An online approval method and system for automatically generating a cancellation material by process configuration
By acquiring contract and loan information, utilizing image recognition and emergency assessment values, and combining this with the focus of approvers, the problems of low efficiency in processing write-off documents and poor approval quality have been solved, achieving efficient and accurate approval of write-off materials.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HANGYIN CONSUMER FINANCE CO LTD
- Filing Date
- 2023-06-30
- Publication Date
- 2026-06-09
AI Technical Summary
In financial institutions, the large number and length of write-off documents lead to low processing efficiency, excessive approval time, and an inability to accurately grasp key information, resulting in the approval of problematic write-off documents.
By acquiring contract and loan information, using image recognition algorithms to determine the probability of a location to be reviewed, and combining this with an urgency assessment value and the focus of the approvers, the reimbursement materials are sorted and tagged to ensure approval quality and efficiency.
This reduces the need for duplicate verification materials, improves approval efficiency, ensures approval quality, enables differentiated assessment of urgency and detail of review, and avoids the occurrence of verification materials with poor approval quality.
Smart Images
Figure CN116777657B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of online approval technology, and in particular relates to an online approval method and system for automatically generating verification materials through process configuration. Background Technology
[0002] In financial institutions, to process the write-off of non-performing loans and bad debts, materials often need to be manually generated and printed into paper documents. The resulting mountains of paper documents require offline signatures and approvals, which is not only wasteful of resources but also time-consuming. To address these technical problems, invention patent CN108376333B, "Document Approval Method, Medium, Device, and Computing Equipment," identifies personnel information contained in the documents to be approved and sends the documents to the corresponding approval recipients based on this information. However, this invention has the following issues:
[0003] In the existing technical solutions, the number of verification documents that approvers need to process is large. Therefore, if the verification documents are not sorted and marked according to their specific circumstances, the processing efficiency of the verification documents may be low, and the approval time may also exceed the limit.
[0004] In existing technical solutions, due to the large number of pages in the reconciliation documents, approvers cannot verify the actual situation of the documents to be approved during the approval process. Therefore, if the key information of the reconciliation documents cannot be read and a summary cannot be generated, the approvers may not be able to accurately and comprehensively grasp the key information of the reconciliation documents, which may lead to the approval of problematic reconciliation documents.
[0005] To address the aforementioned technical problems, this invention provides an online approval method and system for automatically generating verification materials through process configuration. Summary of the Invention
[0006] To achieve the objectives of this invention, the following technical solution is adopted:
[0007] Firstly, it provides an online approval method that configurates processes to automatically generate verification materials.
[0008] An online approval method for automatically generating reimbursement materials through process configuration, characterized by specifically including:
[0009] S11 obtains contract information and loan information to be written off, and if it is determined from the contract information that the contract information has not yet been written off, generates the write-off materials to be approved using the contract information and loan information;
[0010] S12 If the probability of a problem at the location of the verification material to be reviewed is less than a preset value, determined by the image recognition algorithm, proceed to the next step;
[0011] S13 forwards the write-off materials to the approvers, and determines the emergency assessment value of the write-off materials based on the remaining approval time of the write-off materials, the number of remaining approvers, and the number of approval items that the remaining approvers are responsible for, and sorts and tags the write-off materials of the approvers based on the emergency assessment value.
[0012] S14 determines the base time for reviewing the write-off materials based on the type of the write-off materials, the loan amount, and the emergency assessment value, and guides the approver to review the pending review positions of the write-off materials. When the approver's review time is longer than the base time, the approver's focus is assessed based on the facial image of the approver during the review. The approver's review detail is constructed based on the focus, review time, and review time of different pending review positions during the review. Based on the review detail, it is determined whether the review is completed and the completed write-off materials are generated.
[0013] Furthermore, the contract information includes borrower information, loan time and loan amount, agreed repayment time, and contract number; the borrower information includes repayment information, actual repayment time, and overdue time.
[0014] Furthermore, the determination that the contract information has not yet been cancelled specifically includes:
[0015] Based on the contract information, obtain borrower information, loan time and loan amount, and contract number;
[0016] Use the contract number that has been cancelled to determine if there is a contract that has been cancelled and matches the contract number in the contract information. If yes, proceed to the next step; otherwise, determine that it has not yet been cancelled.
[0017] The borrower information of the contract that has been written off and matches the contract number in the contract information is used to determine whether it matches the borrower information in the contract information. If it does, proceed to the next step; if not, it is determined that it has not been written off.
