Fund business data processing method and device, storage medium and program product

Through pre-trained classification model and parameter extraction model, the fund announcements are automatically identified and data extracted, which solves the problem of system response speed decreasing during fund announcement processing, and realizes an efficient and automated processing process, improving system performance and response speed.

CN118965094APending Publication Date: 2024-11-15CHINA MERCHANTS BANK
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410929645.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-11
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

During the process of fund announcement processing, due to the increase in the number of funds and the complexity of announcement types, the server needs to process more requests, resulting in a decrease in system response speed, high computer resource utilization, and increased memory consumption, which affects system performance.

Method used

Use pre-trained classification model and parameter extraction model to automatically identify and extract fund announcements, reduce manual intervention, ensure parameter accuracy through verification, and automate processing after verification, to reduce inefficient manual requests and systems load.

Benefits of technology

Through intelligent and automated processing processes, the overall load of the system during fund announcement processing is reduced, the demand for server resources is reduced, the response speed and processing efficiency are improved, and the system response speed is avoided.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118965094A_ABST
    Figure CN118965094A_ABST
Patent Text Reader

Abstract

The invention discloses a fund business data processing method and device, a storage medium and a program product, and relates to the technical field of computers, and the fund business data processing method comprises the steps of obtaining a to-be-announced file, and inputting the to-be-announced file into a preset classification model to obtain a target announcement type; inputting the target announcement type and the file to be announced into a preset parameter extraction model to obtain a target announcement parameter value; verifying the target announcement type and the target announcement parameter value; and processing target business data of a target fund corresponding to the to-be-announced file on the basis of the target announcement parameter value on the basis of the target announcement type under the condition that the verification is passed. Through the intelligent and automatic processing flow, the overall load of the system in the fund announcement processing process is reduced, the demand for server resources is reduced, and the problem that the response speed of the system in the fund announcement processing process is affected due to increase of the number of fund announcements can be avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a fund business data processing method, device, storage medium and program product. Background Art

[0002] During the fund announcement processing, the wealth processing operation personnel need to confirm the announcement type by querying the announcement-related content information during daily operations, and identify the attributes, transaction switches, rates and other business parameter information of the fund products that need to be maintained based on the announcement type, and issue various operations in the wealth processing operation system as soon as possible to perform different effective processing for different announcements.

[0003] With the increase in the number of funds, the diversification of announcement channels and the complexity of announcement types, the number of fund announcements has grown rapidly, resulting in the server having to process more requests, causing the utilization rate of computer resources to continue to run at a high level and memory consumption to increase, which in turn affects the response speed of the system.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a fund business data processing method, device, storage medium and program product, aiming to solve the problem that the system response speed is affected during the fund announcement processing.

[0006] To achieve the above purpose, the present application proposes a fund business data processing method, the method comprising:

[0007] Obtain a file to be announced, and input the file to be announced into a preset classification model to obtain a target announcement type, wherein the classification model is trained using the fund announcement file as input data and the fund announcement type as a training label;

[0008] Input the target announcement type and the to-be-announced file into a preset parameter extraction model to obtain a target announcement parameter value, wherein the parameter extraction model uses the fund announcement file and the fund announcement type as input data and is trained with the fund announcement parameter value as a training label;

[0009] Verifying the target announcement type and the target announcement parameter value;

[0010] When the verification is passed, the target business data of the target fund corresponding to the file to be announced is processed based on the target announcement type and the target announcement parameter value.

[0011] In one embodiment, the step of verifying the target announcement type and the target announcement parameter value includes:

[0012] Outputting the target announcement parameter value and the target announcement type to a visualization page, and receiving input operations from business personnel based on the visualization page;

[0013] If the input operation carries a value change request, determining that the verification fails;

[0014] If the input operation does not carry a value change request, it is determined that the verification is passed.

[0015] In one embodiment, the step of processing the target business data of the target fund corresponding to the to-be-announced file based on the target announcement type and the target announcement parameter value comprises:

[0016] Determining the type of operation and maintenance of the target fund based on the target announcement type;

[0017] The target business data is processed according to the operation and maintenance operation type and the target announcement parameter value.

[0018] In one embodiment, the parameter extraction model includes parameter extraction sub-models corresponding to different announcement types;

[0019] The step of inputting the target announcement type and the to-be-announced file into a preset parameter extraction model to obtain a target announcement parameter value comprises:

[0020] Determine a target parameter extraction submodel corresponding to the target announcement type from the parameter extraction submodels corresponding to the different announcement types;

[0021] The target announcement type and the file to be announced are input into the target parameter extraction sub-model to obtain the target announcement parameter value.

[0022] In one embodiment, after the step of verifying the target announcement type and the target announcement parameter value, the method further comprises:

[0023] Get the verification result;

[0024] updating the classification model based on the verification result, the target announcement type and the to-be-announced file, and using the updated classification model as a preset classification model;

[0025] The parameter extraction model is updated based on the verification result, the target announcement parameter, the target announcement type and the file to be announced, and the updated parameter extraction model is used as the preset parameter extraction model.

[0026] In one embodiment, if the verification result is that the verification is passed, the step of updating the classification model based on the verification result, the target announcement type and the to-be-announced file and using the updated classification model as the preset classification model includes:

[0027] The target announcement type is used as a training label, and the to-be-announced file is used as input data to train the classification model, and the trained classification model is used as a preset classification model;

[0028] The step of updating the parameter extraction model based on the verification result, the target announcement parameter, the target announcement type and the to-be-announced file, and using the updated parameter extraction model as the preset parameter extraction model, comprises:

[0029] The target announcement type and the to-be-announced file are used as input parameters, and the target announcement parameter value is used as a training label to train the parameter extraction model, and the trained parameter extraction model is used as a preset parameter extraction model.

