Credit evaluation processing method and apparatus

Abstract

The embodiment of the present specification provides a credit evaluation processing method and device, wherein the credit evaluation processing method comprises: in the credit evaluation processing process, inputting a to-be-evaluated image into a text extraction model for text extraction to obtain structured text, performing identity verification on user identity information contained in the structured text, and performing similarity verification on entity fields contained in the structured text and preset entity fields, if the verification passes, inputting the to-be-evaluated image and prompt text into a large language model for verification processing, obtaining a verification result, and synchronizing the structured text and the verification result to a credit evaluation platform.

Description

Technical Field

[0001] This document relates to the field of data processing technology, and in particular to a credit assessment processing method and apparatus. Background Technology

[0002] With the development of internet technology, computers and the internet have become an important part of users' lives and have gradually penetrated into every corner of life. As a result, more and more online services have emerged. For example, in the credit assessment scenario, the past method of manually verifying assessment materials has been abandoned, and credit assessment is carried out with the help of online credit assessment platforms and large models. In this process, how to better meet the needs of users and improve the accuracy and effectiveness of credit assessment has become an increasingly important focus. Summary of the Invention

[0003] This specification provides one or more embodiments of a credit assessment processing method, comprising: inputting an image to be assessed into a text extraction model for text extraction to obtain structured text; verifying the user identity information contained in the structured text, and verifying the similarity between entity fields contained in the structured text and preset entity fields; if the verification passes, inputting the image to be assessed and the prompt text into a large language model for verification processing to obtain a verification result; the verification processing includes: extracting and fusing image features in each verification dimension, and determining the verification result by calculating a verification score on the fused features; and synchronizing the structured text and the verification result to a credit assessment platform for credit assessment processing.

[0004] This specification provides one or more embodiments of a credit assessment processing apparatus, including: a text extraction module configured to input an image to be assessed into a text extraction model for text extraction to obtain structured text; a verification module configured to perform identity verification on user identity information contained in the structured text, and to perform similarity verification on entity fields contained in the structured text with preset entity fields; if the verification passes, a verification processing module is run, configured to input the image to be assessed and the prompt text into a large language model for verification processing to obtain a verification result; the verification processing includes: performing image feature extraction and feature fusion in each verification dimension, and determining the verification result by calculating a verification score on the fused features; and a credit assessment module configured to synchronize the structured text and the verification result to a credit assessment platform for credit assessment processing.

[0005] This specification provides one or more embodiments of a credit assessment processing device, including: a processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to: input an image to be assessed into a text extraction model for text extraction to obtain structured text; verify the user identity information contained in the structured text, and perform similarity verification on entity fields contained in the structured text with preset entity fields; if the verification passes, input the image to be assessed and the prompt text into a large language model for verification processing to obtain a verification result; the verification processing includes: extracting and fusing image features in each verification dimension, and determining the verification result by calculating a verification score on the fused features; and synchronizing the structured text and the verification result to a credit assessment platform for credit assessment processing.

[0006] This specification provides one or more embodiments of a computer-readable storage medium for storing computer-executable instructions, which, when executed, perform the following process: Inputting the image to be evaluated into a text extraction model for text extraction to obtain structured text; verifying the user identity information contained in the structured text, and verifying the similarity between the entity fields contained in the structured text and preset entity fields; if the verification passes, inputting the image to be evaluated and the prompt text into a large language model for verification processing to obtain verification results; the verification processing includes: extracting and fusing image features in each verification dimension, and determining the verification results by calculating verification scores on the fused features; synchronizing the structured text and the verification results to a credit assessment platform for credit assessment processing. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in one or more embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 A schematic diagram illustrating the implementation environment of a credit assessment processing method provided in one or more embodiments of this specification;

[0009] Figure 2 A flowchart of a credit assessment processing method provided for one or more embodiments of this specification;

[0010] Figure 3 A flowchart illustrating a credit assessment processing method applied to a credit limit assessment scenario, provided for one or more embodiments of this specification;

[0011] Figure 4 A schematic diagram of an embodiment of a credit assessment processing apparatus provided for one or more embodiments of this specification;

[0012] Figure 5 This is a schematic diagram of the structure of a credit assessment processing device provided for one or more embodiments of this specification. Detailed Implementation

[0013] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0014] The credit assessment processing method provided in one or more embodiments of this specification is applicable to the implementation environment of a credit assessment system. (Refer to...) Figure 1 The implementation environment includes at least:

[0015] Server 101, text extraction model 102, credit assessment platform 103, large language model 104, and user terminal 105 may also be included;

[0016] The text extraction model 102 and the large language model 104 can run on server 101 or on other servers. They can run on the same server or on different servers. Server 101 can be a single server, a server cluster consisting of several servers, or one or more cloud servers in a cloud computing platform. The text extraction model 102 is used to receive the image to be evaluated sent by server 101 and extract text from the image to be evaluated to obtain image text. The large language model 104 is used to receive the image to be evaluated and the prompt text sent by server 101 and perform real-world verification based on the image to be evaluated and the prompt text to obtain the verification result.