[0018] The loan time and loan amount of the previously cancelled contract are reconfirmed using the borrower information that matches the loan time and loan amount in the contract information. If they match the loan time and loan amount in the contract information, the contract is confirmed to have been cancelled.
[0019] Furthermore, the locations to be reviewed include the stamping location, the location for filling in contract information, and the location for filling in borrower information.
[0020] Furthermore, determining whether there is a possibility of omission based on the number of characters at the location to be reviewed specifically includes:
[0021] Based on the location to be reviewed, determine the text setting type and the number of text settings for each text setting type for the location to be reviewed;
[0022] Obtain the text type and the number of characters of each text type in the location to be reviewed, and combine the text setting type and the number of characters of each text setting type in the location to be reviewed to determine whether there is any possibility of omission.
[0023] Furthermore, based on the aforementioned emergency assessment value, the reimbursement materials of the approvers are sorted and tagged, specifically including:
[0024] Based on the urgency assessment value of the reimbursement materials by the approvers, the reimbursement materials of the approvers are divided into urgent reimbursement materials and general reimbursement materials, and the urgent reimbursement materials are processed to meet the standards.
[0025] The emergency write-off materials are sorted according to their emergency assessment value.
[0026] The general write-off materials are sorted according to the loan amount in the general write-off materials.
[0027] In a second aspect, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that: when the processor runs the computer program, it executes the above-described online approval method for automatically generating verification materials through process configuration.
[0028] Thirdly, the present invention provides a computer storage medium storing a computer program, which, when executed in a computer, causes the computer to execute the above-mentioned online approval method for automatically generating verification materials through process configuration.
[0029] The beneficial effects of this invention are as follows:
[0030] By first determining whether a document has not yet been written off based on contract information, duplicate writing off materials that have already been written off are avoided. This not only reduces the number of approvals required from the approvers but also improves the efficiency of the approval process and ensures the quality of the approval.
[0031] By utilizing the remaining approval time and number of remaining approvers for the reimbursement materials, and combining this with the number of approval items each remaining approver is responsible for, the urgency assessment value of the reimbursement materials is confirmed. This allows for the assessment of the urgency of the reimbursement materials from the perspectives of the approval time limit, the number of remaining approvers, and the quantity of materials. This enables the assessment of the urgency of the reimbursement materials from different angles, which also lays the foundation for differentiated labeling processing.
[0032] By combining the focus of the approver at different locations during the approval process, the approval time, and the approval time at different locations, the level of detail of the approver's review is constructed. This enables the evaluation of the level of detail from two perspectives: the facial image of the approver and the approval time. This avoids the occurrence of low-quality verification materials due to insufficient level of detail in the approval process.
[0033] Other features and advantages will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.
[0034] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description
[0035] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.
[0036] Figure 1 This is a flowchart of an online approval method that configurates and automatically generates verification materials.
[0037] Figure 2 This is a flowchart illustrating the method for generating problem probabilities;
[0038] Figure 3 This is a flowchart illustrating the method for constructing emergency assessment values;
[0039] Figure 4 This is a flowchart outlining the specific steps involved in determining the baseline review time.
[0040] Figure 5 This is a flowchart outlining the specific steps involved in determining the level of detail in the review process;
[0041] Figure 6 It is a framework diagram of a computer system. Detailed Implementation
[0042] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.
[0043] To solve the above problems, according to one aspect of the present invention, such as Figure 1 As shown, according to one aspect of the present invention, an online approval method for automatically generating verification materials through process configuration is provided, characterized in that it specifically includes:
[0044] S11 obtains contract information and loan information to be written off, and if it is determined from the contract information that the contract information has not yet been written off, generates the write-off materials to be approved using the contract information and loan information;
[0045] It should be noted that the contract information includes borrower information, loan time and loan amount, agreed repayment time, and contract number; the borrower information includes repayment information, actual repayment time and overdue time.
[0046] For specific examples, determining that the contract information has not yet been cancelled includes:
[0047] Based on the contract information, obtain borrower information, loan time and loan amount, and contract number;
[0048] Use the contract number that has been cancelled to determine if there is a contract that has been cancelled and matches the contract number in the contract information. If yes, proceed to the next step; otherwise, determine that it has not yet been cancelled.
[0049] The borrower information of the contract that has been written off and matches the contract number in the contract information is used to determine whether it matches the borrower information in the contract information. If it does, proceed to the next step; if not, it is determined that it has not been written off.