[0030] In one embodiment, if the verification result is that the verification fails, the step of updating the classification model based on the verification result, the target announcement type and the to-be-announced file and using the updated classification model as the preset classification model includes:

[0031] Acquire feedback information from business personnel, and use the feedback information from business personnel to update the target announcement parameter value and / or the target announcement type to obtain an updated parameter value and an updated announcement type;

[0032] The classification model is trained using the updated announcement type as a training label and the to-be-announced file as input data, and the trained classification model is used as a preset classification model;

[0033] The step of updating the parameter extraction model based on the verification result, the target announcement parameter, the target announcement type and the to-be-announced file, and using the updated parameter extraction model as the preset parameter extraction model, comprises:

[0034] The updated announcement type and the to-be-announced file are used as input parameters, and the updated parameter value is used as a training label to train the parameter extraction model, and the trained parameter extraction model is used as a preset parameter extraction model.

[0035] In addition, to achieve the above-mentioned purpose, the present application also proposes a fund business data processing device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the fund business data processing method as described above.

[0036] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the fund business data processing method described above are implemented.

[0037] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the fund business data processing method described above.

[0038] In the present application, a file to be announced is obtained, and the file to be announced is input into a preset classification model to obtain a target announcement type, wherein the classification model is trained with the fund announcement file as input data and the fund announcement type as a training label; the target announcement type and the file to be announced are input into a preset parameter extraction model to obtain a target announcement parameter value, wherein the parameter extraction model is trained with the fund announcement file and the fund announcement type as input data and the fund announcement parameter value as a training label; the target announcement type and the target announcement parameter value are verified; and if the verification passes, the target business data of the target fund corresponding to the file to be announced is processed based on the target announcement type and the target announcement parameter value.

[0039] In this application, data of fund announcements is extracted through pre-trained classification models and parameter extraction models. The system can automatically identify the type of announcement file and extract relevant business parameters without human intervention, thereby reducing dependence on human resources and reducing the time for operators to manually query and classify announcements, thereby reducing inefficient manual requests that the server needs to process and reducing the overall load on the system.

[0040] By verifying the parameters, the accuracy of all parameters before application is ensured, avoiding the need to re-extract the fund announcement and then adjust the parameters due to configuration errors, reducing the overall load of the system and further optimizing the use of system resources.

[0041] After the parameter verification is passed, the system can immediately process the parameter value based on the target announcement type and take corresponding measures for specific funds. The entire process is highly automated, reducing processing time and improving response speed. Compared with the previous manual processing process, this mechanism greatly improves the efficiency of processing announcements and reduces the pressure on the server when processing a large number of concurrent requests.

[0042] Therefore, this application reduces the overall system load during the fund announcement processing process through an intelligent and automated processing flow, reduces the demand for server resources, and thus avoids the problem of the system response speed being affected during the fund announcement processing due to the increase in the number of fund announcements. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0045] Figure 1 A flowchart of the first embodiment of the method for processing fund business data in this application;

[0046] Figure 2 A flowchart of the second embodiment of the method for processing fund business data in this application;

[0047] Figure 3 A schematic diagram of the modules of a fund announcement processing system provided in one embodiment of the present application;

[0048] Figure 4 A timing diagram of the fund business data processing process provided by one embodiment of the present application;

[0049] Figure 5 A schematic diagram of a display announcement page provided for an implementation method of the present application;

[0050] Figure 6 This is a schematic diagram of the module structure of the fund business data processing device according to an embodiment of the present application;

[0051] Figure 7 This is a schematic diagram of the device structure of the hardware operating environment involved in the fund business data processing method in the embodiment of the present application.

[0052] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0053] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0054] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0055] The main solution of the embodiment of the present application is: obtain the file to be announced, input the file to be announced into a preset classification model to obtain a target announcement type, wherein the classification model is trained with the fund announcement file as input data and the fund announcement type as a training label; input the target announcement type and the file to be announced into a preset parameter extraction model to obtain a target announcement parameter value, wherein the parameter extraction model is trained with the fund announcement file and the fund announcement type as input data and the fund announcement parameter value as a training label; verify the target announcement type and the target announcement parameter value; if the verification passes, process the target business data of the target fund corresponding to the file to be announced based on the target announcement type and the target announcement parameter value.

[0056] In this embodiment, for ease of description, the following is an explanation using the fund business data processing device as the execution subject. In the fund announcement processing process, the wealth processing operator needs to confirm the announcement type by querying the announcement-related content information during daily operations, and identify the attributes, transaction switches, rates and other business parameter information of the fund products that need to be maintained according to the announcement type, and issue various operations in the wealth processing operation system as soon as possible to perform different effective processing on different announcements.

[0057] With the increase in the number of funds, the diversification of announcement channels and the complexity of announcement types, the number of fund announcements has grown rapidly, resulting in the server having to process more requests, causing the utilization rate of computer resources to continue to run at a high level and memory consumption to increase, which in turn affects the response speed of the system.

[0058] In the embodiment of the present application, data of fund announcements is extracted through pre-trained classification models and parameter extraction models. The system can automatically identify the type of announcement file and extract relevant business parameters without human intervention, thereby reducing dependence on human resources and reducing the time for operators to manually query and classify announcements, thereby reducing inefficient manual requests that the server needs to process and reducing the overall load on the system.

