Fund data management optimization method and system, medium and electronic equipment

Abstract

The invention discloses a fund data management optimization method and system, a medium and electronic equipment. The fund data management optimization method comprises the steps of obtaining fund associated data; screening the fund associated data through a preset detection rule to obtain to-be-audited information, and performing risk assessment through a preset second detection rule according to the to-be-audited information to obtain fund detection data of a corresponding risk level; the fund detection data is subjected to warehousing auditing based on the risk level, whether sub-fund integration data exists in a storage area of a database is detected, traversal is conducted on the passed fund detection data, associated sub-fund is detected, recursive merging processing is conducted, and sub-fund integration data is obtained and sent to the storage area to be output; according to the method, the fund associated data is directly acquired for integration processing, so that the problem of repeated data reference is avoided, more accurate fund comparison and risk assessment are facilitated, compliance risks and communication cost caused by data errors are reduced, and the data accuracy is greatly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of financial data statistics, and in particular to a fund data management optimization method, system, medium and electronic device. BACKGROUND

[0002] The existing financial data statistics, especially in the complex fund investment structure, when statistics the investment distribution of the entire fund industry, the existing statistical method depends on layer-by-layer reporting or aggregation from multiple channels. When the capital source of a sub-fund is not unique but is jointly invested by multiple parent funds (or fund of funds, FOF), how to accurately and redundantly perform data statistics. For example, a sub-fund simultaneously accepts investment from multiple parent funds, and the investment amount of the parent fund needs to be summed up, but in the statistical report, the total assets, investment portfolio, net value and other core information of the sub-fund are repeatedly quoted by multiple parent funds, and different parent funds may have different accounting methods or valuation points, resulting in differences in the valuation of the sub-fund reported by different parent funds. When higher-level regulatory agencies or data service providers aggregate fund industry data, the data of the sub-fund may be repeatedly calculated, resulting in an overestimation of the overall statistical figures, forming data redundancy, and seriously affecting the quality of the data and the effectiveness of the decision-making. SUMMARY

[0003] The present application relates to the field of financial data statistics, and in particular to a fund data management optimization method, system, medium and electronic device.

[0004] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows: a fund data management optimization method, comprising: obtaining fund correlation data, wherein the fund correlation data is complete data uploaded by a fund manager through a transmission port, and the fund correlation data includes fund basic information, partner / fund manager information, investor detailed information, parent fund information, investment situation information and / or reinvestment situation information; the fund basic information includes fund unified social code, name, organizational form, registration place, status, type or management fee deduction method; the investor detailed information includes unified social credit code, identity card, investor name, location, subscription amount or actual amount of investment; the investment situation information includes investment sub-fund related information and investment project related information; The fund correlation data is filtered by a preset first detection rule to obtain to-be-audited information, and risk assessment is performed on the to-be-audited information by a preset second detection rule to obtain fund detection data corresponding to the risk level; The fund detection data is audited based on the risk level, the fund detection data that passes is sent to a database for traversal, whether there is a circular dependency is detected, if not, a circular dependency exception is thrown, if yes, each sub-fund associated with the fund detection data is traversed, recursive merging processing is performed, sub-fund integration data is obtained, and the sub-fund integration data is sent to a storage area of the database for output.

[0005] As a further description of the above technical solution: the fund association data is screened by a first detection rule, which includes: According to the fund association data, whether it matches the first detection rule is detected, if yes, the first detection rule is triggered, and the corresponding fund association data is rejected, if not, the fund association data is sent to a to-be-audited area to obtain the to-be-audited information.

[0006] As a further description of the above technical solution: the risk assessment of the to-be-audited information according to the second detection rule includes: The to-be-audited information is sent to a pre-trained learning model for semantic feature optimization. The matching degree of the to-be-audited information and the second detection rule is judged, the corresponding risk feature is determined, the risk weight of the learning model is adjusted, and the risk level corresponding to the to-be-audited information is determined. The learning model outputs the fund detection data according to the risk level.

[0007] As a further description of the above technical solution: the first detection rule includes at least one of data integrity detection, format specification detection, business logic conflict detection, or blacklist detection. The second detection rule includes at least one of the China Securities Regulatory Commission filing information, the fund industry association data, the exchange announcement, the punishment record, or the negative information.

