Examination and approval limit determination method and device, equipment, medium and product
By extracting features from credit card approval forms and using calculation models, and comparing automatic and assisted approval limits, the problem of inaccurate approval limits in credit card approval processes has been solved, achieving higher approval accuracy and financial security.
Patent Information
- Application Number
- CN202511051311.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-29
- Publication Date
- 2025-11-11
AI Technical Summary
In the current credit card approval process, the increased number of approvals through automated processes makes it difficult to guarantee the accuracy of approved credit limits, thus affecting financial security.
By automating the approval process of the current approval form, extracting the features of the original approval information, calculating the approval amount using the approval amount calculation model, and comparing the automatic and auxiliary approval amounts, the final approval result is determined and a second review is conducted to ensure accuracy.
It improved the accuracy of determining the approval amount, ensured financial security, identified and corrected potential problems in the system, and improved the approval quality of automated processes.
Smart Images
Figure CN120931384A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of artificial intelligence technology, and more particularly to the fields of data processing and financial technology, specifically to a method, apparatus, equipment, medium, and product for determining approval limits. Background Technology
[0002] Credit card limits are the maximum consumer credit granted by banks based on a comprehensive credit assessment of the applicant, involving a dynamic balance of multiple factors. Due to the increasing number of credit card applications, the volume of applications processed through automated workflows has also increased, making it crucial to ensure the accuracy of credit limit approvals in automated processes. Summary of the Invention
[0003] This invention provides a method, apparatus, equipment, medium, and product for determining approval limits, in order to improve the accuracy of approval limit determination.
[0004] According to one aspect of the present invention, a method for determining an approval limit is provided, the method comprising:
[0005] The current approval form is automatically approved to obtain the automatically approved amount.
[0006] Extract features from the original approval information of the current approval form to obtain the target approval features;
[0007] The target approval characteristics are calculated using an approval limit calculation model to obtain an auxiliary approval limit;
[0008] The target approval result is determined based on the automatic approval limit and the auxiliary approval limit.
[0009] According to another aspect of the present invention, an approval limit determination device is provided, the device comprising:
[0010] The automatic approval limit determination module is used to automatically process the approval of the current approval form and obtain the automatic approval limit.
[0011] The target approval feature determination module is used to extract features from the original approval information of the current approval form to obtain the target approval features;
[0012] The auxiliary approval limit determination module is used to calculate the target approval characteristics through the approval limit calculation model to obtain the auxiliary approval limit;
[0013] The target approval result determination module is used to determine the target approval result based on the automatic approval limit and the auxiliary approval limit.
[0014] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:
[0015] At least one processor; and
[0016] A memory communicatively connected to the at least one processor; wherein,
[0017] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the approval limit determination method according to any embodiment of the present invention.
[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the approval limit determination method according to any embodiment of the present invention.
[0019] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the approval limit determination method according to any embodiment of the present invention.
[0020] The technical solution of this invention involves automatically approving the current approval form to obtain an automatic approval limit; extracting features from the original approval information of the current approval form to obtain target approval features; calculating the target approval features using an approval limit calculation model to obtain an auxiliary approval limit; and determining the target approval result based on the automatic approval limit and the auxiliary approval limit. This technical solution, by using an approval limit calculation model to perform a secondary review of the automatic approval limit, facilitates the detection of potential system problems that could lead to inaccurate approval limits, thereby ensuring financial security.
[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description
[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used 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 those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0023] Figure 1 This is a flowchart of a method for determining an approval limit according to an embodiment of the present invention;
[0024] Figure 2This is a flowchart of a method for determining an approval limit according to an embodiment of the present invention;
[0025] Figure 3 This is a schematic diagram of the structure of a neural network model provided according to an embodiment of the present invention;
[0026] Figure 4 This is a schematic diagram of the structure of an approval quota determination device provided in an embodiment of the present invention;
[0027] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the approval limit determination method of this invention. Detailed Implementation
[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.