[0050] The loan time and loan amount of the previously cancelled contract are reconfirmed using the borrower information that matches the loan time and loan amount in the contract information. If they match the loan time and loan amount in the contract information, the contract is confirmed to have been cancelled.
[0051] In this embodiment, by first determining whether the contract information has not yet been processed for cancellation, the duplicate generation of cancellation materials after cancellation is avoided. This not only reduces the number of approvals required by the approvers, but also improves the efficiency of the approval process and ensures the quality of the approval.
[0052] S12 If the probability of a problem at the location of the verification material to be reviewed is less than a preset value, determined by the image recognition algorithm, proceed to the next step;
[0053] Specifically, the locations to be reviewed include the stamping location, the location for filling in contract information, and the location for filling in borrower information.
[0054] Specific examples, such as Figure 2 As shown, the method for generating the problem probability is as follows:
[0055] S21. Obtain an image of the location to be reviewed, extract image features based on the image, and obtain the error probability of the review location based on the image features;
[0056] Specifically, in the actual operation process, the texture features of the image are extracted, and the texture features are used to identify the error probability of the audit position using an image recognition model based on a CNN algorithm.
[0057] S22 obtains the number of characters in the position to be reviewed, and determines whether there is a possibility of omission based on the number of characters in the position to be reviewed. If yes, proceed to step S23; otherwise, construct the problem probability based on the error probability of the review position.
[0058] It is understandable that when the number of characters in the area to be reviewed is less than or greater than a certain number, there is a possibility of omission.
[0059] S23 determines the omission probability of the review location based on the image features of the review location, and constructs the problem probability by combining the error probability of the review location.
[0060] For specific examples, determining whether there is a possibility of omissions based on the number of characters at the location to be reviewed includes:
[0061] Based on the location to be reviewed, determine the text setting type and the number of text settings for each text setting type for the location to be reviewed;
[0062] Obtain the text type and the number of characters of each text type in the location to be reviewed, and combine the text setting type and the number of characters of each text setting type in the location to be reviewed to determine whether there is any possibility of omission.
[0063] S13 forwards the write-off materials to the approvers, and determines the emergency assessment value of the write-off materials based on the remaining approval time of the write-off materials, the number of remaining approvers, and the number of approval items that the remaining approvers are responsible for, and sorts and tags the write-off materials of the approvers based on the emergency assessment value.
[0064] Specific examples, such as Figure 3 As shown, the method for constructing the emergency assessment value is as follows:
[0065] S31 Obtain the remaining approval time for the reimbursement materials, and determine whether the reimbursement materials are urgent based on the remaining approval time. If so, construct the urgency assessment value of the reimbursement materials based on the preset urgency assessment value and the remaining approval time. If not, proceed to step S32.
[0066] Generally, there is a certain remaining approval time from the submission of verification materials to final approval. When the remaining approval time is short, the verification materials are considered to be relatively urgent.
[0067] It should be noted that the preset emergency assessment value can be determined based on the emergency assessment value of the selected benchmark reimbursement material, and the emergency assessment value of the reimbursement material can be constructed based on the difference between the reimbursement material and the benchmark reimbursement material.
[0068] S32 obtains the number of remaining approvers of the reimbursement materials and determines whether the number of remaining approvers is greater than the preset number of personnel. If yes, proceed to step S33; otherwise, proceed to step S34.
[0069] Understandably, when there are a large number of remaining approvers, it is necessary to determine the number of approval items each of the remaining approvers is responsible for.
[0070] S33 Obtain the number of approval items under the responsibility of the remaining approvers of the reimbursement materials to determine whether the reimbursement materials are urgent. If so, construct the urgency assessment value of the reimbursement materials based on the number of remaining approvers, the number of approval items under the responsibility of the remaining approvers, and the preset urgency assessment value. If not, proceed to step S34.
[0071] S34 determines the urgent assessment value of the write-off materials based on the remaining approval time, the number of remaining approvers, and the number of approval items each remaining approver is responsible for.
[0072] For specific examples, the reimbursement materials of the approvers are sorted and tagged based on the urgency assessment value, specifically including:
[0073] Based on the urgency assessment value of the reimbursement materials by the approvers, the reimbursement materials of the approvers are divided into urgent reimbursement materials and general reimbursement materials, and the urgent reimbursement materials are processed to meet the standards.
