A method and system for enterprise financial management for extracting financial data

Abstract

The present invention provides an enterprise financial management method and system for financial data extraction, which relates to the field of data processing technology. In the present invention, the output represents the information extraction result; the initial member information feature representation is output; the updated member information feature representation is formed; the representative information extraction result and the information matching parameters between the network members are output; at least one first data block identification information is determined based on the information matching parameters; and at least one first financial data block to be processed is determined based on the mapping relationship representation data corresponding to the first data block identification information, and the target matching parameters between the first financial data block to be processed and the data extraction identification information are formed; according to the target matching parameters, at least one first financial data block to be processed is subjected to a target data extraction operation to obtain a target financial data block. Based on the above method, the reliability of financial data extraction can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an enterprise financial management method and system for extracting financial data. Background Art

[0002] The amount of financial data is generally large. Therefore, in the process of forming or storing, the massive amount of financial data is generally divided into blocks based on different needs to form different financial data blocks. For example, the data of a project in a time period can be used as a financial data block. Based on this, when data extraction is required, the required financial data blocks can be extracted. However, in the prior art, there is a problem of low reliability for financial data extraction. Summary of the invention

[0003] In view of this, an object of the present invention is to provide an enterprise financial management method and system for extracting financial data, so as to improve the reliability of financial data extraction.

[0004] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions:

[0005] An enterprise financial management method for extracting financial data, comprising:

[0006] When the financial data extraction instruction information is obtained, performing a representative information extraction operation on the data extraction identification information in the financial data extraction instruction information to output a representative information extraction result corresponding to the data extraction identification information;

[0007] Performing key information mining operations on each network member in the representative information extraction result and the optimized data block identification network to output key information feature representations corresponding to the representative information extraction result and initial member information feature representations corresponding to the network members;

[0008] Analyzing the importance characterization parameters corresponding to the network members based on the identification correlation information between the data block identification information corresponding to the network members; and updating the initial member information feature representation according to the importance characterization parameters to form an updated member information feature representation corresponding to the network members;

[0009] Determine matching characterization parameters between the key information feature representation and the updated member information feature representation, and perform interval mapping operation on the matching characterization parameters to output information matching parameters between the representative information extraction result and the relationship network members;

[0010] According to the information matching parameter, at least one first data block identification information is determined from the data block identification information combination; and, based on the mapping relationship representation data corresponding to the first data block identification information, at least one first financial data block to be processed is determined from the financial data block cluster to be processed, and the identification reliability parameter corresponding to the first financial data block to be processed is multiplied by the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information, wherein the mapping relationship representation data is used to reflect the mapping information between the hash function value of the financial data block and the data block identification information;

[0011] According to the target matching parameter, the at least one first financial data block to be processed is subjected to a target data extraction operation to obtain a corresponding target financial data block.

[0012] In some preferred embodiments, in the above-mentioned enterprise financial management method for extracting financial data, the enterprise financial management method further comprises:

[0013] Determine a cluster of financial data blocks to be processed, and use at least one financial data analysis network to perform multiple levels of financial data analysis operations on the financial data blocks to be processed in the cluster of financial data blocks to be processed, so as to form multiple pieces of identification information of each financial data block to be processed and an identification reliability parameter corresponding to each piece of identification information of the financial data block to be processed;

[0014] According to the identification reliability parameter corresponding to the identification information of the data block to be processed, determining the screening data block identification information corresponding to the financial data block to be processed from the plurality of identification information of the data blocks to be processed;

[0015] Determining, from the configuration data block identification information cluster, configuration data block identification information having correlation with the screening data block identification information as the relevant data block identification information;

[0016] Combining the relevant data block identification information and the screening data block identification information to form a data block identification information combination corresponding to the financial data block cluster to be processed;

[0017] Determining identification correlation information between the data block identification information from the data block identification information combination, wherein the identification correlation information includes content correlation information and upper-lower level correlation information;

[0018] Based on the upper-lower level correlation information, with the data block identification information as a member of the relationship network, determining an initial data block identification relationship network;

[0019] Based on the content relevance information, the initial data block identifier relationship network is optimized to form a corresponding optimized data block identifier relationship network.