[0017] Credit assessment platform 103 is used to conduct credit assessments based on the images, text, and verification results synchronized by server 101;

[0018] The user terminal 105 has an application installed. The user can upload an image to be evaluated through the application installed in the user terminal 105. The user terminal 105 uploads the image to be evaluated to the server 101. The user terminal 105 can be a mobile phone, personal computer, tablet computer, e-book reader, VR (Virtual Reality) based information interaction device, vehicle terminal, IoT device, wearable smart device, laptop computer and desktop computer, etc.

[0019] In this implementation environment, during the credit assessment process, server 101 receives the image to be assessed sent by user terminal 105, inputs the image to be assessed into text extraction model 102 for text extraction, and obtains image text. Server 101 performs identity verification based on the user identity information contained in the image text, and performs content verification on the entity information contained in the image text. If the verification is successful, server 101 inputs the image to be assessed and the prompt text into large language model 104 for verification. After obtaining the verification result, the image text and verification result are synchronized to credit assessment platform 103 for efficient and accurate credit assessment.

[0020] This specification provides one or more embodiments of a credit assessment processing method as follows:

[0021] Reference Figure 2 The credit assessment processing method provided in this embodiment specifically includes steps S202 to S208.

[0022] Step S202: Input the image to be evaluated into the text extraction model to extract text and obtain structured text.

[0023] The image to be evaluated in this embodiment refers to a certificate image related to the user's credit assessment. This image can be a photograph of a paper certificate, such as a graduation certificate, property ownership certificate, housing provident fund statement, or individual income tax statement. The text extraction model can be a pre-trained large language model specifically designed for text extraction, or other models with text extraction capabilities.

[0024] The structured text refers to text arranged in a structured form. When the text extraction model outputs structured text, it can output the text in a structured form, such as in the form of graduation certificate (name, graduation certificate number, school, degree, graduation time), property certificate (property owner's name, property certificate number, property area), housing provident fund (housing provident fund holder's name, housing provident fund transaction data), and individual income tax (taxpayer's name, tax details).

[0025] In practice, the image to be evaluated can be sent from the user terminal to the server. Upon receiving the image, the server inputs it into a text extraction model to extract structured text. In one optional implementation of this embodiment, the image to be evaluated is obtained in the following manner:

[0026] Obtain the image to be evaluated sent by the user terminal.

[0027] Optionally, the image to be evaluated is uploaded to the user terminal by the user through an application installed on the user terminal.

[0028] Specifically, users upload images to be evaluated through the credit assessment interface provided by the application installed on their terminals. After receiving the images, the user terminals send them to the server. The server then retrieves the images sent by the user terminals and inputs them into a text extraction model to extract structured text.

[0029] It should be noted that users need to log in to the application before they can upload the image to be evaluated. That is, the user must log in to the application on the user's terminal to grant the user permission to upload the image to be evaluated.

[0030] In the specific execution process, text extraction is performed after obtaining the image to be evaluated. To improve the efficiency and accuracy of text extraction, a pre-trained text extraction model can be used for text recognition and extraction. In one optional implementation method provided in this embodiment, the text extraction model performs text extraction in the following manner:

[0031] The image to be evaluated is subjected to text recognition to obtain the text recognition result;

[0032] Based on the text recognition results, a text extraction strategy is determined, and the text is extracted from the text recognition results according to the text extraction strategy to obtain the image text.

[0033] The text extraction strategy can be a structured extraction strategy, which allows the text recognition results to be extracted in a structured form. It should be noted that different types of images to be evaluated correspond to different text extraction strategies. For example, after obtaining a first text recognition result from a graduation certificate image, a first text extraction strategy is determined based on this first text recognition result, and text extraction is performed on the graduation certificate image according to the first text extraction strategy. Similarly, after obtaining a second text recognition result from a property ownership certificate image, a second text extraction strategy is determined based on this second text recognition result, and text extraction is performed on the property ownership certificate image according to the second text recognition strategy.

[0034] Specifically, in the process of text recognition of the image to be evaluated, the image to be evaluated can be preprocessed first to optimize the image quality. Then, the text regions in the preprocessed image are detected and the detected text regions are converted into text recognition results. Since the image to be evaluated may have various types, such as graduation certificate images, property certificate images, etc., in the process of text extraction, the text extraction strategy can be determined according to the text recognition results. The text extraction strategy corresponding to the text recognition results is used to extract the text from the text recognition results to obtain structured text.

[0035] For example, text recognition is performed on a graduation certificate image to obtain text recognition results such as graduation certificate (name, graduation certificate number, school, academic qualification, graduation time, major name, date of birth, signature). In this case, according to the text extraction strategy corresponding to the graduation certificate image, the major name, date of birth, and signature in the graduation certificate image may not need to be extracted. Only the text corresponding to the name, graduation certificate number, school, academic qualification, and graduation time is extracted.