[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0030] Furthermore, it should be noted that all relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are information and data authorized by the user or fully authorized by all parties. Moreover, the collection, storage, use, processing, transmission, provision, disclosure, and application of the relevant data comply with the relevant laws, regulations, and standards of the relevant regions, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation portals are provided for users to choose to authorize or refuse. Among them, user personal information includes but is not limited to credit information, anti-fraud information, card information, pre-approved credit limit, recommended credit limit, etc.
[0031] In addition, users are provided with corresponding operation entry points to choose whether to agree to or reject the automated decision results; if the user chooses to reject, the process will proceed to the expert decision-making process.
[0032] Figure 1 This is a flowchart illustrating a method for determining an approval limit according to an embodiment of the present invention. This embodiment is applicable to situations involving the determination of approval limits in financial business scenarios, particularly in credit card approval processes. The method can be executed by an approval limit determination device, which can be implemented in hardware and / or software and configured in an electronic device carrying the approval limit determination function, such as a server. Figure 1 As shown, the method includes:
[0033] S110. Perform automatic process approval on the current approval form to obtain the automatic approval amount.
[0034] The "current approval form" refers to the approval form for the user's current credit card application. The "automatic approval limit" refers to the approved credit limit obtained by the automated business process approval system.
[0035] Specifically, the automated workflow approval system automatically approves the current approval form submitted by the user to obtain the automatically approved amount.
[0036] S120. Extract features from the original approval information of the current approval form to obtain the target approval features.
[0037] The original approval information refers to the initial information in the current approval form, including but not limited to customer information, credit information, anti-fraud information, existing card information, pre-approved credit limit, and recommended credit limit. The target approval features refer to the features obtained by feature extraction from the current approval form, which can be represented in matrix or vector form.
[0038] An alternative approach is to vectorize the original approval information of the current approval form to obtain the target approval features.
[0039] Another option is to normalize the original approval information of the current approval form to obtain the target approval features. Specifically, this involves normalizing the credit information, anti-fraud information, existing card information, pre-approved credit limit, and recommended credit limit in the original approval information of the current approval form to obtain the target approval features. It is understandable that normalizing the original approval information allows for effective scaling of the information, thus facilitating subsequent model training.
[0040] S130. The target approval characteristics are calculated using the approval limit calculation model to obtain the auxiliary approval limit.
[0041] The auxiliary approval limit refers to the approval limit calculated using an approval limit calculation model on the current approval form. This approval limit calculation model is obtained by training a simple neural network using historical approval forms.
[0042] Specifically, the target approval features of the current approval form are input into the approval amount calculation model, and the auxiliary approval amount is obtained through model prediction processing.
[0043] S140. Determine the target approval result based on the automatic approval limit and the auxiliary approval limit.
[0044] The target approval result refers to the final approved amount and account opening instructions, etc.
[0045] Specifically, the automatically approved limit and the assisted approval limit are compared. If the difference is small, meaning the difference between the automatically approved limit and the assisted approval limit is within the set range, the approval is granted and the account opening process begins. If the difference is large, meaning the difference between the automatically approved limit and the assisted approval limit exceeds the set range, the process is further transferred to manual processing for manual review to determine the accuracy of the limit.
[0046] The technical solution of this invention involves automatically approving the current approval form to obtain an automatic approval limit; extracting features from the original approval information of the current approval form to obtain target approval features; calculating the target approval features using an approval limit calculation model to obtain an auxiliary approval limit; and determining the target approval result based on the automatic approval limit and the auxiliary approval limit. This technical solution, by using an approval limit calculation model to perform a secondary review of the automatic approval limit, facilitates the detection of potential system problems that could lead to inaccurate approval limits, thereby ensuring financial security.
[0047] Figure 2 This is a flowchart of a method for determining the approval limit according to an embodiment of the present invention. Based on the above embodiments, this embodiment further elaborates on the specific determination method of the approval limit calculation model. Figure 2 As shown, the method includes:
[0048] S210. Perform automatic process approval on the current approval form to obtain the automatic approval amount.
[0049] S220. Extract features from the original approval information of the current approval form to obtain the target approval features.