[0020] In some preferred embodiments, in the above-mentioned enterprise financial management method for extracting financial data, the step of determining the initial data block identification relationship network based on the superior-subordinate correlation information and when the data block identification information is used as a member of the relationship network includes:

[0021] Determine the data block identification information with upper and lower level correlation information from the data block identification information combination to output the corresponding upper and lower level data block identification information combination;

[0022] Based on the correlation type of the upper and lower level correlation information, clustering the data block identification information in the upper and lower level data block identification information combination;

[0023] According to the obtained clustering processing result, the data block identification information in the combination of upper and lower data block identification information is marked as a relationship network member to form an initial data block identification relationship network.

[0024] In some preferred embodiments, in the above-mentioned enterprise financial management method for extracting financial data, the step of optimizing the initial data block identifier relationship network based on the content relevance information to form a corresponding optimized data block identifier relationship network includes:

[0025] Determine, from the initial data block identification relationship network, the relationship network members corresponding to the data block identification information having content correlation information, to output at least one relationship network member combination;

[0026] According to the content relevance information corresponding to the relationship network member combination, the initial data block identifier relationship network is optimized to form an intermediate data block identifier relationship network;

[0027] From the data block identification information combination, data block identification information having content correlation information corresponding to the network members on the intermediate data block identification relationship network is selected to form a corresponding optimized data block identification relationship network.

[0028] In some preferred embodiments, in the above-mentioned enterprise financial management method for extracting financial data, the step of selecting data block identification information having content correlation information with data block identification information corresponding to network members on the intermediate data block identification relationship network from the data block identification information combination to form a corresponding optimized data block identification relationship network includes:

[0029] Selecting data block identification information having content correlation information with data block identification information corresponding to a network member on the intermediate data block identification relationship network from the data block identification information combination to form a relevant data block identification information combination;

[0030] Analyzing the relationship network distribution information of the data block identification information based on the content correlation information corresponding to each data block identification information in the combination of related data block identification information;

[0031] According to the relationship network distribution information, the data block identification information in the relevant data block identification information combination is marked as a relationship network member, and is distributed to the intermediate data block identification relationship network to form a corresponding optimized data block identification relationship network.

[0032] In some preferred embodiments, in the above-mentioned enterprise financial management method for extracting financial data, the steps of determining a cluster of financial data blocks to be processed, and using at least one financial data analysis network to respectively perform multiple levels of financial data analysis operations on the financial data blocks to be processed in the cluster of financial data blocks to be processed, so as to form multiple pieces of identification information of each financial data block to be processed and an identification reliability parameter corresponding to each piece of identification information of the data block to be processed, include:

[0033] Determine a financial data block cluster to be processed;

[0034] Using at least one financial data analysis network, a key information mining operation is performed on the unprocessed financial data blocks in the unprocessed financial data block cluster to form a data block key information feature representation corresponding to the unprocessed financial data blocks at the data level corresponding to the financial data analysis network;

[0035] Based on the characteristic representation of the key information of the data block, the identification information of the data block to be processed corresponding to the financial data block to be processed at the data level and the identification reliability parameter corresponding to the identification information of the data block to be processed are analyzed.