[0036] Here, by leveraging the natural language processing capabilities of the text extraction model, we can gain a deeper understanding of the contextual relationships in structured text and capture semantic changes, thereby improving the accuracy of text extraction and increasing processing efficiency.

[0037] Step S204 involves verifying the user identity information contained in the structured text and verifying the similarity between the entity fields contained in the structured text and preset entity fields.

[0038] After extracting the text from the image to be evaluated into the text extraction model to obtain structured text, this step verifies the user identity information contained in the structured text to detect whether the person to whom the image to be evaluated belongs is the user who uploaded the image to be evaluated. It also verifies the similarity between the entity fields contained in the structured text and the preset entity fields to detect whether the image to be evaluated is similar to historical evaluation images. If they are similar, it means that the image to be evaluated has been used for evaluation and cannot be used again for credit evaluation processing.

[0039] It should be noted that the above step S204, which verifies the user identity information contained in the structured text and verifies the similarity between the entity fields contained in the structured text and the preset entity fields, can also be replaced by: verifying the user identity information contained in the structured text and verifying the fields based on the entity fields contained in the structured text.

[0040] The similarity verification refers to verifying the similarity of fields. This similarity verification can be either a field similarity verification or a field-specific verification. Specifically, the similarity verification result can be determined by calculating the field similarity between entity fields in structured text and preset entity fields.

[0041] The entity fields refer to the specific content contained in the structured text, such as the name, graduation certificate number, school, academic qualification, and / or graduation date in a graduation certificate. In the specific execution process, to improve the credibility of credit assessment processing based on the image to be evaluated, after extracting the structured text from the image, the user identity information and entity fields contained in the image text are verified. Specifically, identity verification is performed based on the user identity information, and similarity verification is performed based on the entity fields. If both verifications pass, the verification is considered successful; if either verification fails or both fail, the verification is considered unsuccessful. The following sections provide detailed explanations of the identity verification process and the similarity verification process:

[0042] (1) Identity verification

[0043] In the specific process of identity verification, the user identity information contained in the structured text is compared with the logged-in user information to determine whether the person to be evaluated is consistent with the user who uploaded the image to be evaluated. In an optional implementation method provided in this embodiment, the user identity information contained in the structured text is verified in the following way:

[0044] Query the login user information of the application that collected the image to be evaluated;

[0045] The user's identity information is compared with the logged-in user information. If they match, the identity verification is deemed successful; otherwise, the identity verification is deemed unsuccessful.

[0046] Specifically, the user who uploaded the image to be evaluated is identified by querying the application's login user information. The user's identity information is then compared with the login user information. If they match, it means that the person to be evaluated and the user who uploaded the image are the same person, and the identity verification is confirmed to be successful. If they do not match, it means that the person to be evaluated and the user who uploaded the image are not the same person, and the identity verification is confirmed to be unsuccessful.

[0047] (2) Similarity verification

[0048] In the specific process of similarity verification, the field similarity between each entity field in the structured text and a preset entity field is calculated. The text similarity is then calculated based on the field weight and field similarity of each entity field. If the text similarity is less than a similarity threshold, the similarity verification is considered passed; if the text similarity is greater than or equal to the similarity threshold, the similarity verification is considered failed. In one optional implementation of this embodiment, the similarity verification is performed based on the entity fields contained in the structured text in the following manner:

[0049] Calculate the field similarity between each entity field in the structured text and a preset entity field;

[0050] Text similarity is calculated based on the field weights and field similarities of each entity field;

[0051] If the text similarity is less than the similarity threshold, then the similarity verification is deemed successful.

[0052] The preset entity fields can be extracted from historical structured text; the historical structured text can be obtained by extracting text from historical evaluation images uploaded by other users in the application that have already undergone credit assessment processing, and the historical evaluation images can be stored in the application's database.

[0053] Specifically, during the similarity verification process, the field similarity between each entity field in the structured text and the preset entity fields is calculated. The text similarity is calculated based on the field weight and field similarity of each entity field. If the text similarity is greater than or equal to the similarity threshold, it indicates that the image to be evaluated has been used for credit assessment in the past and cannot be reused, thus the similarity verification fails. If the text similarity is less than the similarity threshold, it indicates that the image to be evaluated has not been used for credit assessment in the past and can be used for credit assessment, thus the similarity verification passes.

[0054] For example, if the image to be evaluated is a graduation certificate image, text extraction is performed on the graduation certificate image to obtain structured text such as graduation certificate number, school, degree, and graduation time. Each entity field is the specific content of graduation certificate number, school, degree, and graduation time. The field similarity of each entity field with the preset entity fields is calculated. The text similarity is calculated based on the field weight and field similarity of each entity field. If the text similarity is less than the similarity threshold, the similarity verification is considered passed.