[0059] By verifying the parameters, the accuracy of all parameters before application is ensured, avoiding the need to re-extract the fund announcement and then adjust the parameters due to configuration errors, reducing the overall load of the system and further optimizing the use of system resources.

[0060] After the parameter verification is passed, the system can immediately process the parameter value based on the target announcement type and take corresponding measures for specific funds. The entire process is highly automated, reducing processing time and improving response speed. Compared with the previous manual processing process, this mechanism greatly improves the efficiency of processing announcements and reduces the pressure on the server when processing a large number of concurrent requests.

[0061] Therefore, the embodiment of the present application reduces the overall load of the system during the fund announcement processing through an intelligent and automated processing flow, reduces the demand for server resources, and thus avoids the problem of the system response speed being affected during the fund announcement processing due to the increase in the number of fund announcements.

[0062] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a fund business data processing device, etc. The following takes the fund business data processing device as an example to illustrate this embodiment and the following embodiments.

[0063] Based on this, the present application embodiment provides a fund business data processing method, referring to Figure 1 This is a flow chart of the first embodiment of the fund business data processing method of the present application.

[0064] In this embodiment, the fund business data processing method includes steps S10 to S40.

[0065] Step S10, obtaining the file to be announced, and inputting the file to be announced into a preset classification model to obtain a target announcement type, wherein the classification model is trained using the fund announcement file as input data and the fund announcement type as a training label.

[0066] It should be noted that the pending announcement files are fund announcements that require confirmation of the announcement type, and identify the business parameter information such as the attributes, transaction switches, and rates of fund products that need to be maintained based on the announcement type, so that various operations can be issued as soon as possible in the wealth processing operation system based on the parameter information, and different announcements can be processed differently. The pending announcement files to be released are collected from the fund announcement processing system, and the pending announcement files may contain various information changes about the fund, such as rate adjustments, product attribute changes, etc.

[0067] The classification model is obtained by taking the fund announcement files of known announcement types as input data and the fund announcement types as training labels. It should be noted that the structure of the classification model is not limited here, for example, it can be a large language model or a keyword recognition model.

[0068] The document to be announced is input into the preset classification model to obtain the target announcement type. The target announcement type represents the business type of the fund corresponding to the document to be announced (hereinafter referred to as the target fund for convenience of description), including but not limited to suspension of trading, ordinary opening or regular opening, limit adjustment, liquidation, fund contract effectiveness, rate adjustment, etc.

[0069] In this implementation, by using a model to automatically classify documents to be announced, the need for manual classification is reduced, thereby improving efficiency and accuracy.

[0070] Step S20, input the target announcement type and the file to be announced into a preset parameter extraction model to obtain the target announcement parameter value, wherein the parameter extraction model is trained with the fund announcement file and the fund announcement type as input data and the fund announcement parameter value as a training label.

[0071] The parameter extraction model is obtained by training with the fund announcement file and fund announcement type as input data and the fund announcement parameter value as training label. It should be noted that the structure of the parameter extraction model is not limited here, for example, it can be a large language model or a keyword recognition model.

[0072] The target announcement type and the file to be announced are sent to the parameter extraction model to identify and extract the specific parameter values ​​related to fund attributes, trading rules, fees, etc. in the announcement, that is, the target announcement parameter values, which represent the business parameters of the target fund. It should be noted that the business parameters to be extracted for different target announcement types may be different. For example, the business parameters to be extracted for the suspended trading type are the start and end dates of suspension [or resumption] of subscription [or redemption], the start and end dates of suspension [or resumption] of transfer-in [or transfer-out], and the start and end dates of suspension [or resumption] of regular fixed-amount investment, etc.; the business parameters to be extracted for ordinary opening or regular opening are the start and end dates of subscription [or redemption], the subscription [or redemption] opening period, the start and end dates of conversion transfer-in [or transfer-out], and the start and end dates of regular fixed-amount investment, etc. The business parameters corresponding to different target announcement types are not limited here and can be set according to actual needs.

[0073] Step S30, verifying the target announcement type and the target announcement parameter value.

[0074] Verify the target announcement type and target announcement parameter value to detect whether the target announcement parameter value is accurate and meets business requirements, thereby ensuring data accuracy and compliance and preventing business problems caused by incorrect parameter values. It should be noted that there is no restriction on the verification method here, and it can be set according to actual needs. For example, it can be a comparison of existing parameter standards in the database, or a logical check of the target announcement parameter value, such as the rate cannot be negative, or a business rule verification to ensure that the extracted parameter value meets the logic and business requirements.

[0075] In one feasible implementation, the process of verifying the target announcement type and the target announcement parameter value may be:

[0076] a1, output the target announcement parameter value and the target announcement type to the visualization page, and receive the input operation of the business personnel based on the visualization page.

[0077] The target announcement parameter values ​​obtained from the parameter extraction model are displayed on a visualization page so that business personnel can intuitively view the specific values ​​of each parameter. Business personnel can interact on this page, such as viewing, confirming, or proposing changes to parameter values ​​that are considered incorrect.

[0078] By providing a visual interface for business personnel to operate and integrating the flexibility of automated processing and manual review, the rationality of parameter adjustments is ensured.

[0079] a2: If the input operation carries a value change request, it is determined that the verification fails.

[0080] If the input operation of the business personnel on the visualization page includes a request to change the parameter value, this means that the currently extracted parameter value is inaccurate or does not meet expectations. In this case, the verification results of these parameter values ​​are marked as failed.