[0036] In some preferred embodiments, in the above-mentioned enterprise financial management method for extracting financial data, after the step of determining, based on the identification reliability parameter corresponding to the identification information of the data block to be processed, the screening data block identification information corresponding to the financial data block to be processed from the plurality of identification information of the data blocks to be processed, the enterprise financial management method further comprises:

[0037] A hash function mapping operation is performed on the financial data block to be processed to output a hash function value corresponding to the financial data block to be processed; and based on the hash function value and the data block substantial information corresponding to the financial data block to be processed, data block representation information corresponding to the financial data block to be processed is determined; and, based on the screening data block identification information, the financial data block to be processed is clustered, and, based on the obtained clustering result and the hash function value corresponding to the financial data block to be processed, a mapping relationship representation data between the hash function value and the screening data block identification information is determined;

[0038] The step of determining at least one first financial data block to be processed from the cluster of financial data blocks to be processed based on the mapping relationship representation data corresponding to the first data block identification information, and performing a multiplication operation on the identification reliability parameter corresponding to the first financial data block to be processed and the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information comprises:

[0039] Based on the mapping relationship representation data, determine the target hash function value corresponding to the first data block identification information from the hash function value; determine the target data block representation information corresponding to the target hash function value from the data block representation information; based on the target data block representation information, select at least one first financial data block to be processed from the financial data block cluster to be processed corresponding to the data block identification information combination; and multiply the identification reliability parameter corresponding to the first financial data block to be processed by the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information.

[0040] The embodiment of the present invention further provides an enterprise financial management system for extracting financial data, comprising:

[0041] A representative information extraction module is used to perform a representative information extraction operation on the data extraction identification information contained in the financial data extraction indication information when the financial data extraction indication information is obtained, so as to output a representative information extraction result corresponding to the data extraction identification information;

[0042] A key information mining module, used to perform key information mining operations on each network member in the representative information extraction result and the optimized data block identification network, so as to output a key information feature representation corresponding to the representative information extraction result and an initial member information feature representation corresponding to the network member;

[0043] A feature representation updating module is used to analyze the importance representation parameters corresponding to the relationship network members based on the identification correlation information between the data block identification information corresponding to the relationship network members; and, based on the importance representation parameters, update the initial member information feature representation to form an updated member information feature representation corresponding to the relationship network members;

[0044] A matching parameter processing module, used to determine matching characterization parameters between the key information feature representation and the updated member information feature representation, and to perform interval mapping operation on the matching characterization parameters to output information matching parameters between the representative information extraction result and the relationship network members;

[0045] a matching parameter updating module, configured to determine at least one first data block identification information from a combination of data block identification information according to the information matching parameter; and, based on the mapping relationship representation data corresponding to the first data block identification information, determine at least one first financial data block to be processed from a cluster of financial data blocks to be processed, and perform a multiplication operation on an identification reliability parameter corresponding to the first financial data block to be processed and the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information, wherein the mapping relationship representation data is used to reflect the mapping information between the hash function value of the financial data block and the data block identification information;

[0046] The data block extraction module is used to perform a target data extraction operation on the at least one first financial data block to be processed according to the target matching parameter to obtain a corresponding target financial data block.

[0047] In some preferred embodiments, in the above-mentioned enterprise financial management system for extracting financial data, the enterprise financial management system further includes other functional modules for:

[0048] Determine a cluster of financial data blocks to be processed, and use at least one financial data analysis network to perform multiple levels of financial data analysis operations on the financial data blocks to be processed in the cluster of financial data blocks to be processed, so as to form multiple pieces of identification information of each financial data block to be processed and an identification reliability parameter corresponding to each piece of identification information of the financial data block to be processed;

[0049] According to the identification reliability parameter corresponding to the identification information of the data block to be processed, determining the screening data block identification information corresponding to the financial data block to be processed from the plurality of identification information of the data blocks to be processed;

[0050] Determining, from the configuration data block identification information cluster, configuration data block identification information having correlation with the screening data block identification information as the relevant data block identification information;

[0051] Combining the relevant data block identification information and the screening data block identification information to form a data block identification information combination corresponding to the financial data block cluster to be processed;

[0052] Determining identification correlation information between the data block identification information from the data block identification information combination, wherein the identification correlation information includes content correlation information and upper-lower level correlation information;

[0053] Based on the upper-lower level correlation information, with the data block identification information as a member of the relationship network, determining an initial data block identification relationship network;

[0054] Based on the content relevance information, the initial data block identifier relationship network is optimized to form a corresponding optimized data block identifier relationship network.