[0055] In addition, during the similarity verification process, text similarity can be skipped and the field similarity between each entity field in the obtained structured text and the preset entity field can be directly calculated. If the field similarity is less than the similarity threshold, the similarity verification is deemed to have passed.

[0056] It should be noted that the above-mentioned two verification processes—identity verification of the user identity information contained in the structured text and similarity verification based on the entity fields contained in the structured text—can be performed in parallel. For example, after obtaining the structured text, identity verification of the user identity information contained in the structured text and similarity verification based on the entity fields contained in the structured text can be performed simultaneously. Alternatively, one verification can be performed after the other is passed. For example, after the identity verification of the user identity information contained in the structured text is passed, similarity verification based on the entity fields contained in the structured text can be performed. Or, for example, after the similarity verification based on the entity fields contained in the structured text is passed, identity verification of the user identity information contained in the structured text can be performed. This embodiment does not limit this.

[0057] Step S206: If the verification passes, input the image to be evaluated and the prompt text into the large language model for verification processing to obtain the verification result.

[0058] After verifying the user identity information contained in the structured text and performing similarity verification based on the entity fields contained in the structured text, in this step, if both verifications pass, the image to be evaluated and the prompt text are input into the large language model for verification processing to obtain the verification result.

[0059] The verification process refers to verifying the authenticity of the image to be evaluated, or it may also involve performing image enhancement or image modification verification on the image to be evaluated to check whether the image to be evaluated is a forged image, or to check whether the image to be evaluated has been artificially processed, such as checking whether the image has been processed by Photoshop (Adobe Photoshop).

[0060] Optionally, the verification process includes: extracting and fusing image features in each verification dimension, and determining the verification result by calculating the verification score of the fused features; specifically, firstly, image features are extracted from the image to be evaluated in each verification dimension to obtain image features, then the image features are fused to obtain fused features, the verification score is calculated based on the fused features to obtain a verification score, and finally the verification result is determined based on the verification score.

[0061] In the specific process of verifying the image to be evaluated, in order to improve the efficiency and accuracy of image verification, a large language model can be used to verify the image to be evaluated. Furthermore, in order to improve the verification effect of the large language model, the prompt text and the image to be evaluated can be input into the large language model together to guide the large language model to perform verification processing. The prompt text refers to the text used to instruct the large language model to perform a specific task or generate a specific type of answer or output. Here, the prompt text can be used to instruct the large language model to perform verification tasks and generate verification results in a preset format.

[0062] In one optional implementation method provided in this embodiment, the verification process is performed in the following manner:

[0063] Extract image features from each verification dimension of the image to be evaluated;

[0064] The image features are subjected to feature fusion processing to obtain fused features;

[0065] The verification score is calculated based on the fusion features, and the verification result is determined based on the calculated verification score.

[0066] Optionally, the verification result includes verification pass result and verification fail result.

[0067] In the specific execution process, the verification dimension refers to the image feature dimension that affects the authenticity of the image. The verification dimension includes at least one of the following: image noise dimension, pixel dimension, and image structure dimension.

[0068] Based on this, during the verification process of the large language model, the image to be evaluated can be input into the feature extraction module of the large language model to extract image features of each verification dimension. Based on the feature fusion module, the image features are fused to obtain fused features. The fused features are input into the activation function to obtain the verification score. The preset verification result corresponding to the verification score interval where the verification score is located is used as the verification result.

[0069] For example, after fusing noise features, pixel features and / or image structure features of the image to be evaluated to obtain fused features, the verification score of the fused features is calculated, and the preset verification result corresponding to the verification score interval is taken as the verification result.

[0070] Here, the large language model continuously learns from new samples, updating its knowledge base in real time to adapt to and identify emerging material types and image anomaly types. It can identify anomalies in images by learning from vast amounts of data and automatically mark potentially tampered content. Furthermore, combined with multimodal data processing capabilities such as image recognition, the large language model can also assist in the analysis of non-textual information (such as images and tables), further enriching the verification processing dimensions and enhancing overall security. Moreover, utilizing transfer learning and incremental learning techniques, the large language model can quickly adapt to new needs and technological contexts while maintaining existing knowledge, making the entire system more flexible and scalable.

[0071] Step S208: Synchronize the structured text and the verification result with the credit assessment platform for credit assessment processing.

[0072] After inputting the image to be evaluated and the prompt text into the large language model for verification processing to obtain the verification result, in this step, the structured text and verification result are synchronized to the credit assessment platform for credit assessment processing. After this, the credit assessment platform may return a credit assessment failure result or a credit assessment result.

[0073] In the specific execution process, the structured text and verification results can be synchronized to the credit assessment platform via interface calls, so that the credit assessment platform can perform credit assessment processing. In an optional implementation method provided in this embodiment, the credit assessment platform performs credit assessment processing in the following way:

[0074] If the verification result is successful, extract the entity fields used for credit assessment from the structured text; calculate the credit limit variable according to the preset assessment rules and the entity fields, and generate a credit assessment result containing the credit limit variable;

[0075] If the verification result is that the verification failed, a credit assessment failure result will be returned.