[0081] a3: If the input operation does not carry a value change request, it is determined that the verification has passed.

[0082] If the business personnel's input operation does not request a change in the value, that is, if the parameter value extracted by the system is confirmed to be correct, the verification of the target announcement parameter value will be deemed to have passed, and the target parameter value will be deemed to be accurate and meet business requirements and can be used for the next fund business data processing operation.

[0083] In this implementation, business personnel can directly participate in the parameter verification process through the visualization page, which increases the transparency and interactivity of the operation and is conducive to improving the accuracy and trust of data processing. In addition, the combination of automated extraction and manual review not only brings into play the efficiency of machine processing, but also incorporates the flexibility and accuracy of human judgment, achieving complementary advantages.

[0084] Step S40, when the verification passes, the target business data of the target fund corresponding to the announcement file is processed based on the target announcement type and the target announcement parameter value.

[0085] It should be noted that the target business data is the business data of the target fund corresponding to the to-be-announced document. When the target announcement parameter value passes the verification, the corresponding target business data will be processed according to the target announcement type and target announcement parameter value, such as adjusting the fund's transaction status, modifying the fee rate, etc., to ensure the timely and correct update of fund information and business rules, and improve operational efficiency and customer experience.

[0086] In this embodiment, data is extracted from fund announcements through pre-trained classification models and parameter extraction models. The system can automatically identify the type of announcement file and extract relevant business parameters without human intervention, thereby reducing dependence on human resources and reducing the time for operators to manually query and classify announcements, thereby reducing inefficient manual requests that the server needs to process and reducing the overall load on the system.

[0087] By verifying the parameters, the accuracy of all parameters before application is ensured, avoiding the need to re-extract the fund announcement and then adjust the parameters due to configuration errors, reducing the overall load of the system and further optimizing the use of system resources.

[0088] After the parameter verification is passed, the system can immediately process the parameter value based on the target announcement type and take corresponding measures for specific funds. The entire process is highly automated, reducing processing time and improving response speed. Compared with the previous manual processing process, this mechanism greatly improves the efficiency of processing announcements and reduces the pressure on the server when processing a large number of concurrent requests.

[0089] Therefore, this embodiment reduces the overall load of the system during the fund announcement processing through an intelligent and automated processing flow, reduces the demand for server resources, and thus avoids the problem of the system response speed being affected during the fund announcement processing due to the increase in the number of fund announcements.

[0090] In one embodiment, step S40: processing the target business data of the target fund corresponding to the announcement document based on the target announcement parameter value based on the target announcement type, includes steps S401 to S402.

[0091] Step S401, determining the type of operation and maintenance for the target fund based on the target announcement type.

[0092] Based on the identified target announcement type, the specific operation type that needs to be performed on the target fund is further determined. The operation type represents the type of business operation on the target fund. In a specific implementation method, corresponding processing rules or logic can be preset for different operation types to ensure the pertinence and effectiveness of subsequent processing steps. For example, if the announcement type is about rate adjustment, then the operation type may be "rate update"; if it is a change in product features, then the operation may be "product attribute modification."

[0093] Step S402: Process the target business data according to the operation type and target announcement parameter value.

[0094] Specific processing is performed on the target business data according to the operation type and the target announcement parameter value. For example, the rate information can be updated in the fund business data processing system, a certain transaction function can be turned on or off, and the investment strategy restrictions of the fund can be adjusted. It should be noted that all operations are strictly performed in accordance with the requirements of the operation type and parameter value to ensure that the fund status is consistent with the announcement content.

[0095] In this implementation, by clearly distinguishing the types of announcements and determining specific operation and maintenance operations accordingly, accurate processing of the fund is ensured, and the risks caused by operational errors or over-generalization are avoided. In addition, automated operation and maintenance operations reduce the need for manual intervention, significantly improve processing speed and efficiency, and reduce human errors, making fund business data processing more standardized and efficient.

[0096] In one embodiment, the parameter extraction model includes parameter extraction sub-models corresponding to different announcement types. Step S20: input the target announcement type and the to-be-announced file into the preset parameter extraction model to obtain the target announcement parameter value, including steps S201 to S202:

[0097] Step S201, determining a target parameter extraction sub-model corresponding to a target announcement type from parameter extraction sub-models corresponding to different announcement types.

[0098] In this embodiment, the parameter extraction model is designed to include multiple sub-models, each of which can be trained specifically for a specific type of announcement (such as rate adjustments, product changes, dividend notifications, etc.) to more accurately identify and extract key parameters in such announcements. By matching the target announcement type with the corresponding sub-model, the extraction process is ensured to be highly targeted.

[0099] When extracting the target announcement parameters, the target parameter extraction sub-model corresponding to the target announcement type is determined from the parameter extraction sub-models corresponding to different announcement types.

[0100] Step S202, input the target announcement type and the file to be announced into the target parameter extraction sub-model to obtain the target announcement parameter value.

[0101] The target announcement type and the file to be announced are fed into the target parameter extraction sub-model as input. The sub-model uses the patterns and features learned during the training process to accurately extract all important parameter values ​​related to the announcement type from the announcement file, such as rates, effective dates, fund codes, etc. This process is highly automated and aims to capture all key information quickly and accurately.

[0102] In this implementation, by constructing sub-models for different announcement types, the accuracy and efficiency of parameter extraction are improved. Each sub-model focuses on processing the complexity and particularity of a specific announcement type, reducing the risk of cross-type misidentification. In addition, specialized sub-models can more deeply understand and parse specific announcement formats and contents, improve the accuracy of parameter identification, and reduce the need for manual review.