[0008] As a further description of the above technical solution: the fund detection data is audited based on the risk level, which includes: If it is low risk, the fund detection data is sent to an audit area, if it does not pass, it is sent to a risk feature library; If it is medium risk, the fund detection data is sent to an artificial audit area, if it passes, it is sent to the audit area, if it does not pass, it is sent to the risk feature library; If it is high risk, the fund detection data is rejected and sent to the risk feature library.

[0009] As a further description of the above technical solution: before the fund detection data that passes is sent to a database for traversal, it further includes: detecting whether the cached sub-fund integration data exists in the storage area of the database, if yes, outputting the existing sub-fund integration data; if no, traversing the fund detection data in the database to detect whether there is a circular dependency.

[0010] As a further description of the above technical solution: the recursive merging processing of traversing each sub-fund associated with the fund detection data includes: calculating a state mark for the fund detection data in the database; traversing the fund detection data to obtain associated sub-funds; detecting whether each sub-fund is stored in the storage area, if no, performing recursive merging processing to obtain the sub-fund integration data and sending to the storage area; if yes, obtaining a corresponding derivative index, removing the calculation state mark, and outputting from the storage area.

[0011] As a further description of the above technical solution: based on the fund integration data, a corresponding visual chart is output in a preset chart format.

[0012] Further comprising a redundant data optimization system for fund management, comprising: a collection module that obtains fund association data, wherein the fund association data is complete data uploaded by a fund manager through a transmission port, and the fund association data includes fund basic information, partner / fund manager information, investor detail information, parent fund information, investment situation information, and / or reinvestment situation information; the fund basic information includes a fund unified social code, name, organizational form, registration location, status, type, or management fee deduction method; the investor detail information includes a unified social credit code, identity card, investor name, location, subscription amount, or actual subscription amount; and the investment situation information includes investment sub-fund related information and investment project related information; a detection and evaluation module that filters the fund association data through a preset first detection rule to obtain to-be-audited information, and performs risk evaluation on the to-be-audited information through a preset second detection rule to obtain fund detection data of a corresponding risk level; a data optimization module that performs database storage auditing on the fund detection data based on the risk level, sends the passed fund detection data to the database for traversal to detect whether there is a circular dependency, if no, throws a circular dependency exception, if yes, traverses each sub-fund associated with the fund detection data to perform recursive merging processing to obtain sub-fund integration data, and sends to the storage area of the database for output.

[0013] The application further provides a computer readable storage medium storing a computer program for running the optimization method, wherein the computer program enables a computer to perform the optimization method according to any one of the above technical solutions.

[0014] The application further provides an electronic device comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the programs are used to perform the optimization method according to any one of the above technical solutions.

[0015] The above technical solutions have the following advantages or beneficial effects: By directly obtaining fund correlation data for integration processing, replacing the existing calculation method, the problem of repeated reference data is avoided, the obtained fund correlation data is screened and evaluated according to the first detection rule and the second detection rule, more accurate fund comparison and risk assessment are facilitated, the risk level is based on the risk level for warehousing audit, and whether there is a circular dependency is detected, the sub-funds related to the fund detection data with circular dependency are recursively merged and processed, the compliance risk and communication cost caused by data errors are reduced, 60% of redundant data can be reduced, the report processing efficiency can be improved by 4 times, automatic deduplication statistics are realized, and the accuracy of data is greatly improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0017] Figure 1 The flowchart of the optimization method proposed in the application; Figure 2 The flowchart of the risk assessment by the second detection rule in the application; Figure 3 The flowchart of the recursive merging processing in the application; Figure 4 The structural principle diagram of the optimization system proposed in the application. DETAILED DESCRIPTION

[0018] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0019] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application. Figure 1 According to an embodiment of the present application, a fund data management optimization method comprises the following steps: S1, obtaining fund correlation data, wherein the fund correlation data is complete data uploaded by a fund manager through a transmission port, and the fund correlation data comprises fund basic information, partner / fund manager information, investor detail information, parent fund information, investment situation information and / or reinvestment situation information; S2, screening the fund correlation data through a preset first detection rule to obtain to-be-audited information, and performing risk assessment on the to-be-audited information through a preset second detection rule to obtain fund detection data corresponding to a risk level; S3, performing database storage auditing on the fund detection data based on the risk level, sending the passed fund detection data to a database for traversal, detecting whether there is a circular dependency, if not, throwing a circular dependency exception, if yes, traversing each sub-fund associated with the fund detection data to perform recursive merging processing to obtain sub-fund integration data, and sending the sub-fund integration data to a storage area of the database for output.

