated, and the gradient magnitude and direction are mapped into a feature map. Each pixel corresponds to a set of gradient magnitude and direction values;
[0066] Use principal component analysis to encode and reduce the dimension of the extracted features, and input the reduced-dimensional features into a bidirectional long short-term memory network for time series analysis;
[0067] Use the facial key point detection algorithm (68 key point detections in the Dlib library) to identify and align facial feature points, use the pre-trained convolutional neural network to extract facial deep features and encode and reduce the dimension of the extracted facial features;
[0068] Weighted fusion of iris and facial features through multi-head self-attention mechanism:
[0069] , where Q, K, and V are query matrix, key matrix, and value matrix, respectively. k is the dimension of the key, T is the transpose operation;
[0070] Build and train a GAN model, where the generator generates disguised samples and the discriminator distinguishes between real samples and disguised samples;
[0071] The historical iris and facial features are used as real samples. New disguised samples are generated by learning the features of real samples. The discriminator uses a deep convolutional network to perform feature recognition and judge the authenticity of the input samples.
[0072] The collected iris and facial features are input into the discriminator for real-time disguise detection to obtain the authenticity probability value of the sample;
[0073] Set the judgment threshold Q to compare with the authenticity probability to judge the authenticity of the sample:
[0074] If the output authenticity probability value is greater than or equal to the judgment threshold Q, the sample is true, and the user is allowed to register a credit card through disguise detection;
[0075] If the output authenticity probability value is lower than the judgment threshold Q, the sample is disguised and cannot pass the disguise detection. A warning is issued and the user is notified to resample the image;
[0076] Select a threshold range of (0, 1) and increase it in small steps. Calculate the true positive rate (TPR) and false positive rate (FPR) under different thresholds. Generate data points of the ROC curve based on the TPR and FPR under different thresholds. Draw the ROC curve in a two-dimensional coordinate system with the horizontal axis being FPR and the vertical axis being TPR. Observe the shape and position of the ROC curve and find the point with the largest area under the curve (AUC). According to the ROC curve, select a threshold so that the curve is as close to the upper left corner as possible (meaning high TPR and low FPR). Calculate the AUC of the ROC curve. The closer the value is to 1, the better the performance of the model. Select a threshold that maximizes the difference between TPR and FPR, that is, maximizes TPR−FPR.
[0077] The Sobel operator is used for preliminary edge detection, and the dynamic circular convolution algorithm is used to separate the iris from the background, which improves the image processing effect and accuracy. The Hough circle transform is used to fit the pupil and iris boundaries, providing high-quality boundary data for subsequent feature analysis. In the process of users using credit cards, accurate user portraits can help financial institutions provide services that better meet user needs. The texture features of the iris are extracted through the image gradient guided filter, and the principal component analysis and bidirectional long short-term memory network are combined for feature encoding and time series analysis. The system can capture the dynamic changes of user features. This in-depth analysis is of great significance for preventing fraud and improving the quality of user services.
[0078] In contract management, the authenticity and security of data are of vital importance. Generative adversarial networks (GANs) effectively improve the accuracy of disguise detection and ensure the authenticity of registered users through adversarial learning between generators and discriminators. Through adversarial learning, the effect of disguise detection can be continuously improved to ensure the reliability of detection results. Real-time disguise detection is performed through the GAN model to ensure the efficiency and security of the user registration process. Real-time disguise detection can effectively prevent disguise attacks and improve the security and reliability of the registration process.