[0103] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can refer to the above introduction, and will not be repeated later. Figure 2 Step S30: After verifying the target announcement type and the target announcement parameter value, the method further includes steps S50 to S70:

[0104] Step S50, obtaining the verification result.

[0105] After verifying the target announcement type and target announcement parameter value, the verification result of the verification operation feedback is obtained. The verification result is the result that characterizes whether the target announcement parameter value and target announcement type need to be adjusted. It is the basis for subsequent model updates and fund business adjustments, and ensures the real-time and accuracy of model training data.

[0106] Step S60, updating the classification model based on the verification result, the target announcement type and the file to be announced, and using the updated classification model as the preset classification model.

[0107] The model will be fine-tuned or retrained based on the verification results to ensure that similar announcements can be classified more accurately in the future. The updated model will replace the original preset classification model to continuously improve the ability to recognize announcement types.

[0108] It should be noted that the verification result may be that the target announcement parameters and the target announcement type pass, or it may be that the verification fails (the verification failure may be that the target announcement parameters fail, or it may be that the target announcement type fails). In a specific implementation, when the verification result is that the target announcement type and the target announcement parameter value pass, the classification model is updated based on the target announcement type and the file to be announced; or when the verification result is that the target announcement parameter fails, the target announcement parameters are adjusted, and the classification model is updated using the adjusted target announcement parameters, target announcement type and the file to be announced; or the model update operation may be performed in both cases.

[0109] Step S70, updating the parameter extraction model based on the verification result, target announcement parameters, target announcement type and files to be announced, and using the updated parameter extraction model as the preset parameter extraction model.

[0110] Based on the verification results, target announcement parameters, announcement type and specific content of the original announcement file, the parameter extraction model is optimized in a targeted manner.

[0111] It should be noted that, when the verification result is that the target announcement type and the target announcement parameter value have passed, the parameter extraction model may be updated based on the target announcement type and the file to be announced; or, when the verification result is that the target announcement parameter verification has failed, after adjusting the target announcement parameters, the parameter extraction model may be updated using the adjusted target announcement parameters, target announcement type and the file to be announced; or, the model update operation may be performed in both cases.

[0112] In this embodiment, by feeding back the verification results to the model update process, the model can achieve self-learning and evolution, so that the model can better adapt to changes in the content of the announcement, gradually improve the accuracy of identifying the announcement type and parameter values, and improve the model's ability to handle new types of announcements or complex situations. In addition, by promptly correcting problems exposed by the model in actual applications, the error rate in future announcement processing can be reduced, and the overall efficiency and reliability of announcement processing can be improved.

[0113] In one embodiment, if the verification result is passed, step S60: based on the verification result, the target announcement type and the file to be announced, the classification model is updated, and the updated classification model is used as the preset classification model, including step S601.

[0114] Step S601, training a classification model with the target announcement type as a training label and the file to be announced as input data, and using the trained classification model as a preset classification model.

[0115] When the verification results show that the target announcement type and target announcement parameter value have passed the verification, the target announcement type is used as the training label and the pending announcement file is used as the input data to fine-tune or supplement the existing classification model to ensure that the model can better understand and remember the classification instance and enhance its accuracy in judging similar announcement types in the future. The updated model becomes the new preset classification model and is ready to be applied to subsequent announcement classification.

[0116] In this implementation, step S70: updating the parameter extraction model based on the verification result, target announcement parameter, target announcement type and file to be announced, and using the updated parameter extraction model as the preset parameter extraction model, includes step S701.

[0117] Step S701, taking the target announcement type and the file to be announced as input parameters, and taking the target announcement parameter value as a training label to train the parameter extraction model, and using the trained parameter extraction model as the preset parameter extraction model.

[0118] If the target announcement type and target announcement parameter value are verified, it means that the parameter extraction model has successfully identified and extracted the correct parameter value. The target announcement type and the file to be announced are used as input parameters, and the extracted correct target announcement parameter value is used as the training label to train the parameter extraction model to enhance the model's ability to extract parameters under specific announcement types and file contents, thereby improving the accuracy and stability of parameter extraction. The updated model is also set as the new preset parameter extraction model for future parameter extraction tasks.

[0119] In this implementation, by continuously and dynamically adjusting the classification model and parameter extraction model, the model can better cope with edge cases and complex situations, enhance the stability and reliability of the system, and improve the automation level and efficiency of the overall announcement processing.

[0120] In one embodiment, if the verification result is verification failure, step S60: based on the verification result, the target announcement type and the file to be announced, the classification model is updated, and the updated classification model is used as the preset classification model, including steps S602 to S603.

[0121] Step S602, obtaining feedback information from business personnel, and using the feedback information from business personnel to update target announcement parameter values ​​and / or target announcement types to obtain updated parameter values ​​and updated announcement types.

[0122] If the verification result is verification failure, it means that the target announcement type or the target announcement parameter value is inaccurate. Obtain feedback from business personnel on the target announcement parameter value and / or target announcement type. Update the original target announcement type or target announcement parameter value based on the feedback received to ensure that the target announcement type or target announcement parameter value is consistent with the actual file to be announced.

[0123] Step S603, using the updated announcement type as a training label and the file to be announced as input data to train the classification model, and using the trained classification model as a preset classification model.

[0124] The updated announcement type is used as a new training label, and the file to be announced is used as input data to train the classification model again, so that the model can learn the professional judgment of business personnel and improve its classification accuracy when handling similar announcements in the future. After the training is completed, the updated model is set as the new default model.