[0020] In the embodiment, the fund manager fills in and uploads fund association data. If the fund is a parent and child fund structure, the parent fund manager fills in the parent fund data, and the child fund manager fills in the child fund data, avoiding multiple and repeated filling by the child fund manager. The fund association data is obtained through a transmission port identifier to ensure the uniqueness of the fund association data and the integrity of the entity in the database. The obtained fund association data is screened and evaluated according to the first detection rule and the second detection rule to filter out obviously illegal and false information. The learning model is used to screen and evaluate the existing risk features, which facilitates more accurate fund comparison and risk assessment. Based on the risk level, the data is audited and stored, and the depth-first algorithm is used to detect whether there is a circular dependency. If there is no circular dependency, a circular dependency exception is thrown to avoid subsequent integration into a logical paradox. If there is, the child fund associated with the fund detection data with circular dependency is recursively merged and processed to prevent multiple threads from reading and polluting the data, and to reduce redundant data. By directly obtaining the fund association data, the related child fund is recursively merged and processed, replacing the existing calculation method, avoiding the problem of repeated reference data, reducing the compliance risk and communication cost caused by data errors, and reducing redundant data by 60%, improving report processing efficiency by 4 times, achieving automatic deduplication statistics, and greatly improving the accuracy of data.

[0021] The fund basic information includes the fund unified social code, name, organizational form, registered place, status, type or management fee deduction method; the investor detailed information includes the unified social credit code, identity card, investor name, location, subscription amount or paid-in amount; the investment situation information includes the investment child fund related information and the investment project related information.

[0022] The fund association data is screened by the first detection rule, which includes: According to the fund association data, it is detected whether it matches the first detection rule. If yes, the first detection rule is triggered, and the corresponding fund association data is rejected. If not, the fund association data is sent to the area to be audited to obtain the information to be audited.

[0023] In the embodiment, the first detection rule is a set hard rule, including at least one of data integrity detection, format specification detection, business logic conflict detection or blacklist detection. If the first detection rule is triggered, the corresponding fund association data is rejected immediately and recorded, and the specific rejection reason is returned to the fund manager for targeted correction. If it is not triggered, the fund association data is sent to the area to be audited.

[0024] In the matching detection, the fund association data is sequentially detected by the first detection rule, and the subsequent check is terminated as soon as any of the data integrity detection, format specification detection, business logic conflict detection or blacklist detection triggers, so as to avoid invalid computing power consumption, and the rejection reason is generated according to the triggered detection condition. The first detection rule can quickly filter out obviously invalid fund association data, avoid entering the subsequent learning model for evaluation and recursive processing, and reduce the waste of algorithm computing power.

[0025] Referring to Figure 2 According to the to-be-audited information, the risk assessment by the second detection rule includes: S21, sending the to-be-audited information to the pre-trained learning model for semantic feature optimization; S22, judging the matching degree of the to-be-audited information and the second detection rule, determining the corresponding risk feature, adjusting the risk weight of the learning model, and determining the risk level corresponding to the to-be-audited information; S23, the learning model outputs the fund detection data according to the risk level.

[0026] In this embodiment, the pre-trained learning model and the second detection rule are used for risk assessment, and the to-be-audited information is analyzed and optimized at the semantic level. The second detection rule includes at least one of the information recorded by the CSRC, the data of the fund industry association, the exchange announcement, the punishment record or the negative information, so as to identify the hidden risk. The learning model continuously receives the to-be-audited information fed back by the audit, and iteratively optimizes the weight distribution of the risk feature according to the to-be-audited information. The learning model can quickly complete the risk assessment of the to-be-audited information, and the judgment based on the vector similarity can avoid the subjective deviation of manual audit.