[0055] In some preferred embodiments, in the enterprise financial management system for extracting financial data, the determining of the initial data block identification relationship network based on the superior-subordinate correlation information and when the data block identification information is used as a member of the relationship network includes:

[0056] Determine the data block identification information with upper and lower level correlation information from the data block identification information combination to output the corresponding upper and lower level data block identification information combination;

[0057] Based on the correlation type of the upper and lower level correlation information, clustering the data block identification information in the upper and lower level data block identification information combination;

[0058] According to the obtained clustering processing result, the data block identification information in the combination of upper and lower data block identification information is marked as a relationship network member to form an initial data block identification relationship network.

[0059] An enterprise financial management method and system for financial data extraction provided by an embodiment of the present invention can first perform a representative information extraction operation on the data extraction identification information in the financial data extraction indication information to output the representative information extraction result corresponding to the data extraction identification information; perform a key information mining operation on each network member in the representative information extraction result and the optimized data block identification network to output the key information feature representation corresponding to the representative information extraction result and the initial member information feature representation corresponding to the network member; analyze the importance representation parameters corresponding to the network member based on the identification correlation information between the data block identification information corresponding to the network member; and, based on the importance representation parameters, perform an update operation on the initial member information feature representation to form an updated member information feature representation corresponding to the network member; determine the key information feature representation. The matching characterization parameters between the feature representation and the updated member information feature representation are obtained, and the matching characterization parameters are interval mapped to output the information matching parameters between the representative information extraction result and the relationship network members; at least one first data block identification information is determined from the data block identification information combination according to the information matching parameters; and at least one first financial data block to be processed is determined from the financial data block cluster to be processed based on the mapping relationship characterization data corresponding to the first data block identification information, and the identification reliability parameter corresponding to the first financial data block to be processed is multiplied by the information matching parameter to form the target matching parameter between the first financial data block to be processed and the data extraction identification information; according to the target matching parameter, at least one first financial data block to be processed is subjected to the target data extraction operation to obtain the corresponding target financial data block. Based on the above content, before the financial data block is extracted, the reliability of the target matching parameters obtained is higher due to the sequential operation of table information extraction, key information mining and feature representation update, thereby ensuring the reliability of financial data extraction.

[0060] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A structural block diagram of an enterprise financial management platform for extracting financial data provided by an embodiment of the present invention.

[0062] Figure 2 A flowchart of the steps included in the enterprise financial management method for extracting financial data provided by an embodiment of the present invention.

[0063] Figure 3 A schematic diagram of various modules included in the enterprise financial management system for financial data extraction provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0065] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] like Figure 1 As shown, an embodiment of the present invention provides an enterprise financial management platform for extracting financial data. The enterprise financial management platform for extracting financial data may include a memory and a processor.

[0067] In detail, the memory and the processor are electrically connected directly or indirectly to achieve data transmission or interaction. For example, they can be electrically connected to each other through one or more communication buses or signal lines. The memory can store at least one software function module (computer program) that can exist in the form of software or firmware. The processor can be used to execute the executable computer program stored in the memory, thereby implementing the enterprise financial management method for financial data extraction provided by the embodiment of the present invention (as described later).

[0068] Optionally, in some embodiments, the memory may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc.

[0069] Optionally, in some embodiments, the processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0070] Optionally, in some implementations, the enterprise financial management platform for extracting financial data may be a server with data processing capabilities.

[0071] Combination Figure 2 The embodiment of the present invention also provides an enterprise financial management method for financial data extraction, which can be applied to the above-mentioned enterprise financial management platform for financial data extraction. The method steps defined in the process related to the enterprise financial management method for financial data extraction can be implemented by the enterprise financial management platform for financial data extraction.

[0072] The following will Figure 2 The specific process shown is explained in detail.