[0125] In this implementation, step S70: updating the parameter extraction model based on the verification result, target announcement parameter, target announcement type and file to be announced, and using the updated parameter extraction model as the preset parameter extraction model, including step S702.

[0126] Step S702, taking the updated announcement type and the file to be announced as input parameters, taking the updated parameter value as a training label to train the parameter extraction model, and taking the trained parameter extraction model as the preset parameter extraction model.

[0127] Get the updated announcement type and parameter value, use the updated announcement type and the file to be announced as input parameters, and use the updated parameter value as the training label to train the parameter extraction model to ensure that the model can accurately learn and extract the parameter value adjusted according to the opinions of business experts, further improving the practicality and accuracy of the model. The updated parameter extraction model is also used as the preset parameter extraction model for subsequent processing.

[0128] In this implementation, by dynamically updating the model, it is ensured that the model can quickly absorb new knowledge and adjustments, thereby improving the accuracy of judgments on announcement types and parameter values.

[0129] For example, in order to help understand the technical concept or technical principle of fund business data processing after combining this embodiment with the above-mentioned embodiment 1 and embodiment 2, this embodiment provides a fund announcement processing method and system in the fund announcement processing process, which can better display the announcement content in the business system, and perform classification and keyword preprocessing, so as to facilitate efficient and fast processing and open up the last link of fund announcement application. The system includes an announcement aggregation platform, an intelligent analysis engine, and an announcement processing module. The module schematic diagram is shown in FIG. Figure 3As shown, the announcement aggregation platform is used to centrally collect announcements and notify the intelligent analysis of changes in announcement information; the engine intelligent analysis engine, a customized artificial intelligence processing engine, is used to filter announcements, determine the announcement type based on the announcement content, and extract the corresponding business parameters based on the announcement type, and provide them to the announcement processing module for use; the announcement processing module receives the announcement content, announcement type, and extracted business parameters sent by the intelligent analysis engine, and displays them to the user to facilitate the user to confirm the announcement, and issues corresponding execution instructions based on the announcement type and extracted business parameters.

[0130] In this embodiment, refer to Figure 4 , the implementation process of fund business data processing can be:

[0131] 1. (Timed trigger) Pay attention to the synchronization of fund list, and send all the funds (such as fund code, fund name) processed by the business system to the intelligent analysis engine.

[0132] 2. Announcement collection: collect, unify and store announcements issued by fund companies or securities firms, including but not limited to announcement ID (Identifier), fund code, announcement name or title, announcement content, etc. It is recommended that this can be achieved through an announcement collection platform or crawler.

[0133] 3. Notify the announcement change information and notify the intelligent analysis engine of the announcement information collected in step 2. For reference, after step 2 is stored in the database, the database change triggered sends a synchronization message (that is, obtains the file to be announced) through the database processing middleware such as canal and Kafka.

[0134] 4. Filter the fund announcements you want to follow. Use the fund list provided in step 1 to filter out the synchronization messages of relevant announcements for subsequent analysis.

[0135] 5. Determine the announcement classification based on the announcement content. According to step 4, filter and receive the announcement name (title) and announcement content to determine the announcement classification. Specifically, a large language model can be used, and the model can be improved based on the feedback provided in step 12 (that is, the announcement file is input into the preset classification model to obtain the target announcement type).

[0136] Exemplarily, some announcement types and their corresponding announcement keywords may be shown in Table 1 below. It should be noted that the announcement types and keywords shown in Table 1 below are only examples, and the actual announcement types and keywords may be set according to actual needs.

[0137] Table 1 Some announcement types and their corresponding announcement keywords

[0138]

[0139] 6. Determine the business parameters to be extracted based on the announcement content and announcement classification. Step 5 determines the target announcement type. Different announcement types determine the corresponding business parameters to be extracted in this step to obtain the target announcement parameter value (that is, the target announcement type and the file to be announced are input into the preset parameter extraction model to obtain the target announcement parameter value). Exemplarily, the business parameters corresponding to some announcement types can be shown in Table 2 below. It should be noted that the announcement types and business parameters shown in Table 2 below are only examples. The actual announcement types and business parameters can be set according to actual needs:

[0140] Table 2 Some announcement types and their corresponding business parameters

[0141]

[0142]

[0143] 7. Push announcement content, announcement type, and extract parameters. Send the announcement content received in step 4, the announcement category determined in step 5, and the business parameters extracted in step 6 to the announcement processing module.

[0144] 8. Display announcement related information (announcement content, announcement type, extraction parameters), please refer to Figure 5 , you can refer to the announcement page Figure 5 Based on this display page, business personnel can use the Modify button to adjust the announcement type or business parameter extraction value and provide feedback to the intelligent analysis engine; they can also use the Delete button to delete abnormal announcements; use the Confirm button to confirm the announcement analysis results and issue execution instructions.

[0145] 9. Adjust the announcement type or extraction parameters. If necessary, the business personnel can modify the announcement type or extract parameters through the announcement display page (that is, verify the target announcement type and target announcement parameter value).

[0146] 10. Feedback adjustment content: the announcement processing module will feed back the business personnel's adjustment content to the intelligent analysis engine.

[0147] 11. Send new announcement types and extraction parameters. The intelligent analysis engine will re-extract business parameters based on the announcement type adjusted by the business personnel, and re-push the announcement content, announcement type, and extraction parameters.