[0027] Specifically, the learning model can be selected as a Sentence-Bert model, a Bert-Base-Chinses model or a ConSERT model. The Bert-Base-Chinses model is a general Chinese pre-training language model, which is a basic version of the BERT architecture, and is used to capture the language rules of Chinese, mainly through 12 layers of Transformer, 768 hidden layer dimensions, and 12 attention heads. The Sentence-Bert model encodes each sentence independently through a twin / triplet network structure, and then obtains a fixed-size sentence vector representation, which is used to solve the bottleneck of BERT in semantic similarity calculation, clustering and other tasks. Each sentence can be independently encoded without concatenating two sentences into the input, but is input into the same BERT model at the same time, and the output of the BERT is pooled to obtain a fixed sentence vector, such as the MEAN strategy, which takes the average of all word vectors. During training, classification loss, cosine similarity loss or triple loss is used, the purpose of which is to make the distance between semantically similar sentences in the vector space closer, and the distance between semantically different sentences farther. The ConSERT model is a contrastive learning framework that improves the quality of BERT sentence representation, especially in solving the anisotropic problem (i.e., the word / sentence vector distribution is not uniform, collapsed in a narrow conical space), and is excellent in optimizing the BERT sentence representation through data augmentation and contrastive learning without changing the structure of the BERT model itself. Among them, data augmentation is to apply multiple perturbations to the same sentence, such as cropping, word order scrambling, word covering, deletion, etc., to generate multiple different views. Contrastive learning is to let the vector representations of different views of the same sentence be as close as possible in space as positive samples, and as far away as possible from the vector representations of other sentences as negative samples during model training.

[0028] The fund detection data is audited based on the risk level and includes: If it is low risk, the fund detection data is sent to the storage audit area, and if it is not passed, it is sent to the risk feature library; If it is medium risk, the fund detection data is sent to the manual audit area, if it is passed, it is sent to the storage audit area, and if it is not passed, it is sent to the risk feature library; If it is high risk, the fund detection data is rejected and sent to the risk feature library.

[0029] In the embodiment, the risk level output according to the second detection rule corresponds to different audit processes. For low-risk fund detection data, the fund detection data is directly sent to the storage audit area, and if the audit is passed, the fund detection data is sent to the database, and if the audit is not passed, the fund detection data is synchronized to the risk feature library and the specific reason for not passing is recorded. For medium-risk fund detection data, the fund detection data is first sent to the manual audit area for manual audit, and if the audit is passed, the fund detection data is sent to the storage audit area, and if the audit is not passed, the fund detection data is synchronized to the risk feature library after the corresponding risk reason is determined. For high-risk fund detection data, the fund detection data is directly rejected to be stored and automatically synchronized to the risk feature library, which is used to optimize the second detection rule and adjust the risk weight of the learning model.

[0030] Before the passed fund detection data is sent to the database for traversal, the following steps are further included. Detecting whether there is cached sub-fund integration data in the storage area of the database, If yes, the existing sub-fund integration data is outputted. If no, the fund detection data in the database is traversed to detect whether there is a circular dependency.

[0031] In the embodiment, before the fund detection data is traversed, it is detected whether there is already merged sub-fund integration data in the storage area of the database. If yes, the sub-fund integration data is outputted, and if no, the fund detection data in the database is traversed to detect whether there is a circular dependency.

[0032] Referring to Figure 3 , the recursive merging processing of traversing each sub-fund associated with the fund detection data includes the following steps. S31, marking the calculation state of the fund detection data in the database; S32, traversing the fund detection data to obtain the associated sub-fund; S33, detecting whether each sub-fund is stored in the storage area, and if no, performing recursive merging processing to obtain sub-fund integration data and sending the sub-fund integration data to the storage area; S34, if yes, obtaining the corresponding derivative index, removing the calculation state mark, and outputting from the storage area.