[0050] S230. The target approval characteristics are calculated using the approval limit calculation model to obtain the auxiliary approval limit.
[0051] S240. Determine the target approval result based on the automatic approval limit and the auxiliary approval limit.
[0052] In this embodiment, the approval limit calculation model is trained in the following way: historical approval forms are obtained, and feature filtering and feature extraction are performed on the historical approval forms to obtain sample approval features of the sample approval forms; the sample approval features are used to train the neural network model to obtain the approval limit calculation model.
[0053] Among them, "historical approval forms" refer to approval forms that have already been approved, such as those from the past month. "Sample approval forms" refer to approval forms obtained after filtering historical approval forms. "Sample approval features" refer to the features extracted from the approval information of sample approval forms. It should be noted that the approval limit calculation model uses a simple neural network model, including an input layer, an output layer, and at least one hidden layer. Calculating the approval limit based on a simple fully connected neural network is convenient and fast.
[0054] Specifically, historical approval forms are obtained, and those that do not meet the requirements are removed to obtain sample approval forms. Features are extracted from these sample forms to obtain sample approval features. These sample approval features are then used to iteratively train a neural network model to obtain a predicted approval amount. Based on a preset damage function, such as a mean squared error loss function, the training loss is determined according to the predicted approval amount and the actual approval amount of the sample approval forms. This process continues until a training stop condition is met. The neural network model at the point of training stoppage is then used as the approval amount calculation model. The training stop condition is either that the number of iterations reaches a set number or that the training loss stabilizes within a set range. The set number of iterations and the set range can be set by those skilled in the art based on actual circumstances.
[0055] An optional approach involves performing feature filtering and feature extraction on historical approval forms to obtain sample approval features for sample approval forms. This includes: removing approval forms from historical approval forms whose customer information does not meet customer requirements to obtain candidate approval forms; removing approval forms from candidate approval forms whose approval amount does not meet the amount requirements to obtain sample approval forms; and extracting features from the sample approval information of the sample approval forms to obtain sample approval features for the sample approval forms.
[0056] Specifically, approval forms whose customer information does not meet customer requirements are removed from historical approval forms to obtain candidate approval forms. For example, approval forms for minors, cross-default customers, and customers who refuse medical examinations are removed to obtain candidate approval forms. Approval forms whose approval amounts do not meet the requirements are removed from the candidate approval forms. For example, approval forms with approval amounts that are too large or too small are removed to obtain sample approval forms. Feature extraction is performed on the sample approval information of the sample approval forms. For example, the sample approval information of the sample approval forms is vectorized to obtain the sample approval features of the sample approval forms.
[0057] Furthermore, feature extraction is performed on the sample approval information of the sample approval form to obtain the sample approval features of the sample approval form, including: normalizing the sample approval information of the sample approval form to obtain the sample approval features; wherein, the sample approval information includes credit information, anti-fraud information, existing card information, pre-approved limit, recommended limit and approved limit.
[0058] Alternatively, the neural network model includes an input layer, an output layer, and at least one hidden layer. Preferably, the neural network model in this invention includes an input layer, an output layer, and a hidden layer, such as... Figure 3 As shown. Among them, Figure 3 The approved amount is the predicted amount output by the model.
[0059] Accordingly, the neural network model is trained using sample approval features to obtain an approval limit calculation model, including: inputting sample approval features into the input layer to obtain input features; inputting the input features into at least one hidden layer to obtain hidden features; inputting the hidden features into the output layer to obtain the predicted limit; calculating the training loss based on the predicted limit and the labeled limit; iteratively training the neural network model based on the training loss, and dynamically adjusting the ratio of random deactivation (Dropout) and the bias term of the activation function in the hidden layer during the iterative training process, and performing backpropagation training to obtain the approval limit calculation model.
[0060] Input features are the features obtained after processing by the input layer, and can be represented in matrix or vector form. Hidden features are the features obtained after processing by the hidden layer, and can also be represented in matrix or vector form. Predicted credit limit refers to the credit limit output by the neural network model. Labeled credit limit is the true credit limit value corresponding to the sample approval form.