[0148] 12. Optimize the analysis model based on the adjustment content of the feedback. The intelligent analysis engine optimizes the model and improves the analysis accuracy according to the classification or extraction parameter adjustments made by the business personnel (that is, obtain the verification results; update the classification model based on the verification results, target announcement type and files to be announced, and use the updated classification model as the preset classification model; update the parameter extraction model based on the verification results, target announcement parameters, target announcement type and files to be announced, and use the updated parameter extraction model as the preset parameter extraction model).

[0149] 13. Confirm the announcement. The business personnel confirm that the announcement classification and extraction parameters are correct and confirm that the announcement can take effect.

[0150] 14. Action instructions are issued according to the confirmed announcement type and business parameters. The announcement processing module uses command mode or strategy mode to perform different effectiveness processing on announcements of different categories, such as transaction control, limit control, etc. (that is, if the verification is passed, the target business data of the target fund corresponding to the announcement file is processed based on the target announcement type and the target announcement parameter value).

[0151] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the fund business data processing method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0152] This application also provides a fund business data processing device, please refer to Figure 6 , the fund business data processing device comprises:

[0153] The first input module 10 is used to obtain a file to be announced, and input the file to be announced into a preset classification model to obtain a target announcement type, wherein the classification model is trained using the fund announcement file as input data and the fund announcement type as a training label;

[0154] The second input module 20 is used to input the target announcement type and the to-be-announced file into a preset parameter extraction model to obtain a target announcement parameter value, wherein the parameter extraction model uses the fund announcement file and the fund announcement type as input data and is trained with the fund announcement parameter value as a training label;

[0155] A verification module 30, used to verify the target announcement type and the target announcement parameter value;

[0156] The processing module 40 is used to process the target business data of the target fund corresponding to the to-be-announced file based on the target announcement type and the target announcement parameter value when the verification passes.

[0157] Optionally, the verification module 30 is further used for:

[0158] Outputting the target announcement parameter value and the target announcement type to a visualization page, and receiving input operations from business personnel based on the visualization page;

[0159] If the input operation carries a value change request, determining that the verification fails;

[0160] If the input operation does not carry a value change request, it is determined that the verification is passed.

[0161] Optionally, the processing module 40 is further configured to:

[0162] Determining the type of operation and maintenance of the target fund based on the target announcement type;

[0163] The target business data is processed according to the operation and maintenance operation type and the target announcement parameter value.

[0164] Optionally, the parameter extraction model includes parameter extraction sub-models corresponding to different announcement types; the second input module 20 is further used for:

[0165] Determine a target parameter extraction submodel corresponding to the target announcement type from the parameter extraction submodels corresponding to the different announcement types;

[0166] The target announcement type and the file to be announced are input into the target parameter extraction sub-model to obtain the target announcement parameter value.

[0167] Optionally, the fund business data processing device further includes a model updating module, which is used to:

[0168] Get the verification result;

[0169] updating the classification model based on the verification result, the target announcement type and the to-be-announced file, and using the updated classification model as a preset classification model;

[0170] The parameter extraction model is updated based on the verification result, the target announcement parameter, the target announcement type and the file to be announced, and the updated parameter extraction model is used as the preset parameter extraction model.

[0171] Optionally, if the verification result is that the verification passes, the model updating module is further used to:

[0172] The target announcement type is used as a training label, and the to-be-announced file is used as input data to train the classification model, and the trained classification model is used as a preset classification model;

[0173] The step of updating the parameter extraction model based on the verification result, the target announcement parameter, the target announcement type and the to-be-announced file, and using the updated parameter extraction model as the preset parameter extraction model, comprises:

[0174] The target announcement type and the to-be-announced file are used as input parameters, and the target announcement parameter value is used as a training label to train the parameter extraction model, and the trained parameter extraction model is used as a preset parameter extraction model.

[0175] Optionally, if the verification result is that the verification fails, the model updating module is further used to:

[0176] Acquire feedback information from business personnel, and use the feedback information from business personnel to update the target announcement parameter value and / or the target announcement type to obtain an updated parameter value and an updated announcement type;

[0177] The classification model is trained using the updated announcement type as a training label and the to-be-announced file as input data, and the trained classification model is used as a preset classification model;

[0178] The step of updating the parameter extraction model based on the verification result, the target announcement parameter, the target announcement type and the to-be-announced file, and using the updated parameter extraction model as the preset parameter extraction model, comprises:

[0179] The updated announcement type and the to-be-announced file are used as input parameters, and the updated parameter value is used as a training label to train the parameter extraction model, and the trained parameter extraction model is used as a preset parameter extraction model.

[0180] The fund business data processing device provided by the present application adopts the fund business data processing method in the above embodiment, which can solve the technical problem that the system response speed is affected during the fund announcement processing. Compared with the prior art, the beneficial effects of the fund business data processing device provided by the present application are the same as the beneficial effects of the fund business data processing method provided by the above embodiment, and other technical features in the fund business data processing device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0181] The present application provides a fund business data processing device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the fund business data processing method in the above-mentioned embodiment one.

[0182] Reference below Figure 7 , which shows a schematic diagram of the structure of a fund business data processing device suitable for implementing the embodiment of the present application. The fund business data processing device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The fund business data processing device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0183] like Figure 7 As shown, the fund business data processing device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the fund business data processing device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the fund business data processing device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a fund business data processing device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or provided instead.