[0033] In the embodiment, since the information in the database is unique although the sub-foundation is invested by multiple parent foundations, each foundation or project is only calculated once, so by marking the calculation state of the fund detection data entering the integration process, the fund detection data of the completed integration is avoided to be read, causing data redundancy, by traversing the fund detection data, the fund basic information, investor detailed information and investment situation information are obtained according to the corresponding fund association data, the directly held asset information is processed, the associated investment sub-foundation related information is obtained, it is detected whether each sub-foundation is stored in the storage area, if not, recursive processing is performed, the sub-foundation integration data is obtained and sent to the storage area, wherein the storage area displays the fund association data and the fund basic information of the sub-foundation in the form of a table, until there is no sub-foundation, the loop ends, the corresponding derivative index is obtained, the calculation state mark is removed, and the sub-foundation integration data is output from the storage area.

[0034] In a specific embodiment, the fund basic information about integrated circuits in the fund to be queried is checked first to see whether the fund about integrated circuits in the storage area has completed statistics. If yes, the fund is directly output, and if not, it is detected whether the fund detection data has integrated circuits. If yes, the fund about integrated circuits can be cyclically counted, and if not, a loop dependency exception is thrown. When the fund basic information about integrated circuits is counted, the state of the fund basic information about integrated circuits in the calculation is marked first, then the basic information of the fund is obtained, the information result about integrated circuits is initialized, and the directly related assets are processed. The fund basic information of the sub-foundation is queried, the related information is counted if there is a related sub-foundation, the associated sub-foundation is recursively merged and processed, and is stored in the storage area. Until the loop ends, the related derivative index such as the proportion of capital expenditure to sales is calculated, the calculation mark, the sub-foundation integration data, and the output are removed.

[0035] Based on the fund integration data, the corresponding visual chart is output in a preset chart format.

[0036] In the embodiment, the fund integration data in the database is converted into intuitive visual charts and interactive dashboards by using a business intelligence tool (BI), so as to facilitate business analysis and decision-making. By connecting the data source, the data channel is opened, the connection information of the database is filled in, such as the server address, the database name, the user name, the password, etc., based on the connected data source, the preset chart format is selected, the association operation is used, and the mysql database query statement is mainly used; the data display of the database is imported for large-screen display, the fields to be analyzed are selected, the fields are customized, the data is filtered according to the needs, the required visual table is formed, and business analysis and decision-making are facilitated.

[0037] Reference Figure 4Also included is an embodiment of a redundant data optimization system for fund management, comprising: a collection module 1 that acquires fund-related data, wherein the fund-related data is complete data uploaded by a fund manager through a transmission port, and the fund-related data includes fund basic information, partner / fund manager information, investor detail information, parent fund information, investment situation information, and / or reinvestment situation information; the fund basic information includes a fund unified social code, name, organizational form, registration location, status, type, or management fee deduction method; the investor detail information includes a unified social credit code, ID card, investor name, location, amount of capital contribution or amount of capital actually paid; and the investment situation information includes investment sub-fund related information and investment project related information; a detection and evaluation module 2 that screens the fund-related data through a preset first detection rule to obtain to-be-audited information, and performs risk assessment on the to-be-audited information through a preset second detection rule to obtain fund detection data corresponding to a risk level; a data optimization module 3 that performs database auditing on the fund detection data based on the risk level, sends the passed fund detection data to a database for traversal, detects whether there is a circular dependency, and if not, throws a circular dependency exception, and if so, traverses each sub-fund associated with the fund detection data, performs recursive merging processing to obtain sub-fund integration data, and sends the sub-fund integration data to a storage area of the database for output.

[0038] In this embodiment, the collection module 1 acquires fund-related data, the detection and evaluation module 2 receives the fund-related data sent by the collection module 1, detects whether it matches the first detection rule to obtain to-be-audited information, sends the to-be-audited information to a pre-trained learning model, judges the matching degree of the to-be-audited information with the second detection rule, outputs fund detection data, and the data optimization module 3 receives the fund detection data sent by the detection and evaluation module 2, performs database auditing, sends the passed fund detection data to a database, detects whether there is cached sub-fund integration data in the storage area of the database, if not, traverses the fund detection data in the database, detects whether there is a circular dependency, if so, traverses each sub-fund associated with the fund detection data, performs recursive merging processing to obtain corresponding sub-fund integration data.