[0074] The emergency write-off materials are sorted according to their emergency assessment value.
[0075] The general write-off materials are sorted according to the loan amount in the general write-off materials.
[0076] In this embodiment, by utilizing the remaining approval time and number of remaining approvers of the reimbursement materials, and combining this with the number of approval items handled by the remaining approvers, the urgency assessment value of the reimbursement materials is confirmed. This allows for the assessment of the urgency of the reimbursement materials from the perspectives of the approval time limit, the number of remaining approvers, and the quantity of materials. This enables the assessment of the urgency of the reimbursement materials from different angles, which also lays the foundation for differentiated labeling processing.
[0077] S14 determines the base time for reviewing the write-off materials based on the type of the write-off materials, the loan amount, and the emergency assessment value, and guides the approver to review the pending review positions of the write-off materials. When the approver's review time is longer than the base time, the approver's focus is assessed based on the facial image of the approver during the review. The approver's review detail is constructed based on the focus, review time, and review time of different pending review positions during the review. Based on the review detail, it is determined whether the review is completed and the completed write-off materials are generated.
[0078] Specific examples, such as Figure 4 As shown, the specific steps for determining the basic review time are as follows:
[0079] S41 obtains the type of the write-off material, the number of pages and words of the write-off material, the number of pending review positions and the number of words in the write-off material, and determines the approval setting time for the write-off material based on the type of the write-off material, the number of pages and words of the write-off material, and the number of pending review positions and the number of words in the write-off material;
[0080] Specifically, the approval time for the verification materials can be determined by using a pre-defined form based on the type of verification materials, combined with the number of pages and words in the verification materials, as well as the number and word count of the pending review sections. Alternatively, the approval time for each pending review section can be determined based on the pre-defined approval time for each item's word count, thus obtaining the approval time for the verification materials.
[0081] S42 and determine whether the approval setting time needs to be modified based on the emergency assessment value of the reimbursement materials. If yes, proceed to step S44; otherwise, proceed to step S43.
[0082] S43 determines whether the approval setting time needs to be modified based on the approval authority of the approver of the reimbursement materials. If yes, proceed to step S44; otherwise, determine the basic approval time of the reimbursement materials based on the approval setting time of the reimbursement materials.
[0083] S44 determines the basic approval time for the write-off materials based on the approval setting time, emergency assessment value, loan amount, and the approval authority of the approver.
[0084] It should be noted that the greater the approval authority of the approver, the longer the basic approval time for the reimbursement materials.
[0085] Specific examples, such as Figure 5 As shown, the specific steps for determining the level of detail in the review are as follows:
[0086] S51 obtains the approval time of different pending review positions when the approver is making an approval, and determines whether the approval time of the different pending review positions meets the requirements. If yes, proceed to step S53; otherwise, proceed to step S52.
[0087] For example, if the approval time for location A to be reviewed is 3 minutes and the approval time for location B to be reviewed is 2 minutes, and the required approval time for location A to be reviewed is 2 minutes and the required approval time for location B to be reviewed is 1 minute, then it is determined that the approval time for both locations to be reviewed meets the requirements.
[0088] S52 is based on the number of audit locations that meet the requirements in the approval time of the different audit locations, the number of audit locations that do not meet the requirements in the approval time of the audit locations, and the ratio of the approval time of the audit locations that do not meet the requirements to the set approval time. Combined with the approval time of the approver when making the approval, an approval time assessment is obtained, and it is determined whether the approval time assessment meets the requirements. If yes, proceed to step S53. If no, it is determined that the approval time of the approver cannot meet the requirements, and the audit cannot be completed and the verification materials for the audit completed cannot be generated.
[0089] It should be noted that the approval time assessment is determined using a model based on the HHO-ELM algorithm. The inputs of the model are the number of approval locations that meet the requirements in the approval time of different approval locations, the number of approval locations that do not meet the requirements in the approval time of different approval locations, the ratio of the approval time of the approval location that does not meet the requirements to the set approval time, and the approval time of the approver when making the approval. The output is the approval time assessment.