[0079] Furthermore, the collecting of user data, calculating user credit scores and dynamically adjusting credit scores, generating preliminary financial contracts and deploying them on the blockchain to form smart financial contracts refers to collecting multi-dimensional data of registered users and pre-processing the collected data, extracting key features from the data, including the user's income level, debt-to-income ratio, credit card utilization rate and credit history length, and normalizing different features;
[0080] Use the Logistic regression model as the credit scoring model, use historical user data as the training set to input the credit scoring model for iterative training, define the loss function to iteratively optimize the model parameters, and stop iterating when the loss of the credit scoring model no longer decreases significantly during continuous iterations, output the model parameters and update the credit scoring model;
[0081] Inputting user feature data into a credit scoring model to obtain a corresponding user credit score;
[0082] Set up to obtain new user data once a quarter, retrain the credit scoring model based on the latest user data and adjust the model parameters, and automatically adjust the credit score of each user according to the latest model;
[0083] Automatically generate financial contract terms based on the user's credit score, use Solidity to write smart financial contract code, embed the financial contract terms into the smart financial contract, select the Ethereum blockchain as the deployment platform, deploy the smart financial contract through the Ethereum mainnet and record the hash value of each financial contract on the chain.
[0084] By collecting multi-dimensional data such as the user's income level, debt-to-income ratio, credit card utilization rate, and length of credit history, a more comprehensive user profile can be constructed. This fine-grained data analysis helps to accurately assess the user's credit risk and provides a basis for subsequent credit scoring. Logistic regression models are widely used in binary classification problems due to their good interpretability and efficient computing power. In credit scoring, it can effectively identify potential default risk users. By using historical user data for iterative training and continuously updating model parameters, the model can be ensured to adapt to changes in the market and user behavior. This enables the model to not only accurately predict risks, but also respond agilely to market changes. User data is updated and the model is retrained every quarter to ensure that the credit score reflects the user's latest credit status in real time. This dynamic adjustment mechanism enhances the flexibility and accuracy of financial contract management. By regularly adjusting the score, users can improve their credit status through good financial behavior during the contract execution period, thereby enjoying better financial services. Automatically generate and execute contract terms through smart contracts, reduce manual intervention and human errors, and improve the speed and accuracy of contract execution. Generate personalized financial contract terms based on user credit scores to provide financial products and services that better meet user needs.
[0085] S2. Create a smart financial contract account to set the billing cycle and account type, conduct a combination analysis of multiple credit card financial contracts held by the user, and formulate the optimal financial contract combination strategy;
[0086] Specifically, the creation of a smart financial contract account and setting of a billing cycle and an account type refers to using a revolving credit account as the default account type, calculating each user's initial credit limit based on the minimum credit limit set by the bank and the user's credit score, binding the user's initial credit limit and the user account and defining them as financial contract parameters, calling the Ethereum platform's API to automatically deploy the smart financial contract to complete account creation and setting the user's bill update cycle to monthly updates.
[0087] By using a revolving credit account as the default account type and combining it with the deployment of smart contracts on the Ethereum platform, account management is automated and transparent. This automation not only reduces human errors, but also improves the operational efficiency of financial institutions. The user's initial credit limit is automatically calculated based on the user's credit score and the minimum credit limit set by the bank. This personalized credit limit setting method not only meets the bank's risk management needs, but also improves the user experience and satisfaction. By setting the billing cycle to monthly updates, users can better manage their financial expenditures and repayment plans. This mechanism encourages users to develop good financial habits, thereby reducing overdue risks and improving overall financial health. The monthly updated billing cycle allows financial institutions to adjust account parameters in a timely manner based on the user's latest credit performance and changes in the market environment. This dynamic adjustment mechanism enhances risk control capabilities and reduces potential bad debt risks. Through the Ethereum platform's API, the deployment of financial contracts is efficient, transparent and automated. The automatic execution feature of smart contracts avoids delays and errors that may be caused by human intervention, thereby improving the efficiency and security of contract execution.