[0039] Also included is a computer-readable storage medium that stores a computer program for running an optimization method, wherein the computer program causes a computer to perform the following steps: S1, acquiring fund-related data, wherein the fund-related data is complete data uploaded by a fund manager through a transmission port, and the fund-related data includes fund basic information, partner / fund manager information, investor detail information, parent fund information, investment situation information, and / or reinvestment situation information; S2. Fund-related data is filtered through a preset first detection rule to obtain information to be reviewed. Based on the information to be reviewed, a risk assessment is performed through a preset second detection rule to obtain fund detection data with the corresponding risk level. S3. Based on the risk level, the fund detection data is reviewed for entry into the database. The approved fund detection data is sent to the database for traversal to check for circular dependencies. If no circular dependency exists, a circular dependency exception is thrown. If so, the sub-funds associated with each fund detection data are traversed and recursively merged to obtain the sub-fund integrated data, which is then sent to the database storage area for output.

[0040] The computer-readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of a computer program from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a computer-readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the computer-readable storage medium. Of course, the computer-readable storage medium can also be a component of the processor. The processor and the computer-readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the ASIC can reside in a user equipment. Of course, the processor and the computer-readable storage medium can also exist as discrete components in a communication device.

[0041] Specifically, the computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The storage medium can be any available medium accessible to general-purpose or special-purpose computers.

[0042] It also includes an electronic device, comprising: One or more processors; memory; and One or more programs, wherein the programs are stored in memory and configured to be executed by one or more processors, the programs being used to perform the following steps: S1. Obtain fund-related data, which is the complete data uploaded by the fund manager through the transmission port. The fund-related data includes basic fund information, partner / fund manager information, investor details, parent fund information, investment information and / or reinvestment information. S2. Fund-related data is filtered through a preset first detection rule to obtain information to be reviewed. Based on the information to be reviewed, a risk assessment is performed through a preset second detection rule to obtain fund detection data with the corresponding risk level. S3. Based on the risk level, the fund detection data is reviewed for entry into the database. The approved fund detection data is sent to the database for traversal to check for circular dependencies. If no circular dependency exists, a circular dependency exception is thrown. If so, the sub-funds associated with each fund detection data are traversed and recursively merged to obtain the sub-fund integrated data, which is then sent to the database storage area for output.

[0043] Memory is used to store computer programs. This memory may include high-speed random access memory (RAM) and may also include non-volatile memory (Non-volatile memory). Volatile Memory (NVM), such as at least one disk storage device, and can also be a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.

[0044] A processor is used to execute computer programs stored in memory. The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0045] Alternatively, the memory can be either standalone or integrated with the processor.

[0046] When memory is a device independent of the processor, electronic devices may also include a bus. This bus is used to connect the memory and the processor. This bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc.

[0047] It should be noted that, through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the prior art, can be embodied in the form of software products. These computer software products can be stored in computer-readable storage media, such as ROM / RAM, magnetic disks, optical disks, etc., and include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or certain portions of the embodiments. In this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, 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. In the absence of further restrictions, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0048] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An optimized method for fund data management, characterized in that, include: Obtain fund-related data, wherein the fund-related data is complete data uploaded by the fund manager through the transmission port. The fund-related data includes basic fund information, partner information, fund manager information, investor details, parent fund information, investment information, and / or reinvestment information. The basic fund information includes the fund's unified social credit code, name, organizational form, place of registration, status, type, or management fee calculation method. The investor details include the unified social credit code, ID card, investor name, region, and subscribed or paid-in amount. The investment information includes information related to invested sub-funds and investment projects. The fund-related data is filtered through a preset first detection rule to obtain information to be reviewed. Based on the information to be reviewed, a risk assessment is performed through a preset second detection rule to obtain fund detection data with the corresponding risk level. Based on the risk level, the fund detection data is reviewed for entry into the database. Passing fund detection data is sent to the database for traversal to check for circular dependencies. If no circular dependency exists, a circular dependency exception is thrown. If so, each sub-fund associated with the fund detection data is traversed, and a recursive merging process is performed. This merging process includes checking if the parent fund's sub-funds have investments in other sub-funds with corresponding statistical information. The recursive method includes using counting and accumulation to statistically analyze the sub-fund information until the parent fund's sub-funds have investments in sub-funds that do not have corresponding investments in other sub-funds. The resulting integrated sub-fund data is then sent to the database's storage area for output.