[0090] For example, the specific steps for constructing the model based on the HHO-ELM algorithm are as follows:
[0091] (1) Initialization: In the ELM network model, a training sample set is given. This training sample set is determined by the number of pending review positions that meet the requirements in the approval time of different pending review positions, the number of pending review positions that do not meet the requirements in the approval time of the pending review positions, the ratio of the approval time of the pending review positions that do not meet the requirements to the approval time set, the approval time of the approver when making the approval, and the corresponding approval time evaluation quantity; the input weight matrix between the input layer and the hidden layer and the hidden layer bias b are randomly generated; the number of hidden layers L and the activation function g(x) of ELM are set. The HHO algorithm randomly assigns N Harris Eagles in the D-dimensional objective function solution space, and sets the parameter E o J is initialized, setting the population size N of Harris Eagles and the maximum number of iterations T. max Individual Harris Hawk X is composed of input weights and hidden layer biases, and the initial population is generated using ICMIC mapping;
[0092] 2) Calculation of training error: For each individual in the population, using the ELM network model with its randomly generated input weights, hidden layer biases, and training sample set, the output matrix H of the hidden layer is calculated, then the output weight matrix β is calculated, and the mean squared error (MSE) of each individual is calculated. The MSE is then used as the fitness function of HHO, and the minimum value of MSE is obtained.
[0093] (3) Iterative update: Update the escape energy E. Based on the relative magnitude of the escape energy E and the random number r, select the corresponding update strategy to update the position of the eagle flock. A new population can be obtained through the random expansion domain opposition learning strategy.
[0094] (4) Determine the termination condition: Determine whether the maximum number of iterations has been reached. If it has, output the optimal input weights and hidden layer biases of ELM, thereby constructing the HHO-ELM prediction model.
[0095] S53. Obtain facial images of the approver at different positions to be reviewed during the approval process, and determine the focus of the approver at different positions to be reviewed during the approval process based on the facial images. Determine whether the focus of different positions to be reviewed meets the requirements. If yes, proceed to step S55; otherwise, proceed to step S54.
[0096] S54 obtains the focus level of the approver at different locations to be reviewed during the approval process, and determines the locations to be reviewed that do not meet the focus requirements based on the focus level, and treats them as locations with insufficient focus. Based on the number of locations with insufficient focus, the minimum and average focus levels of the locations with insufficient focus, and combined with the number of locations to be reviewed that meet the focus requirements and the average focus level, a focus evaluation value is obtained. It is then determined whether the focus evaluation value meets the requirements. If yes, proceed to step S55. If no, it is determined that the approver's approval time cannot meet the requirements, and the review cannot be completed and the verification materials for the completed review cannot be generated.
[0097] S55 determines the level of detail in the review by the approver based on the focus assessment value and the approval time assessment value.
[0098] In this embodiment, the review detail of the approver is constructed by combining the focus of different review positions during the approval process, the approval time, and the approval time of different review positions. This enables the review detail to be evaluated from two perspectives: the facial image being reviewed and the approval time. This avoids the occurrence of low-quality verification materials due to the failure of the review detail to meet the requirements.
[0099] On the other hand, such as Figure 6 As shown, the present invention provides a computer system, including: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that: when the processor runs the computer program, it executes the above-described online approval method for automatically generating verification materials through process configuration.
[0100] The above-mentioned online approval method for automatically generating verification materials through process configuration specifically includes:
[0101] Obtain contract information and loan information to be written off, and when it is determined from the contract information that the contract information has not yet been written off, generate the write-off materials to be approved using the contract information and loan information;
[0102] When the probability of a problem in the pending audit location of the verification material is less than a preset value, the process proceeds to the next step.
[0103] The reimbursement materials are forwarded to the approvers, and the urgency assessment value of the reimbursement materials is determined based on the remaining approval time, the number of remaining approvers, and the number of approval items that the remaining approvers are responsible for. The reimbursement materials of the approvers are sorted and tagged based on the urgency assessment value.
[0104] Obtain the type of the verification material, the number of pages and words in the verification material, and the number and number of words in the pending review positions of the verification material; and determine the approval setting time for the verification material based on the type of the verification material, the number of pages and words in the verification material, and the number and number of words in the pending review positions of the verification material.
[0105] When it is determined that the approval setting time needs to be revised based on the emergency assessment value of the write-off materials, the basic approval time of the write-off materials is determined based on the approval setting time of the write-off materials, the emergency assessment value, the loan amount, and the approval authority of the approver.
[0106] The approver is guided to review the pending review positions of the verification materials. When the approver's review time exceeds the basic review time, the focus of the approver is assessed based on the facial image of the approver during the review. The review detail of the approver is constructed based on the focus of different pending review positions, the review time, and the review time of different pending review positions. Based on the review detail, it is determined whether the review is completed and the verification materials are generated after the review is completed.