[0061] It's important to note that the structure of a fully connected neural network is not fixed. Generally, a neural network includes an input layer, hidden layers, and an output layer. A fully connected neural network has only one input layer and one output layer; the layers between the input and output layers are hidden layers. Each layer of a neural network has a number of neurons. Neurons between layers are interconnected, but neurons within the same layer are not interconnected. Furthermore, neurons in the next layer connect to all neurons in the previous layer. Neural networks with a large number of hidden layers (usually >2) are called deep neural networks. Deep neural networks have stronger expressive power than shallow networks; a neural network with only one hidden layer can fit any function, but it requires a very large number of neurons. The expression corresponding to the hidden layer is: z = wx + b; The expression f(z) uses the sigmoid activation function. The purpose of activation functions is to increase the non-linearity of the neural network model. Without an activation function, each layer is equivalent to matrix multiplication. Even if you stack several layers, it's still just matrix multiplication. Besides avoiding simple matrix multiplication, the sigmoid activation function maps the input to the range (0,1), which is called normalization, thus preventing any input from reaching infinity. However, there are many other activation functions, such as tanh and ReLU. b is the bias term, which is actually the intercept of the function, used to better fit the data.
[0062] Backpropagation typically occurs when the output value calculated by forward propagation differs from the expected output value. Therefore, the weights W need to be adjusted. This means the input X has a corresponding true value Y (label value). The loss between the neural network's output Y (predicted value) and the true value Y is what backpropagation deals with. The entire network training process is a continuous process of reducing the loss. Deep learning generally uses backpropagation to update weights. Based on the loss function value generated by forward propagation, backpropagation optimizes the parameters of each layer from the output to the input. During this process, gradient descent is used to optimize the parameters. Solving the parameters in a neural network is still a problem of finding the optimal solution in a planning context.
[0063] In this invention, the mean squared error loss function is preferably used to calculate the training loss based on the predicted amount and the labeled amount. Finding the optimal loss value involves differentiation; for a bivariate function, this means finding the partial derivatives. Gradient descent is commonly used for evaluation, where the gradient is:
[0064] Updated weights:
[0065]
[0066] By continuously iterating in this way, the loss value gradually decreases as it approaches the output value, until a certain threshold or number of iterations is reached, at which point training stops. This is how the desired solution is found.
[0067] Finding the optimal loss value involves differentiation; for a bivariate function, this means finding the partial derivatives. Gradient descent is commonly used for evaluation, i.e., gradient:
[0068] Updated weights:
[0069] This iterative process continues, and as the value approaches the output, the loss decreases until a set threshold or number of iterations is reached. At this point, training stops, and the neural network model at the point of termination is used as the model for calculating the approval limit. The threshold and number of iterations can be set by those skilled in the art based on the specific circumstances.
[0070] Here, η is the learning rate, which can usually be set arbitrarily, generally between (0, 1). The learning rate directly affects how quickly the model converges to a local minimum (i.e., reaches the best accuracy). Generally, the larger the learning rate, the faster the neural network learns. If the learning rate is too small, the network is likely to get stuck in a local optimum; however, if it is too large, exceeding the extreme value, the loss will stop decreasing and oscillate repeatedly at a certain point. In other words, if a suitable learning rate is chosen, not only can the model be trained in a shorter time, but various computational resources can also be saved.
[0071] In each training batch, overfitting can be significantly reduced by ignoring half of the feature detectors (setting half of the hidden layer nodes to 0). This method is called Dropout, which reduces the interaction between feature detectors (hidden layer nodes). Dropout, in other words, stops the activation value of a neuron during forward propagation with a certain probability p. This makes the model more generalizable because it does not rely too much on certain local features.