[0073] Step S110, when the financial data extraction instruction information is obtained, a representative information extraction operation is performed on the data extraction identification information contained in the financial data extraction instruction information to output a representative information extraction result corresponding to the data extraction identification information.

[0074] In an embodiment of the present invention, the enterprise financial management platform for financial data extraction can, upon obtaining financial data extraction indication information, perform a representative information extraction operation on the data extraction identification information contained in the financial data extraction indication information to output a representative information extraction result corresponding to the data extraction identification information. The representative information extraction result can be screened out representative words, sentences, audio clips, image frames, etc.

[0075] Step S120, performing key information mining operations on each network member in the representative information extraction result and the optimized data block identification network to output key information feature representation corresponding to the representative information extraction result and initial member information feature representation corresponding to the network member.

[0076] In an embodiment of the present invention, the enterprise financial management platform for financial data extraction can perform key information mining operations on each network member in the representative information extraction result and the optimized data block identification network to output key information feature representations corresponding to the representative information extraction result and initial member information feature representations corresponding to the network members. The key information mining operation can be implemented by a configured neural network, and the feature representation can be in the form of a vector.

[0077] Step S130, based on the identification correlation information between the data block identification information corresponding to the relationship network members, analyzing the importance characterization parameters corresponding to the relationship network members; and, according to the importance characterization parameters, updating the initial member information feature representation to form an updated member information feature representation corresponding to the relationship network members.

[0078] In an embodiment of the present invention, the enterprise financial management platform for financial data extraction can analyze the importance characterization parameters corresponding to the relationship network members based on the identification correlation information between the data block identification information corresponding to the relationship network members. Exemplarily, different importance characterization parameters are pre-configured for different identification correlation information; and, based on the importance characterization parameters, the initial member information feature representation is updated, such as multiplied, to form an updated member information feature representation corresponding to the relationship network member.

[0079] Step S140, determining matching characterization parameters between the key information feature representation and the updated member information feature representation, and performing interval mapping operation on the matching characterization parameters to output information matching parameters between the representative information extraction result and the relationship network members.

[0080] In an embodiment of the present invention, the enterprise financial management platform for financial data extraction can determine the matching characterization parameters between the key information feature representation and the updated member information feature representation (such as determining the corresponding cosine similarity), and perform interval mapping operations on the matching characterization parameters (such as mapping to an interval of 0 to 1, etc.) to output the representative information extraction results and the information matching parameters between the relationship network members.

[0081] Step S150, based on the information matching parameters, at least one first data block identification information is determined from the data block identification information combination; and based on the mapping relationship representation data corresponding to the first data block identification information, at least one first financial data block to be processed is determined from the financial data block cluster to be processed, and the identification reliability parameter corresponding to the first financial data block to be processed is multiplied by the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information.

[0082] In an embodiment of the present invention, the enterprise financial management platform for financial data extraction can determine at least one first data block identification information (such as data block identification information corresponding to an information matching parameter greater than a preset matching parameter) from a combination of data block identification information based on the information matching parameter; and, based on the mapping relationship representation data corresponding to the first data block identification information, determine at least one first financial data block to be processed from a cluster of financial data blocks to be processed, and perform multiplication operation on the identification reliability parameter corresponding to the first financial data block to be processed and the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information. The mapping relationship representation data is used to reflect the mapping information between the hash function value of the financial data block and the data block identification information.

[0083] Step S160: performing a target data extraction operation on the at least one first financial data block to be processed according to the target matching parameter to obtain a corresponding target financial data block.

[0084] In an embodiment of the present invention, the enterprise financial management platform for financial data extraction can perform a target data extraction operation on the at least one first to-be-processed financial data block according to the target matching parameter to obtain a corresponding target financial data block. For example, the first to-be-processed financial data block with the largest target matching parameter can be used as the target financial data block.