[0107] On the other hand, the present invention provides a computer storage medium storing a computer program, which, when executed in a computer, causes the computer to execute the above-mentioned online approval method for automatically generating verification materials through process configuration.
[0108] The above-mentioned online approval method for automatically generating verification materials through process configuration specifically includes:
[0109] Obtain contract information and loan information to be written off, and when it is determined from the contract information that the contract information has not yet been written off, generate the write-off materials to be approved using the contract information and loan information;
[0110] When the probability of a problem in the pending audit location of the verification material is less than a preset value, the process proceeds to the next step.
[0111] The reimbursement materials are forwarded to the approvers, and the urgency assessment value of the reimbursement materials is determined based on the remaining approval time, the number of remaining approvers, and the number of approval items that the remaining approvers are responsible for. The reimbursement materials of the approvers are sorted and tagged based on the urgency assessment value.
[0112] Obtain the type of the verification material, the number of pages and words in the verification material, and the number and number of words in the pending review positions of the verification material; and determine the approval setting time for the verification material based on the type of the verification material, the number of pages and words in the verification material, and the number and number of words in the pending review positions of the verification material.
[0113] Based on the type of write-off materials, loan amount, and emergency assessment value, a base time for reviewing the write-off materials is determined, and the approvers are guided to review the pending sections of the write-off materials. When the approvers' review time exceeds the base time,
[0114] Obtain the approval time of different pending review positions when the approver is conducting the approval, and proceed to the next step when it is determined that the approval time of each different pending review position meets the requirements;
[0115] For example, if the approval time for location A to be reviewed is 3 minutes and the approval time for location B to be reviewed is 2 minutes, and the required approval time for location A to be reviewed is 2 minutes and the required approval time for location B to be reviewed is 1 minute, then it is determined that the approval time for both locations to be reviewed meets the requirements.
[0116] The system acquires facial images of the approver at different locations awaiting review during the approval process. Based on these images, it determines the approver's level of focus at each location. If not all locations meet the required level of focus, the system acquires the required level of focus at each location and identifies locations where focus is insufficient. Based on the number of locations with insufficient focus, the minimum and average level of focus at these locations, combined with the number of locations meeting the required level of focus and the average level of focus, a focus assessment value is obtained. If the focus assessment value meets the requirements, the system determines the approver's level of detail in review based on the focus assessment value and the approval time assessment value. Based on this level of detail, the system determines whether the review is complete and generates the verification materials for completed review.
[0117] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.
[0118] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.
[0119] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.
Claims
1. An online approval method for automatically generating verification materials through process configuration, characterized in that, Specifically, it includes: Obtain contract information and loan information to be written off, and when it is determined from the contract information that the contract information has not yet been written off, generate the write-off materials to be approved using the contract information and loan information; When the probability of a problem at the location of the verification material to be reviewed is less than a preset value, determined by the image recognition algorithm, proceed to the next step. The reimbursement materials are forwarded to the approvers, and the urgency assessment value of the reimbursement materials is determined based on the remaining approval time, the number of remaining approvers, and the number of approval items that the remaining approvers are responsible for. The reimbursement materials of the approvers are sorted and tagged based on the urgency assessment value. Based on the type of the write-off materials, the loan amount, and the emergency assessment value, the basic review time for the write-off materials is determined, and the approver is guided to review the pending review positions of the write-off materials. When the approver's review time exceeds the basic review time, the focus of the approver is assessed based on the facial image of the approver during the review. Based on the focus of the approver at different pending review positions, the review time, and the review time at different pending review positions, the approver's review detail is constructed, and based on the review detail, it is determined whether the review is completed and the completed write-off materials are generated. The method for generating the probability of the problem is as follows: Obtain an image of the location to be reviewed, extract image features based on the image, and obtain the error probability of the review location based on the image features; Obtain the number of characters in the position to be reviewed, and determine whether there is a possibility of omission based on the number of characters in the position to be reviewed. If yes, proceed to the next step; otherwise, construct the problem probability based on the error probability of the review position. The omission probability of the review location is determined based on the image features of the review location, and the problem probability is constructed by combining the error probability of the review location.
2. The online approval method for automatically generating verification materials through process configuration as described in claim 1, characterized in that, The contract information includes borrower information, loan time and loan amount, agreed repayment time, and contract number; the borrower information includes repayment information, actual repayment time, and overdue time.