[0088] Furthermore, the combination analysis of multiple credit card financial contracts held by the user and the formulation of the optimal financial contract combination strategy refers to collecting data related to all credit card financial contracts held by the user, including the credit limit, interest rate, annual fee, reward policy, usage frequency and historical payment record of each credit card, and integrating all data into the central database after pre-processing;
[0089] Identify the features that influence the credit card portfolio, including interest rates, usage frequency, annual fees, and rewards policies, and quantify the features into feature vectors;
[0090] Use the linear programming model to optimize the financial contract portfolio and define the objective function with the financial contract costs and benefits as variables:
[0091] , where Z is the optimization target, the net income of the financial contract portfolio, z i is the estimated annual income of the ith financial contract, reflecting the expected income that each financial contract will bring to users within one year, ensuring that the contribution of each contract can be quantified so that effective comparison and trade-off can be made during the optimization process. i is the interest rate of the ith financial contract. By introducing u i , the formula can take into account the interest rate differences of different contracts, so as to select the contract combination with the lowest cost in the optimization process, m i is the credit limit of the ith financial contract, ensuring that the user's capital needs are taken into account during portfolio optimization to avoid selecting contracts with too low credit limits that affect the user's experience.j is the annual fee of the jth financial contract, which can minimize the annual fee cost in the optimization process and avoid users from paying unnecessary high fees. e represents the total number of financial contracts of the credit card held by the user, and R i It is the quantitative value of the reward policy, which is calculated by the reward points accumulated by the user under i contracts and the monetary value corresponding to each point. By converting the reward points of different contracts into actual monetary values in a quantitative way and incorporating them into the optimization calculation, it can help users obtain more reward returns in the contract portfolio and increase net income. i is the decision variable, calculated by the linear programming optimization algorithm, l j is the selection variable of the jth financial contract, with a value of 0 or 1, l j =1 means selecting the jth financial contract and including it in the optimization portfolio, l j = 0 means not to select the jth financial contract. By introducing l i The formula can flexibly select the optimal contract combination, making the optimization process more accurate and efficient. z0, f0 and R0 are standardization factors, which enable parameters of different units to be compared and optimized on the same scale. i , f i and R i Dimensionless, z0 is the average annual expected return value of all contracts, f0 is determined by calculating the average annual fee of all financial contracts, and R0 is determined by calculating the average of the quantitative values of the reward policies of all financial contracts, which improves the accuracy of the calculation and avoids the calculation error caused by the difference in parameter units;
[0092] Define the constraints as:
[0093] , where B is the upper limit of the total cost acceptable to the user, which can prevent the user from exceeding the budget and increasing the financial burden, L is the minimum total benefit expected by the user, ensuring that the user's benefit requirements are met and improving user satisfaction, l i is the decision variable, f i is the annual fee of the ith financial contract, C is the upper limit of the total credit limit acceptable to the user. By setting C, the formula can ensure that the user's total credit limit will not exceed the controllable range and avoid potential credit risks. N max The maximum number of contracts that users are allowed to choose from limits the number of financial contracts that users can choose, avoiding management difficulties caused by too many contracts, simplifying the user's selection process, and ensuring that the final recommended credit card combination is easy to manage and use;
[0094] Solve the model according to the linear programming algorithm and find the financial contract combination that maximizes the objective function as the optimal financial contract combination;
[0095] Select the credit card corresponding to the financial contract in the financial contract combination, combine the credit cards and recommend the best credit card combination to the user.
[0096] By collecting data related to all credit card financial contracts held by users, we are able to build a comprehensive user credit card usage profile. This includes not only basic credit limits, interest rates, and annual fees, but also complex information such as reward policies, usage frequency, and historical payment records. This comprehensiveness ensures our accuracy in the analysis phase and avoids bias caused by missing information. At the same time, by preprocessing and integrating this information into a central database, we can effectively eliminate data redundancy and inconsistency and provide a unified analysis basis. Feature extraction is to quantify complex credit card attributes into analyzable feature vectors so that the characteristics of each credit card can be compared in the same dimension. For example, converting key features such as interest rates, annual fees, and usage frequency into vector form can provide a clear analysis basis for subsequent models. This quantification process not only enhances the comparability of data, but also makes the differences between different credit cards more obvious, providing the possibility for refined optimization. The application of linear programming models in financial contract combinations can help users find the most cost-effective combination among many credit card products. By defining net income as the objective function and setting multiple constraints, the model can identify the optimal solution in a huge combination space. This not only helps users reduce the total cost of using credit cards (such as interest and annual fees), but also maximizes reward benefits (such as points and cash back). Through combination analysis, the system can identify the best strategy in a complex credit card combination, allowing users to have greater flexibility in choosing and using credit cards. For example, the system can recommend users to use specific cards on specific occasions to obtain higher rewards, or adjust the credit card usage strategy based on the frequency of use to avoid unnecessary expenses. Recommending the optimal credit card combination is to directly convert the results of the combination analysis into user behavior guidance. By providing personalized service recommendations, users can not only maximize the financial benefits of using credit cards, but also optimize consumption habits and credit card usage patterns based on recommendations, reduce unnecessary financial losses, and users can manage credit cards more effectively, optimize financial benefits, reduce usage costs, and achieve personalized financial goals.