[0076] Specifically, if the verification result is successful, the entity fields used for credit assessment can be extracted from the structured text, the credit limit variable corresponding to the entity field can be found from the preset assessment rules, and a credit assessment result containing the credit limit variable can be generated.

[0077] For example, if the credit assessment platform recognizes that the verification result is successful, it extracts the first degree information from the structured text, finds the corresponding credit limit variable x from the preset assessment rules, and thus determines the credit limit variable x when the extracted educational information in the structured text is the first degree information.

[0078] Furthermore, during the credit assessment process on the credit assessment platform, a credit score can be generated, and a credit limit variable can be calculated based on the credit score. In another optional implementation provided in this embodiment, the credit assessment platform performs credit assessment in the following manner:

[0079] If the verification result is successful, extract the entity fields used for credit assessment from the structured text; calculate the credit score according to the scoring rules and the extracted entity fields, and generate a credit assessment result containing credit limit variables based on the credit score;

[0080] If the verification result is that the verification failed, a credit assessment failure result will be returned.

[0081] Specifically, when the credit assessment platform identifies the verification result as passed, credit assessment processing is carried out. In the process of credit assessment, entity fields used for credit assessment are extracted from the structured text, credit scores are calculated according to the scoring rules and the extracted entity fields, and credit assessment results containing credit limit variables are generated based on the credit scores.

[0082] In the specific execution process, after the credit assessment platform generates the credit limit variable, it generates a credit assessment result containing the credit limit variable and returns it to the server. Correspondingly, the server receives the credit assessment result containing the credit limit variable returned by the credit assessment platform and updates the user's remaining credit limit. In an optional implementation method provided in this embodiment, the remaining credit limit is updated in the following way:

[0083] The user's current credit limit is calculated based on the user's available credit limit and the credit assessment results returned by the credit assessment platform;

[0084] The available credit limit is updated based on the current credit limit to obtain an updated credit limit.

[0085] Specifically, the system receives the credit assessment results returned by the credit assessment platform, which include credit limit variables, and queries the user's available credit limit. Based on the credit limit variables in the credit assessment results and the available credit limit, the system calculates the current credit limit and updates the available credit limit based on the current credit limit to obtain an updated credit limit.

[0086] For example, if the credit limit variable in the received credit assessment result is n, and the available credit limit of the user is x, then x+n is taken as the current credit limit, and the remaining credit limit is updated to x+n.

[0087] Furthermore, to improve user experience, after updating the user's remaining credit limit, credit limit update information containing the updated credit limit can be generated and sent to the user's terminal. In one optional implementation of this embodiment, after the step of updating the available credit limit based on the current credit limit is performed, the method further includes:

[0088] Generate a credit limit update notification that includes the updated credit limit;

[0089] The credit limit update notification is sent to the user's terminal so that the credit limit update information can be displayed through an application installed on the user's terminal.

[0090] Furthermore, if the credit assessment platform detects that the verification result has failed, it can also return a credit assessment failure result to the server. Correspondingly, the server receives the credit assessment failure result returned by the credit assessment platform and performs manual review of the image to be assessed to avoid misjudgment and improve reliability. In one optional implementation of this embodiment, after the step of synchronizing the structured text and the verification result with the credit assessment platform for credit assessment is executed, the following is also included:

[0091] If the credit assessment fails, the image to be assessed is marked, and the marked image is sent to the reviewer for review.

[0092] Specifically, if a credit assessment failure result is received from the credit assessment platform, the image to be assessed is marked and sent to the reviewer for manual review. Alternatively, after receiving a failure result from the credit assessment platform, the image to be assessed can be sent directly to the reviewer for secondary review without marking it.

[0093] It should be noted that, through large-scale data training, the large language model can also uncover potential trends and patterns during the verification process of images to be evaluated, providing important references for subsequent risk warnings and decision support. For example, the large language model records and reports frequently occurring image anomaly types during the verification process, allowing reviewers to take corresponding preventative measures for these types; image anomaly types could include anomalies such as names being altered in graduation certificate images.

[0094] It should also be noted that the above step S208, which synchronizes the structured text and the verification result to the credit assessment platform for credit assessment processing, can also be replaced by: synchronizing the structured text to the credit assessment platform for credit assessment processing if the verification is passed. That is, after obtaining the verification result, the server identifies the verification result. If the identified verification result is that the verification is passed, the server synchronizes the structured text to the credit assessment platform for credit assessment processing.