[0085] Based on the foregoing, before extracting the financial data block, the table information extraction operation, key information mining operation, and feature representation update operation are performed in sequence, so that the reliability of the obtained target matching parameters is higher, thereby ensuring the reliability of financial data extraction.

[0086] Optionally, in some implementations, the enterprise financial management method for extracting financial data may further include the following contents:

[0087] Determine a cluster of financial data blocks to be processed, and use at least one financial data analysis network to perform multiple levels of financial data analysis operations on the financial data blocks to be processed in the cluster of financial data blocks to be processed, so as to form multiple levels of identification information of each financial data block to be processed and identification reliability parameters corresponding to each identification information of the data block to be processed; illustratively, the financial data analysis operations at multiple levels can be implemented by multiple financial data analysis networks to respectively determine each identification information of the financial data blocks to be processed, and the identification information of the data blocks to be processed can be semantic description data, application scenario data, project type data, abnormality degree data, etc. of the financial data blocks to be processed;

[0088] According to the identification reliability parameter corresponding to the identification information of the data block to be processed, among the plurality of identification information of the data blocks to be processed, determining the screening data block identification information corresponding to the financial data block to be processed, such as screening out one or more with the largest identification reliability parameter;

[0089] Determining, from the configuration data block identification information cluster, configuration data block identification information having correlation with the screening data block identification information as the relevant data block identification information;

[0090] Combining the relevant data block identification information and the screening data block identification information to form a data block identification information combination corresponding to the financial data block cluster to be processed;

[0091] Determine the identification correlation information between the data block identification information from the data block identification information combination, wherein the identification correlation information includes content correlation information (such as similar data, dissimilar data, etc.) and upper-lower level correlation information (such as upper-level overview, lower-level details, etc.);

[0092] Based on the upper-lower level correlation information, with the data block identification information as a member of the relationship network, determining an initial data block identification relationship network;

[0093] Based on the content relevance information, the initial data block identifier relationship network is optimized to form a corresponding optimized data block identifier relationship network.

[0094] Optionally, in some implementations, the step of determining the initial data block identification relationship network based on the upper-lower level correlation information and with the data block identification information as a member of the relationship network may further include the following contents:

[0095] Determine the data block identification information having upper and lower level correlation information from the data block identification information combination to output the corresponding upper and lower level data block identification information combination, wherein the upper and lower level data block identification information combination may include upper level data block identification information and lower level data block identification information;

[0096] Based on the correlation type of the upper and lower level correlation information, clustering the data block identification information in the upper and lower level data block identification information combination;

[0097] According to the obtained clustering processing result, the data block identification information in the upper and lower data block identification information combination is marked as a relationship network member to form an initial data block identification relationship network. In this way, the upper and lower correlation information is included in the initial data block identification relationship network.

[0098] Optionally, in some implementations, the step of optimizing the initial data block identifier relationship network based on the content relevance information to form a corresponding optimized data block identifier relationship network may further include the following contents:

[0099] Determine, from the initial data block identification relationship network, the relationship network members corresponding to the data block identification information having content correlation information, to output at least one relationship network member combination;

[0100] According to the content relevance information corresponding to the network member combination, the initial data block identifier network is optimized to form an intermediate data block identifier network, that is, the content relevance information is added to the initial data block identifier network;

[0101] From the data block identification information combination, data block identification information having content correlation information corresponding to the network members on the intermediate data block identification relationship network is selected to form a corresponding optimized data block identification relationship network.

[0102] Alternatively, in some implementations, the step of selecting data block identification information having content correlation information with data block identification information corresponding to network members on the intermediate data block identification relationship network from the data block identification information combination to form a corresponding optimized data block identification relationship network may further include the following contents:

[0103] Selecting data block identification information having content correlation information with data block identification information corresponding to a network member on the intermediate data block identification relationship network from the data block identification information combination to form a relevant data block identification information combination;

[0104] Based on the content relevance information corresponding to each piece of data block identification information in the combination of related data block identification information, analyzing the relationship network distribution information of the data block identification information, that is, the distribution position, so as to achieve further classification based on the content relevance information;

[0105] According to the relationship network distribution information, the data block identification information in the relevant data block identification information combination is marked as a relationship network member, and is distributed to the intermediate data block identification relationship network to form a corresponding optimized data block identification relationship network.