[0097] S3. Securely store financial contracts and implement access control, and display financial contracts through a visual interface;
[0098] Specifically, the secure storage of financial contracts and implementation of access control refers to encrypting the financial contract combination, using the Advanced Encryption Standard for symmetric encryption and storing the hash value of the encrypted financial contract on the blockchain, using the RBAC mechanism to define user roles and set different permission levels based on business needs, allowing only authorized users to access financial contracts, logging each access and conducting regular audits.
[0099] The encrypted storage of contracts is achieved through the Advanced Encryption Standard (AES) and blockchain technology, ensuring the confidentiality and immutability of data. The application of the RBAC mechanism provides flexible permission management and efficient auditing functions, improves the security and compliance of the system, significantly improves the contract management efficiency of financial institutions and enterprises, reduces the risk of data leakage, and provides a solid technical guarantee for the development of the industry.
[0100] Furthermore, displaying the financial contract through a visual interface refers to using a front-end framework (such as React.js or Vue.js) to develop a user-friendly visual interface, using a data visualization tool (such as D3.js) to display the key content and indicators of the financial contract, decrypting the financial contract for verified user requests, and displaying the plaintext content on the interface, and using charts and text boxes to display the financial contract change records in real time.
[0101] By using modern front-end frameworks such as React.js or Vue.js, efficient, dynamic and responsive user interfaces can be created. This interface can intuitively display financial contract information, allowing users to quickly understand contract terms and reduce reading and decision-making time. Data visualization tools such as D3.js can transform complex financial data into intuitive charts and graphs, allowing users to quickly understand and analyze key indicators of contracts (such as yield, fee structure, risk assessment, etc.). Through encryption and decryption mechanisms, only verified users can access and view the details of financial contracts, ensuring the security and confidentiality of sensitive information. The display of plain text content allows users to understand the contract terms and details in detail, increasing the transparency and trust of the system, and displaying contract change records in real time. Users can track all historical modifications of the contract to ensure the transparency and integrity of the contract.
[0102] Example 2, reference Figure 2 , which is the second embodiment of the present invention, and which is different from the previous embodiment, provides a credit card financial contract management system, which includes:
[0103] The identity verification module is used to collect and process the user's facial and iris images to verify the authenticity of the user's identity;
[0104] The contract generation module is used to collect multi-dimensional data of verified users and calculate user scores, and automatically generate financial contract terms based on the user's credit score;
[0105] The contract combination module is used to analyze multiple credit card financial contracts held by the user and optimize the user's credit card financial contract combination based on the analysis results;
[0106] The storage and display module is used to securely store financial contract combinations and perform access control, and visually display the content of financial contracts to verified users.
[0107] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.
[0108] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.
[0109] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.