[0095] In summary, the credit assessment processing method provided in this embodiment first inputs the image to be assessed into a text extraction model for text extraction. Leveraging the text extraction capabilities of the model, relatively accurate structured text is extracted. Next, the user identity information contained in the structured text is verified, and similarity verification is performed based on the entity fields contained in the structured text to ensure the security of the credit assessment. If the verification passes, the image to be assessed and the prompt text are input into a large language model for verification processing to check the authenticity of the image and ensure that it has not been abnormally modified. After obtaining the verification result, the structured text and verification result are synchronized to the credit assessment platform for credit assessment processing. This improves the automation and accuracy of credit assessment and enables efficient assessment and flexible updating of credit limits.

[0096] The following example uses a credit assessment processing method provided in this embodiment to illustrate its application in a credit limit assessment processing scenario. Figure 3 The credit assessment processing method provided in this embodiment will be further explained below. Figure 3 The credit assessment and processing method applied to credit limit assessment and processing scenarios includes the following steps.

[0097] Step S302: Input the image to be evaluated into the text extraction model to extract text and obtain structured text.

[0098] Step S304: Query the login user information of the application that collects the images to be evaluated.

[0099] Step S306: Compare the user identity information contained in the structured text with the logged-in user information. If they match, the identity verification is confirmed to be successful.

[0100] Step S308: Calculate the field similarity between each entity field in the structured text and the preset entity fields.

[0101] Step S310: If the field similarity is less than the similarity threshold, the similarity verification is deemed successful.

[0102] Step S312: Extract image features of each verification dimension of the image to be evaluated, and perform feature fusion processing to obtain fused features.

[0103] Step S314: Calculate the verification score based on the fusion features, and determine the verification result based on the verification score.

[0104] Step S316: If the verification is successful, synchronize the structured text to the credit assessment platform for credit assessment processing.

[0105] Step S318: Calculate the user's current credit limit based on the user's available credit limit and the credit assessment results returned by the credit assessment platform.

[0106] Step S320: Update the available credit limit based on the current credit limit to obtain the updated credit limit.

[0107] Step S322: Generate a credit limit update reminder that includes the updated credit limit.

[0108] Step S324: Send a credit limit update reminder to the user terminal so that the credit limit update information can be displayed through the application installed on the user terminal.

[0109] It should be noted that any one or more steps in steps S302 to S324 can be combined with any one or more steps in steps S202 to S208 to form a new implementation method according to the needs of implementation and deployment. In addition, any one or more technical features in steps S302 to S324 can be selected and combined with any one or more technical features provided in steps S202 to S208 to form a new implementation method according to the actual deployment needs. Alternatively, any one or more technical features in steps S302 to S324 can be replaced with any one or more technical features provided in steps S202 to S208 to form a new implementation method according to the actual deployment needs. These will not be elaborated on here.

[0110] This specification provides an embodiment of a credit assessment processing device as follows:

[0111] In the above embodiments, a credit assessment processing method is provided, and correspondingly, a credit assessment processing apparatus is also provided, which will be described below with reference to the accompanying drawings.

[0112] Reference Figure 4 This illustration shows a schematic diagram of an embodiment of a credit assessment processing device provided in this embodiment.

[0113] Since the apparatus embodiments correspond to the method embodiments, the descriptions are relatively simple. For relevant parts, please refer to the corresponding descriptions of the method embodiments provided above. The apparatus embodiments described below are merely illustrative.

[0114] This embodiment provides a credit assessment processing device, the device comprising:

[0115] The text extraction module 402 is configured to input the image to be evaluated into the text extraction model for text extraction to obtain structured text;

[0116] The verification module 404 is configured to verify the user identity information contained in the structured text and to verify the similarity between the entity fields contained in the structured text and preset entity fields.

[0117] If the verification passes, the verification processing module 406 is run. The verification processing module 406 is configured to input the image to be evaluated and the prompt text into a large language model for verification processing to obtain the verification result. The verification processing includes: performing image feature extraction and feature fusion for each verification dimension, and determining the verification result by calculating the verification score of the fused features.

[0118] The credit assessment module 408 is configured to synchronize the structured text and the verification results to the credit assessment platform for credit assessment processing.

[0119] This manual provides an example of a credit assessment processing device as follows:

[0120] Corresponding to the credit assessment processing method described above, based on the same technical concept, one or more embodiments of this specification also provide a credit assessment processing device for executing the credit assessment processing method provided above. Figure 5 This is a schematic diagram of the structure of a credit assessment processing device provided for one or more embodiments of this specification.

[0121] This embodiment provides a credit assessment processing device, including:

[0122] like Figure 5As shown, credit assessment processing devices can vary significantly due to differences in configuration or performance. They may include one or more processors 501 and memory 502, with memory 502 storing one or more application programs or data. Memory 502 can be temporary or persistent storage. The application programs stored in memory 502 may include one or more modules (not shown), each module including a series of computer-executable instructions from the credit assessment processing device. Furthermore, processor 501 may be configured to communicate with memory 502, executing the series of computer-executable instructions in memory 502 on the credit assessment processing device. The credit assessment processing device may also include one or more power supplies 503, one or more wired or wireless network interfaces 504, one or more input / output interfaces 505, one or more keyboards 506, etc.