[0072] The reasons for using dropout are as follows: 1) Averaging: Dropping out different hidden neurons is similar to training different networks; the entire dropout process is equivalent to averaging across many different neural networks. 2) Reducing complex co-adaptation relationships between neurons: Because the dropout procedure causes two neurons to not always appear in the same dropout network, weight updates no longer rely on the combined effect of hidden nodes with fixed relationships, preventing situations where certain features are only effective under other specific features. This forces the network to learn more robust features that also exist in random subsets of other neurons. In other words, if the neural network is making a prediction, it should not be too sensitive to certain cue fragments; even if specific cues are lost, it should be able to learn some common features from many other cues.
[0073] Specifically, during the iterative training of the neural network model using sample approval features, the learning rate can be initialized to 0.001, and Dropout can be temporarily omitted to avoid inaccurate weights in the early stages of the model. The bias term of the activation function, i.e., the b-bias, is initialized to 0. If the data becomes overfitted later, the Dropout ratio needs to be increased to allow some hidden layer nodes to be excluded from scheduling, thus reducing overfitting. If the training effect is poor and fails to match the expected approval amount, the bias value of the activation function can be increased to obtain a more accurate approval amount calculation model. It is understandable that the weights can be dynamically adjusted to ensure accuracy.
[0074] This invention determines the approval limit through an approval limit calculation model and an automated approval process system, which is highly reliable compared to manually calculating the approval limit.
[0075] Figure 4 This is a schematic diagram of an approval limit determination device according to an embodiment of the present invention. This embodiment is applicable to situations involving approval limit determination in financial business scenarios, particularly in credit card approval processes. The device can be implemented in hardware and / or software and can be configured in an electronic device that carries the approval limit determination function, such as a server. Figure 4 As shown, the device includes:
[0076] The automatic approval limit determination module 310 is used to automatically process the approval of the current approval form and obtain the automatic approval limit.
[0077] The target approval feature determination module 320 is used to extract features from the original approval information of the current approval form to obtain the target approval features;
[0078] The auxiliary approval limit determination module 330 is used to calculate the target approval characteristics through the approval limit calculation model to obtain the auxiliary approval limit;
[0079] The target approval result determination module 340 is used to determine the target approval result based on the automatic approval limit and the auxiliary approval limit.
[0080] The technical solution of this invention involves automatically approving the current approval form to obtain an automatic approval limit; extracting features from the original approval information of the current approval form to obtain target approval features; calculating the target approval features using an approval limit calculation model to obtain an auxiliary approval limit; and determining the target approval result based on the automatic approval limit and the auxiliary approval limit. This technical solution, by using an approval limit calculation model to perform a secondary review of the automatic approval limit, facilitates the detection of potential system problems that could lead to inaccurate approval limits, thereby ensuring financial security.
[0081] Optionally, the target approval feature determination module 320 is specifically used for:
[0082] Feature extraction is performed on the original approval information of the current approval form to obtain the target approval features, including:
[0083] The original approval information of the current approval form is normalized to obtain the target approval features.
[0084] Optionally, the device also includes a module for determining the approval limit calculation model, including:
[0085] The sample approval feature determination unit is used to obtain historical approval forms, perform feature filtering and feature extraction on historical approval forms, and obtain the sample approval features of the sample approval forms.
[0086] The approval quota calculation model training unit is used to train the neural network model using sample approval features to obtain the approval quota calculation model.
[0087] Optionally, the sample approval feature determination unit is used for:
[0088] Remove approval forms from the historical approval forms that do not meet the customer requirements to obtain candidate approval forms;
[0089] The approval forms whose approval amounts do not meet the requirements are removed from the candidate approval forms to obtain the sample approval forms;
[0090] Feature extraction is performed on the sample approval information of the sample approval form to obtain the sample approval features of the sample approval form.
[0091] Optionally, the neural network model includes an input layer, an output layer, and at least one hidden layer. The training unit for the approval quota calculation model is specifically used for:
[0092] Input the sample approval features into the input layer to obtain the input features;
[0093] Input features are fed into at least one hidden layer to obtain hidden features;
[0094] The hidden features are input into the output layer to obtain the predicted credit limit;
[0095] Calculate the training loss based on the predicted and labeled amounts;
[0096] The neural network model is iteratively trained based on the training loss. During the iterative training process, the ratio of random deactivation Dropout in the hidden layer and the bias term of the activation function are dynamically adjusted, and backpropagation training is performed to obtain the approval quota calculation model.