[0110] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
Claims
1. A credit card financial contract management method, characterized by: include, Verify the identity of registered users through biometric technology, collect user data to calculate user credit scores and dynamically adjust credit scores, generate preliminary financial contracts and deploy them on the blockchain to form smart financial contracts; Create a smart financial contract account to set the billing cycle and account type, conduct a combination analysis of multiple credit card financial contracts held by the user, and formulate the optimal financial contract combination strategy; Securely store financial contracts and implement access control, and display financial contracts through a visual interface; Wherein, verifying the identity of the registered user through biometric technology includes: Collect facial and iris images of users when they register for a credit card, and tag each user's data with a unique identifier; After preprocessing the image, the multi-scale decomposition technology is used to decompose the iris and facial images into multiple sub-images of different scales and the sub-images are enhanced; Perform recognition processing on the enhanced image; After the image is preprocessed, a multi-scale decomposition technique is used to decompose the iris and facial image into a plurality of sub-images of different scales and enhance the sub-images, including: By formula: Performing image enhancement on the sub-image; in, Where ε(I) is the final enhanced image, N is the number of scales of multi-scale decomposition, and w A (I) is the weight coefficient at the Ath scale, α and β are adjustment parameters, γ A is the smoothing coefficient associated with the Ath scale, δ is a constant, M(I A ) is the result of multi-scale transformation, F A (x,y) is the enhancement function at the Ath scale, H A (x,y) is the spatial filter of the Ath scale, ▽I A (x, y) is the gradient of the image at the Ath scale, LCε(I) is the local contrast enhancement function, μI A (x, y) is the local mean of the A-th scale image at position (x, y), σI A (x,y) is the local standard deviation and ρ is a constant.
2. The credit card financial contract management method according to claim 1, characterized in that: The step of performing recognition processing on the enhanced image includes: The Sobel operator is used to perform preliminary edge detection on the enhanced image, identify the edge features of the iris and pupil areas, and separate the iris from the background in the image through the dynamic circular convolution algorithm; The pupil and iris boundaries are fitted using the Hough circle transform method, and the gradient of the iris image is calculated using the image gradient guided filter on the fitted iris image; Use convolution operation to process the entire iris image to obtain the gradient value of each pixel, calculate the direction and magnitude of the gradient, and map the gradient magnitude and direction into a feature map. Each pixel corresponds to a set of gradient magnitude and direction values. Use principal component analysis to encode and reduce the dimension of the extracted features, and input the reduced-dimensional features into a bidirectional long short-term memory network for time series analysis; Use pre-trained convolutional neural network to extract facial deep features, encode and reduce the dimension of the extracted facial features, and perform weighted fusion of iris and facial features through multi-head self-attention mechanism; Judging the authenticity of the sample based on the fused iris and facial features means building and training a GAN model, inputting the collected iris and facial features into the discriminator for real-time camouflage detection to obtain the authenticity probability value of the sample; Set the judgment threshold Q and compare it with the authenticity probability value to judge the authenticity of the sample: If the output authenticity probability value is greater than or equal to the judgment threshold Q, the sample is true, and the user is allowed to register a credit card through disguise detection; If the output authenticity probability value is lower than the judgment threshold Q, the sample is disguised and cannot pass the disguise detection. A warning is issued and the user is notified to resample the image.
3. The credit card financial contract management method according to claim 2, characterized in that: The collecting of user data, calculating user credit scores and dynamically adjusting credit scores, generating preliminary financial contracts and deploying them on the blockchain to form smart financial contracts refers to collecting multi-dimensional data of registered users and pre-processing the collected data, extracting key features from the data, including the user's income level, debt-to-income ratio, credit card utilization rate and credit history length, and normalizing different features; Use the Logistic regression model as the credit scoring model, use historical user data as the training set to input the credit scoring model for iterative training, define the loss function to iteratively optimize the model parameters, and stop iterating when the loss of the credit scoring model no longer decreases significantly during continuous iterations, output the model parameters and update the credit scoring model; Inputting user feature data into a credit scoring model to obtain a corresponding user credit score; Set up to obtain new user data once a quarter, retrain the credit scoring model based on the latest user data and adjust the model parameters, and automatically adjust the credit score of each user according to the latest model; Automatically generate financial contract terms based on the user's credit score, use Solidity to write smart financial contract code, embed the financial contract terms into the smart financial contract, select the Ethereum blockchain as the deployment platform, deploy the smart financial contract through the Ethereum mainnet and record the hash value of each financial contract on the chain.