[0123] In one specific embodiment, the credit assessment processing apparatus includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the credit assessment processing apparatus, and is configured to be executed by one or more processors. The one or more programs include computer-executable instructions for performing the following:

[0124] The image to be evaluated is input into the text extraction model for text extraction to obtain structured text;

[0125] The structured text contains user identity information for identity verification, and the structured text contains entity fields for similarity verification with preset entity fields.

[0126] If the verification passes, the image to be evaluated and the prompt text are input into a large language model for verification processing to obtain the verification result. The verification processing includes: extracting and fusing image features in each verification dimension, and determining the verification result by calculating the verification score of the fused features.

[0127] The structured text and the verification results are synchronized with the credit assessment platform for credit assessment processing.

[0128] This specification provides an embodiment of a computer-readable storage medium as follows:

[0129] Corresponding to the credit assessment processing method described above, and based on the same technical concept, one or more embodiments of this specification also provide a computer-readable storage medium.

[0130] The computer-readable storage medium provided in this embodiment is used to store computer-executable instructions, which, when executed, implement the following process:

[0131] The image to be evaluated is input into the text extraction model for text extraction to obtain structured text;

[0132] The structured text contains user identity information for identity verification, and the structured text contains entity fields for similarity verification with preset entity fields.

[0133] If the verification passes, the image to be evaluated and the prompt text are input into a large language model for verification processing to obtain the verification result. The verification processing includes: extracting and fusing image features in each verification dimension, and determining the verification result by calculating the verification score of the fused features.

[0134] The structured text and the verification results are synchronized with the credit assessment platform for credit assessment processing.

[0135] It should be noted that the embodiments of a computer-readable storage medium described in this specification and the embodiments of a credit assessment processing method described in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can be referred to the implementation of the corresponding method described above, and the repeated parts will not be described again.

[0136] This specification provides an example of a computer program product as follows:

[0137] Corresponding to the credit assessment processing method described above, and based on the same technical concept, one or more embodiments of this specification also provide a computer program product.

[0138] A computer program product includes a computer program / instructions that, when executed by a processor, perform the following steps:

[0139] The image to be evaluated is input into the text extraction model for text extraction to obtain structured text;

[0140] The structured text contains user identity information for identity verification, and the structured text contains entity fields for similarity verification with preset entity fields.

[0141] If the verification passes, the image to be evaluated and the prompt text are input into a large language model for verification processing to obtain the verification result. The verification processing includes: extracting and fusing image features in each verification dimension, and determining the verification result by calculating the verification score of the fused features.

[0142] The structured text and the verification results are synchronized with the credit assessment platform for credit assessment processing.

[0143] It should be noted that the embodiments of a computer program product described in this specification and the embodiments of a credit assessment processing method described in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can be referred to the implementation of the corresponding method described above, and the repeated parts will not be described again.

[0144] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. Each embodiment focuses on describing the differences from other embodiments. For example, the device embodiments, equipment embodiments, computer-readable storage medium embodiments, and computer program product embodiments are all similar to the method embodiments, so the descriptions are relatively simple. For reading the relevant content of the device embodiments, equipment embodiments, computer-readable storage medium embodiments, and computer program product embodiments, please refer to the description of the method embodiments.

[0145] 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.

[0146] In the 1930s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many improvements to the methodology today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that an improvement to the methodology cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0147] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0148] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0149] For ease of description, the above apparatus is described by dividing it into various functional units. Of course, when implementing the embodiments of this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0150] Those skilled in the art will understand that one or more embodiments of this specification can be provided as a method, system, or computer program product. Therefore, one or more embodiments of this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0151] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0152] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0153] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0154] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0155] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0156] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0157] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising at least one…" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0158] One or more embodiments of this specification can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a particular task or implement a particular abstract data type. One or more embodiments of this specification can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0159] The above description is merely an embodiment of this document and is not intended to limit the scope of this document. Various modifications and variations can be made to this document by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this document should be included within the scope of the claims of this document.

Claims

1. A credit assessment processing method, characterized in that, The method includes: The image to be evaluated is input into the text extraction model and the text is extracted according to the text extraction strategy to obtain structured text; the image to be evaluated is uploaded by the user through the credit assessment interface provided by the application after the user logs in to the application. The user identity information contained in the structured text is compared with the login user information of the application. If they match, the similarity of the entity fields contained in the structured text with other historical entity fields uploaded by other users is verified. The historical entity fields are obtained by extracting text from historical evaluation images that have undergone credit assessment processing. If the text similarity calculated based on the entity field and the historical entity field is less than the similarity threshold, the similarity verification is deemed successful. The image to be evaluated and the prompt text are then input into the large language model for image enhancement verification to obtain the verification result. The image enhancement verification includes: extracting and fusing image features in the image structure dimension, and determining the verification result by calculating the verification score of the fused features. The prompt text is used to instruct the large language model to perform the image enhancement verification task and generate the verification result in a preset format. If the verification result is that the verification fails, the image to be evaluated is marked, and the marked image is sent to the reviewer for review.