2. The method according to claim 1, characterized in that: The fund-related data is filtered using the first detection rule, including: Based on the fund association data, it is checked whether it matches the first detection rule. If it does, the first detection rule is triggered to reject the corresponding fund association data. If not, the fund association data is sent to the review area to obtain the review information.

3. The method according to claim 1, characterized in that: The risk assessment based on the information to be reviewed using the second detection rule includes: The information to be reviewed is sent to a pre-trained learning model for semantic feature optimization. Determine the matching degree between the information to be reviewed and the second detection rule, identify the corresponding risk characteristics, adjust the risk weight of the learning model, and determine the risk level corresponding to the information to be reviewed; The learning model outputs the fund detection data based on the risk level.

4. The method according to claim 1, characterized in that: The first detection rule includes at least one of data integrity detection, format standardization detection, business logic conflict detection, or blacklist detection; The second detection rule includes at least one of the following: China Securities Regulatory Commission filing information, Asset Management Association of China data, exchange announcements, penalty records, or negative information.

5. The method according to claim 1, characterized in that: The process of reviewing the fund detection data for inclusion in the database based on the risk level includes: If the risk is low, the fund detection data will be sent to the entry review area. Data that passes the entry review area will be saved to the database. Data that fails the review will be sent to the risk feature database. If the risk level is medium, the fund detection data will be sent to the manual review area. Data that passes the review in the entry review area will be saved to the database. If it passes the review, it will be sent to the entry review area; if it fails the review, it will be sent to the risk feature database. If the risk level is high, the fund detection data will be rejected and sent to the risk feature database.

6. The method according to claim 1, characterized in that: Before sending the passed fund detection data to the database for traversal, the process also includes: Check if the cached sub-fund integration data exists in the storage area of ​​the database. If so, output the integrated data of the existing sub-funds; If not, iterate through the fund detection data in the database to check for circular dependencies.

7. The method according to claim 1, characterized in that: The recursive merging process of traversing each sub-fund associated with the fund detection data includes: The status label is calculated for the fund detection data in the database; Traverse the fund detection data to obtain the associated sub-funds; Check whether each of the sub-funds is stored in the storage area. If not, perform recursive merging to obtain the integrated data of the sub-funds and send it to the storage area. If so, obtain the corresponding derived index, remove the calculation status flag, and output it from the storage area.

8. The method according to claim 1, characterized in that: Based on the integrated fund data, corresponding visual charts are output according to a preset chart format.

9. A redundant data optimization system for fund management, characterized in that, include: The data acquisition module obtains fund-related data, which is complete data uploaded by the fund manager through the transmission port. This fund-related data includes basic fund information, partner / fund manager information, investor details, parent fund information, investment information, and / or reinvestment information. The basic fund information includes the fund's unified social credit code, name, organizational form, registered address, status, type, or management fee calculation method. The investor details include the unified social credit code, ID card number, investor name, location, and subscribed or paid-in capital. The investment information includes information related to invested sub-funds and investment projects. The detection and evaluation module filters the fund-related data according to a preset first detection rule to obtain information to be reviewed. Based on the information to be reviewed, a risk assessment is performed according to a preset second detection rule to obtain fund detection data with the corresponding risk level. The data optimization module reviews the fund detection data for entry into the database based on the risk level. The approved fund detection data is sent to the database for traversal to check for circular dependencies. If no circular dependency exists, a circular dependency exception is thrown. If so, the sub-funds associated with each fund detection data are traversed and recursively merged to obtain integrated sub-fund data, which is then sent to the storage area of ​​the database for output.

10. A computer-readable storage medium and electronic device, characterized in that, The storage medium stores a computer program for running the optimization method; The electronic device includes: One or more processors; memory; and One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs being used to perform the optimization method as described in any one of claims 1-8.

Images