4. The credit card financial contract management method according to claim 3, characterized in that: The creation of a smart financial contract account and setting of a billing cycle and an account type refers to using a revolving credit account as the default account type, calculating each user's initial credit limit based on the minimum credit limit set by the bank and the user's credit score, binding the user's initial credit limit to the user account and defining them as financial contract parameters, calling the Ethereum platform's API to automatically deploy a smart financial contract to complete account creation and setting the update cycle of the user's bill to monthly updates.
5. The credit card financial contract management method according to claim 4, characterized in that: The said combination analysis of multiple credit card financial contracts held by the user and formulation of the optimal financial contract combination strategy refers to collecting relevant data of all credit card financial contracts held by the user, including the credit limit, interest rate, annual fee, reward policy, usage frequency and historical payment record of each credit card, and integrating all data into the central database after pre-processing; Identify the features that influence the credit card portfolio, including interest rates, usage frequency, annual fees, and rewards policies, and quantify the features into feature vectors; Use the linear programming model to optimize the financial contract portfolio and define the objective function with the financial contract costs and benefits as variables: In the formula, Z is the optimization target, the net income of the financial contract portfolio, z i is the annual expected return of the ith financial contract, u i is the interest rate of the ith financial contract, m i is the credit limit of the ith financial contract, f j is the annual fee of the jth financial contract, e represents the total number of financial contracts of the credit card held by the user, R i is the quantitative value of the reward policy, l i is the decision variable, l j is the selection variable of the jth financial contract, z0, f0 and R0 are standardization factors; Define the constraints as: In the formula, B is the upper limit of the total cost that the user can accept, L is the minimum total benefit expected by the user, and l i is the decision variable, f i is the annual fee of the ith financial contract, C is the upper limit of the total credit limit acceptable to the user, N max The maximum number of contracts that the user is allowed to select; Solve the model according to the linear programming algorithm and find the financial contract combination that maximizes the objective function as the optimal financial contract combination; Select the credit card corresponding to the financial contract in the financial contract combination, combine the credit cards and recommend the best credit card combination to the user.
6. The credit card financial contract management method according to claim 5, characterized in that: The secure storage of financial contracts and implementation of access control refers to encrypting the financial contract combination, using the Advanced Encryption Standard for symmetric encryption and storing the hash value of the encrypted financial contract on the blockchain, using the RBAC mechanism to define user roles and set different permission levels based on business needs, allowing only authorized users to access financial contracts, logging each access and conducting regular audits.
7. The credit card financial contract management method according to claim 6, characterized in that: The display of financial contracts through a visual interface refers to using a front-end framework to develop a user-friendly visual interface, using data visualization tools to display the key content and indicators of the financial contract, decrypting the financial contract for verified user requests, and displaying the plain text content on the interface, and using charts and text boxes to display financial contract change records in real time.
8. A credit card financial contract management system based on the credit card financial contract management method according to any one of claims 1 to 7, characterized in that: include, The identity verification module is used to collect and process the user's facial and iris images to verify the authenticity of the user's identity; The contract generation module is used to collect multi-dimensional data of verified users and calculate user scores, and automatically generate financial contract terms based on the user's credit score; The contract combination module is used to analyze multiple credit card financial contracts held by the user and optimize the user's credit card financial contract combination based on the analysis results; The storage and display module is used to securely store financial contract combinations and perform access control, and visually display the content of financial contracts to verified users.
9. A computer device comprising: A memory and a processor; the memory stores a computer program, wherein the processor implements the steps of the credit card financial contract management method according to any one of claims 1 to 7 when executing the computer program.
Citation Information
Patent Citations
Method and system of texture features for biometric authentication
CN103390153A
Bank financial management personnel authority setting method and apparatus based on combination of blockchain and iris recognition
CN108881259A
Financial product risk control method based on block chain technology
CN109102390A
Financial data processing method and device, equipment and storage medium
CN115018576A
Consumption financial product service construction method based on distributed automation
CN116645197A