[0184] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0185] The fund business data processing device provided by the present application adopts the fund business data processing method in the above embodiment, which can solve the technical problem that the system response speed is affected during the fund announcement processing. Compared with the prior art, the beneficial effects of the fund business data processing device provided by the present application are the same as the beneficial effects of the fund business data processing method provided by the above embodiment, and other technical features in the fund business data processing device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0186] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0187] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0188] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the fund business data processing method in the above-mentioned embodiment.

[0189] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0190] The above-mentioned computer-readable storage medium may be included in the fund business data processing device; or it may exist independently without being assembled into the fund business data processing device.

[0191] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the fund business data processing device, the fund business data processing device: obtains the file to be announced, inputs the file to be announced into a preset classification model to obtain a target announcement type, wherein the classification model is trained with the fund announcement file as input data and the fund announcement type as a training label; inputs the target announcement type and the file to be announced into a preset parameter extraction model to obtain a target announcement parameter value, wherein the parameter extraction model is trained with the fund announcement file and the fund announcement type as input data and the fund announcement parameter value as a training label; verifies the target announcement type and the target announcement parameter value; and if the verification passes, processes the target business data of the target fund corresponding to the file to be announced based on the target announcement type and the target announcement parameter value.

[0192] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0193] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0194] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0195] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned fund business data processing method, and can solve the technical problem that the system response speed is affected during the fund announcement processing. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the fund business data processing method provided by the above-mentioned embodiment, and will not be repeated here.

[0196] The present application also provides a computer program product, including a computer program, which implements the steps of the fund business data processing method as described above when executed by a processor.

[0197] The computer program product provided by this application can solve the technical problem that the system response speed is affected during the fund announcement processing. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the fund business data processing method provided by the above embodiment, which will not be repeated here.

[0198] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A fund business data processing method, characterized in that: The method includes: Obtain a file to be announced, and input the file to be announced into a preset classification model to obtain a target announcement type, wherein the classification model is trained using the fund announcement file as input data and the fund announcement type as a training label; Input the target announcement type and the to-be-announced file into a preset parameter extraction model to obtain a target announcement parameter value, wherein the parameter extraction model uses the fund announcement file and the fund announcement type as input data and is trained with the fund announcement parameter value as a training label; Verifying the target announcement type and the target announcement parameter value; When the verification is passed, the target business data of the target fund corresponding to the file to be announced is processed based on the target announcement type and the target announcement parameter value.

2. The method according to claim 1, characterized in that The step of verifying the target announcement type and the target announcement parameter value comprises: Outputting the target announcement parameter value and the target announcement type to a visualization page, and receiving input operations from business personnel based on the visualization page; If the input operation carries a value change request, determining that the verification fails; If the input operation does not carry a value change request, it is determined that the verification is passed.

3. The method according to claim 1, characterized in that The step of processing the target business data of the target fund corresponding to the to-be-announced file based on the target announcement type and the target announcement parameter value comprises: Determining the type of operation and maintenance of the target fund based on the target announcement type; The target business data is processed according to the operation and maintenance operation type and the target announcement parameter value.

4. The method according to claim 1, characterized in that The parameter extraction model includes parameter extraction sub-models corresponding to different announcement types; The step of inputting the target announcement type and the to-be-announced file into a preset parameter extraction model to obtain a target announcement parameter value comprises: Determine a target parameter extraction submodel corresponding to the target announcement type from the parameter extraction submodels corresponding to the different announcement types; The target announcement type and the file to be announced are input into the target parameter extraction sub-model to obtain the target announcement parameter value.

5. The method according to any one of claims 1 to 4, characterized in that After the step of verifying the target announcement type and the target announcement parameter value, the method further comprises: Get the verification result; updating the classification model based on the verification result, the target announcement type and the to-be-announced file, and using the updated classification model as a preset classification model; The parameter extraction model is updated based on the verification result, the target announcement parameter, the target announcement type and the file to be announced, and the updated parameter extraction model is used as the preset parameter extraction model.

6. The method according to claim 5, characterized in that If the verification result is that the verification is passed, the step of updating the classification model based on the verification result, the target announcement type and the to-be-announced file, and using the updated classification model as the preset classification model includes: The target announcement type is used as a training label, and the to-be-announced file is used as input data to train the classification model, and the trained classification model is used as a preset classification model; The step of updating the parameter extraction model based on the verification result, the target announcement parameter, the target announcement type and the to-be-announced file, and using the updated parameter extraction model as the preset parameter extraction model, comprises: The target announcement type and the to-be-announced file are used as input parameters, and the target announcement parameter value is used as a training label to train the parameter extraction model, and the trained parameter extraction model is used as a preset parameter extraction model.

7. The method according to claim 5, characterized in that If the verification result is that the verification fails, the step of updating the classification model based on the verification result, the target announcement type and the to-be-announced file and using the updated classification model as the preset classification model includes: Acquire feedback information from business personnel, and use the feedback information from business personnel to update the target announcement parameter value and / or the target announcement type to obtain an updated parameter value and an updated announcement type; The classification model is trained using the updated announcement type as a training label and the to-be-announced file as input data, and the trained classification model is used as a preset classification model; The step of updating the parameter extraction model based on the verification result, the target announcement parameter, the target announcement type and the to-be-announced file, and using the updated parameter extraction model as the preset parameter extraction model, comprises: The updated announcement type and the to-be-announced file are used as input parameters, and the updated parameter value is used as a training label to train the parameter extraction model, and the trained parameter extraction model is used as a preset parameter extraction model.

8. A fund business data processing device, characterized in that: The device comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the fund business data processing method according to any one of claims 1 to 7.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the fund business data processing method according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the fund business data processing method according to any one of claims 1 to 7 are implemented.