[0106] Alternatively, in some implementations, the steps of determining a financial data block cluster to be processed, and performing multiple levels of financial data analysis operations on the financial data blocks to be processed in the financial data block cluster to be processed using at least one financial data analysis network to form multiple pieces of identification information of each financial data block to be processed and an identification reliability parameter corresponding to each piece of identification information of the data block to be processed may further include the following contents:

[0107] Determine a financial data block cluster to be processed;

[0108] Using at least one financial data analysis network, a key information mining operation is performed on the unprocessed financial data blocks in the unprocessed financial data block cluster to form a data block key information feature representation corresponding to the unprocessed financial data blocks at the data level corresponding to the financial data analysis network;

[0109] Based on the characteristic representation of the key information of the data block, the identification information of the data block to be processed corresponding to the financial data block to be processed at the data level and the identification reliability parameter corresponding to the identification information of the data block to be processed are analyzed.

[0110] Optionally, in some implementations, after the step of determining the screening data block identification information corresponding to the to-be-processed financial data block from the plurality of to-be-processed data block identification information based on the identification reliability parameter corresponding to the to-be-processed data block identification information, the enterprise financial management method for financial data extraction may further include the following contents:

[0111] A hash function mapping operation is performed on the financial data block to be processed to output a hash function value corresponding to the financial data block to be processed; and based on the hash function value and data block substantial information corresponding to the financial data block to be processed, i.e., data block attribute information, such as the name of the data block, data block representation information corresponding to the financial data block to be processed is determined, and the data block representation information may include the hash function value and the data block substantial information; and, based on the screening data block identification information, the financial data block to be processed is clustered, i.e., classified, and, based on the clustering processing result and the hash function value corresponding to the financial data block to be processed, the mapping relationship representation data between the hash function value and the screening data block identification information is determined.

[0112] Optionally, in some implementations, the step of determining at least one first financial data block to be processed from the financial data block cluster to be processed based on the mapping relationship representation data corresponding to the first data block identification information, and performing a multiplication operation on the identification reliability parameter corresponding to the first financial data block to be processed and the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information may further include the following contents:

[0113] Based on the mapping relationship representation data, determining a target hash function value corresponding to the first data block identification information from the hash function value;

[0114] Determine the target data block representation information corresponding to the target hash function value from the data block representation information; select at least one first financial data block to be processed from the financial data block cluster to be processed corresponding to the data block identification information combination based on the target data block representation information; and perform multiplication operation on the identification reliability parameter corresponding to the first financial data block to be processed and the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information.

[0115] Combination Figure 3 The embodiment of the present invention further provides an enterprise financial management system for extracting financial data, which can be applied to the above-mentioned enterprise financial management platform for extracting financial data. The enterprise financial management system for extracting financial data may include:

[0116] A representative information extraction module is used to perform a representative information extraction operation on the data extraction identification information contained in the financial data extraction indication information when the financial data extraction indication information is obtained, so as to output a representative information extraction result corresponding to the data extraction identification information;

[0117] A key information mining module, used to perform key information mining operations on each network member in the representative information extraction result and the optimized data block identification network, so as to output a key information feature representation corresponding to the representative information extraction result and an initial member information feature representation corresponding to the network member;

[0118] A feature representation updating module is used to analyze the importance representation parameters corresponding to the relationship network members based on the identification correlation information between the data block identification information corresponding to the relationship network members; and, based on the importance representation parameters, update the initial member information feature representation to form an updated member information feature representation corresponding to the relationship network members;

[0119] A matching parameter processing module, used to determine matching characterization parameters between the key information feature representation and the updated member information feature representation, and to perform interval mapping operation on the matching characterization parameters to output information matching parameters between the representative information extraction result and the relationship network members;