2. The credit assessment processing method according to claim 1, characterized in that, The step of comparing the user identity information contained in the structured text with the login user information of the application includes: Query the login user information of the application that collected the image to be evaluated; The user identity information is compared with the logged-in user information. If they match, the identity verification is confirmed to be successful.

3. The credit assessment processing method according to claim 1, characterized in that, The similarity verification of entity fields contained in the structured text with historical entity fields uploaded by other users includes: Calculate the field similarity between each entity field in the structured text and the historical entity fields; Text similarity is calculated based on the field weights and field similarities of each entity field; If the text similarity is less than the similarity threshold, then the similarity verification is deemed successful.

4. The credit assessment processing method according to claim 1, characterized in that, Also includes: If the verification result is that the verification is successful, extract the entity fields used for credit assessment from the structured text; The credit limit variable is calculated based on the preset evaluation rules and the entity fields, and a credit evaluation result containing the credit limit variable is generated.

5. The credit assessment processing method according to claim 1, characterized in that, If the verification result is successful, perform the following operations: The structured text and the verification results are synchronized with the credit assessment platform for credit assessment processing. Accordingly, after the step of synchronizing the structured text and the verification result with the credit assessment platform for credit assessment processing is executed, the process further includes: The user's current credit limit is calculated based on the user's available credit limit and the credit assessment results returned by the credit assessment platform; The available credit limit is updated based on the current credit limit to obtain an updated credit limit.

6. The credit assessment processing method according to claim 5, characterized in that, After the step of updating the available credit limit based on the current credit limit to obtain the updated credit limit is executed, the method further includes: Generate a credit limit update notification that includes the updated credit limit; The credit limit update notification is sent to the user's terminal so that the credit limit update information can be displayed through the application installed on the user's terminal.

7. The credit assessment processing method according to claim 1, characterized in that, The text extraction is performed in the following manner: The image to be evaluated is subjected to text recognition to obtain the text recognition result; The text recognition result is extracted according to the text extraction strategy to obtain the structured text.

8. The credit assessment processing method according to claim 1, characterized in that, Before the step of inputting the image to be evaluated into the text extraction model to extract text according to the text extraction strategy and obtaining structured text, the following steps are also included: Obtain the image to be evaluated sent by the user terminal; The image to be evaluated is uploaded to the user terminal by the user through an application installed on the user terminal.

9. A credit assessment processing device, characterized in that, The device includes: The text extraction module is configured to input the image to be evaluated into the text extraction model and extract the text according to the text extraction strategy to obtain structured text; the image to be evaluated is uploaded by the user through the credit assessment interface provided by the application after the user logs in to the application. The verification module is configured to compare the user identity information contained in the structured text with the login user information of the application. If they match, the module performs a similarity verification on the entity fields contained in the structured text and other historical entity fields uploaded by other users. The historical entity fields are obtained by extracting text from historical evaluation images that have undergone credit assessment processing. If the text similarity calculated based on the entity field and the historical entity field is less than the similarity threshold, the similarity verification is deemed successful, and the verification processing module is run. The verification processing module is configured to input the image to be evaluated and the prompt text into a large language model for image enhancement verification to obtain the verification result. The image enhancement verification includes: extracting and fusing image features in the image structure dimension, and determining the verification result by calculating a verification score on the fused features. The prompt text is used to instruct the large language model to perform the image enhancement verification task and generate the verification result in a preset format. The credit assessment module is configured to mark the image to be assessed if the verification result is that the verification fails, and send the marked image to the reviewer for review processing.

10. A credit assessment processing device, characterized in that, The device includes: A processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to: The image to be evaluated is input into the text extraction model and the text is extracted according to the text extraction strategy to obtain structured text; the image to be evaluated is uploaded by the user through the credit assessment interface provided by the application after the user logs in to the application. The user identity information contained in the structured text is compared with the login user information of the application. If they match, the similarity of the entity fields contained in the structured text with other historical entity fields uploaded by other users is verified. The historical entity fields are obtained by extracting text from historical evaluation images that have undergone credit assessment processing. If the text similarity calculated based on the entity field and the historical entity field is less than the similarity threshold, the similarity verification is deemed successful. The image to be evaluated and the prompt text are then input into the large language model for image enhancement verification to obtain the verification result. The image enhancement verification includes: extracting and fusing image features in the image structure dimension, and determining the verification result by calculating the verification score of the fused features. The prompt text is used to instruct the large language model to perform the image enhancement verification task and generate the verification result in a preset format. If the verification result is that the verification fails, the image to be evaluated is marked, and the marked image is sent to the reviewer for review.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store computer-executable instructions that, when executed, implement the steps of the method of claim 1.

Images