3. The online approval method for automatically generating verification materials through process configuration as described in claim 2, characterized in that, The determination that the contract information has not yet been cancelled specifically includes: Based on the contract information, obtain borrower information, loan time and loan amount, and contract number; Use the contract number that has been cancelled to determine if there is a contract that has been cancelled and matches the contract number in the contract information. If yes, proceed to the next step; otherwise, determine that it has not yet been cancelled. The borrower information of the contract that has been written off and matches the contract number in the contract information is used to determine whether it matches the borrower information in the contract information. If it does, proceed to the next step; if not, it is determined that it has not been written off. The loan time and loan amount of the previously cancelled contract are reconfirmed using the borrower information that matches the loan time and loan amount in the contract information. If they match the loan time and loan amount in the contract information, the contract is confirmed to have been cancelled.
4. The online approval method for automatically generating verification materials through process configuration as described in claim 1, characterized in that, The areas to be reviewed include the stamping area, the area for filling in contract information, and the area for filling in borrower information.
5. The online approval method for automatically generating verification materials through process configuration as described in claim 1, characterized in that, The determination of whether there is a possibility of omission based on the number of characters at the location to be reviewed specifically includes: Based on the location to be reviewed, determine the text setting type and the number of text settings for each text setting type for the location to be reviewed; Obtain the text type and the number of characters of each text type in the location to be reviewed, and combine the text setting type and the number of characters of each text setting type in the location to be reviewed to determine whether there is any possibility of omission.
6. The online approval method for automatically generating verification materials through process configuration as described in claim 1, characterized in that, The method for constructing the emergency assessment value is as follows: S31 Obtain the remaining approval time for the reimbursement materials, and determine whether the reimbursement materials are urgent based on the remaining approval time. If so, construct the urgency assessment value of the reimbursement materials based on the preset urgency assessment value and the remaining approval time. If not, proceed to step S32. S32 obtains the number of remaining approvers of the reimbursement materials and determines whether the number of remaining approvers is greater than the preset number of personnel. If yes, proceed to step S33; otherwise, proceed to step S34. S33 Obtain the number of approval items under the responsibility of the remaining approvers of the reimbursement materials to determine whether the reimbursement materials are urgent. If so, construct the urgency assessment value of the reimbursement materials based on the number of remaining approvers, the number of approval items under the responsibility of the remaining approvers, and the preset urgency assessment value. If not, proceed to step S34. S34 determines the urgent assessment value of the write-off materials based on the remaining approval time, the number of remaining approvers, and the number of approval items each remaining approver is responsible for.
7. The online approval method for automatically generating verification materials through process configuration as described in claim 6, characterized in that, Based on the aforementioned emergency assessment value, the reimbursement materials of the approvers are sorted and tagged, specifically including: Based on the urgency assessment value of the reimbursement materials by the approvers, the reimbursement materials of the approvers are divided into urgent reimbursement materials and general reimbursement materials, and the urgent reimbursement materials are processed to meet the standards. The emergency write-off materials are sorted according to their emergency assessment value. The general write-off materials are sorted according to the loan amount in the general write-off materials.
8. The online approval method for automatically generating verification materials through process configuration as described in claim 1, characterized in that, The specific steps for determining the basic review time are as follows: S41 obtains the type of the write-off material, the number of pages and words of the write-off material, the number of pending review positions and the number of words in the write-off material, and determines the approval setting time for the write-off material based on the type of the write-off material, the number of pages and words of the write-off material, and the number of pending review positions and the number of words in the write-off material; S42 and determine whether the approval setting time needs to be modified based on the emergency assessment value of the reimbursement materials. If yes, proceed to step S44; otherwise, proceed to step S43. S43 determines whether the approval setting time needs to be modified based on the approval authority of the approver of the reimbursement materials. If yes, proceed to step S44; otherwise, determine the basic approval time of the reimbursement materials based on the approval setting time of the reimbursement materials. S44 determines the basic approval time for the write-off materials based on the approval setting time, emergency assessment value, loan amount, and the approval authority of the approver.
9. A computer system, comprising: A memory and processor connected by communication, and a computer program stored on the memory and capable of running on the processor, characterized in that: when the processor runs the computer program, it executes an online approval method for automatically generating verification materials through process configuration as described in any one of claims 1-8.
10. A computer storage medium storing a computer program that, when executed in a computer, causes the computer to execute an online approval method for automatically generating verification materials based on a process configuration, as described in any one of claims 1-8.