[0097] During the iterative training of the neural network model using sample approval features, the ratio of Dropout in the hidden layer and the bias term of the activation function are dynamically adjusted to obtain the approval amount calculation model.
[0098] Optionally, the sample approval feature determination unit is specifically used for:
[0099] The sample approval information of the sample approval form is normalized to obtain the sample approval features.
[0100] The approval limit determination device provided in this embodiment of the invention can execute the approval limit determination method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.
[0101] According to embodiments of the present invention, the present invention also provides an electronic device, a readable storage medium, and a computer program product.
[0102] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the approval limit determination method of this invention. Figure 5 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.
[0103] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.
[0104] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.
[0105] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the approval limit determination method.
[0106] In some embodiments, the approval limit determination method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the approval limit determination method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the approval limit determination method by any other suitable means (e.g., by means of firmware).
[0107] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.
[0108] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0109] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.
[0110] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).
[0111] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.
[0112] A computing system can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product within the cloud computing service system to address the shortcomings of traditional physical hosts and virtual private servers, such as high management difficulty and weak business scalability.
[0113] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.
[0114] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.
Claims
1. A method for determining the approval limit, characterized in that, include: The current approval form is automatically approved to obtain the automatically approved amount. Extract features from the original approval information of the current approval form to obtain the target approval features; The target approval characteristics are calculated using an approval limit calculation model to obtain an auxiliary approval limit; The target approval result is determined based on the automatic approval limit and the auxiliary approval limit.
2. The method according to claim 1, characterized in that, Feature extraction is performed on the original approval information of the current approval form to obtain the target approval features, including: The original approval information of the current approval form is normalized to obtain the target approval features.
3. The method according to claim 1, characterized in that, The approval limit calculation model was trained in the following way: Obtain historical approval forms, perform feature filtering and feature extraction on the historical approval forms, and obtain sample approval features of the sample approval forms; The neural network model was trained using sample approval features to obtain the approval amount calculation model.
4. The method according to claim 3, characterized in that, The historical approval forms are subjected to feature filtering and feature extraction to obtain sample approval features of the sample approval forms, including: Remove approval forms from the historical approval forms that do not meet the customer requirements to obtain candidate approval forms; The approval forms whose approval amounts do not meet the requirements are removed from the candidate approval forms to obtain sample approval forms; Feature extraction is performed on the sample approval information of the sample approval form to obtain the sample approval features of the sample approval form.
5. The method according to claim 3, characterized in that, The neural network model includes an input layer, an output layer, and at least one hidden layer. The neural network model is trained using sample approval features to obtain an approval limit calculation model, including: The sample approval features are input into the input layer to obtain the input features; Input features are fed into at least one hidden layer to obtain hidden features; The hidden features are input into the output layer to obtain the predicted credit limit; Calculate the training loss based on the predicted and labeled amounts; The neural network model is iteratively trained based on the training loss. During the iterative training process, the ratio of random deactivation (Dropout) and the bias term of the activation function in the hidden layer are dynamically adjusted, and backpropagation training is performed to obtain the approval quota calculation model.
6. The method according to claim 4, characterized in that, Feature extraction is performed on the sample approval information of the sample approval form to obtain the sample approval features of the sample approval form, including: The sample approval information of the sample approval form is normalized to obtain sample approval features.
7. A device for determining an approval limit, characterized in that, include: The automatic approval limit determination module is used to automatically process the approval of the current approval form and obtain the automatic approval limit. The target approval feature determination module is used to extract features from the original approval information of the current approval form to obtain the target approval features; The auxiliary approval limit determination module is used to calculate the target approval characteristics through the approval limit calculation model to obtain the auxiliary approval limit; The target approval result determination module is used to determine the target approval result based on the automatic approval limit and the auxiliary approval limit.
8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the approval limit determination method according to any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the approval quota determination method according to any one of claims 1-6.
10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method for determining the approval amount according to any one of claims 1-6.