[0120] a matching parameter updating module, configured to determine at least one first data block identification information from a combination of data block identification information according to the information matching parameter; and, based on the mapping relationship representation data corresponding to the first data block identification information, determine at least one first financial data block to be processed from a cluster of financial data blocks to be processed, and perform a multiplication operation on an identification reliability parameter corresponding to the first financial data block to be processed and the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information, wherein the mapping relationship representation data is used to reflect the mapping information between the hash function value of the financial data block and the data block identification information;

[0121] The data block extraction module is used to perform a target data extraction operation on the at least one first financial data block to be processed according to the target matching parameter to obtain a corresponding target financial data block.

[0122] Optionally, in some implementations, the enterprise financial management system further includes other functional modules for:

[0123] Determine a cluster of financial data blocks to be processed, and use at least one financial data analysis network to perform multiple levels of financial data analysis operations on the financial data blocks to be processed in the cluster of financial data blocks to be processed, so as to form multiple pieces of identification information of each financial data block to be processed and an identification reliability parameter corresponding to each piece of identification information of the financial data block to be processed;

[0124] According to the identification reliability parameter corresponding to the identification information of the data block to be processed, determining the screening data block identification information corresponding to the financial data block to be processed from the plurality of identification information of the data blocks to be processed;

[0125] Determining, from the configuration data block identification information cluster, configuration data block identification information having correlation with the screening data block identification information as the relevant data block identification information;

[0126] Combining the relevant data block identification information and the screening data block identification information to form a data block identification information combination corresponding to the financial data block cluster to be processed;

[0127] Determining identification correlation information between the data block identification information from the data block identification information combination, wherein the identification correlation information includes content correlation information and upper-lower level correlation information;

[0128] Based on the upper-lower level correlation information, with the data block identification information as a member of the relationship network, determining an initial data block identification relationship network;

[0129] Based on the content relevance information, the initial data block identifier relationship network is optimized to form a corresponding optimized data block identifier relationship network.

[0130] Optionally, in some implementations, the determining of the initial data block identification relationship network based on the upper-lower level correlation information and with the data block identification information as a member of the relationship network includes:

[0131] Determine the data block identification information with upper and lower level correlation information from the data block identification information combination to output the corresponding upper and lower level data block identification information combination;

[0132] Based on the correlation type of the upper and lower level correlation information, clustering the data block identification information in the upper and lower level data block identification information combination;

[0133] According to the obtained clustering processing result, the data block identification information in the combination of upper and lower data block identification information is marked as a relationship network member to form an initial data block identification relationship network.

[0134] In summary, the enterprise financial management method and system for financial data extraction provided by the present invention can first perform a representative information extraction operation on the data extraction identification information in the financial data extraction indication information to output the representative information extraction result corresponding to the data extraction identification information; perform a key information mining operation on each network member in the representative information extraction result and the optimized data block identification network to output the key information feature representation corresponding to the representative information extraction result and the initial member information feature representation corresponding to the network member; based on the identification correlation information between the data block identification information corresponding to the network member, analyze the importance representation parameters corresponding to the network member; and, based on the importance representation parameters, update the initial member information feature representation to form an updated member information feature representation corresponding to the network member; determine the key information The matching characterization parameters between the feature representation and the updated member information feature representation are performed, and the matching characterization parameters are interval mapped to output the information matching parameters between the representative information extraction result and the relationship network members; at least one first data block identification information is determined from the data block identification information combination according to the information matching parameters; and at least one first financial data block to be processed is determined from the financial data block cluster to be processed based on the mapping relationship characterization data corresponding to the first data block identification information, and the identification reliability parameter corresponding to the first financial data block to be processed is multiplied by the information matching parameter to form the target matching parameter between the first financial data block to be processed and the data extraction identification information; according to the target matching parameter, at least one first financial data block to be processed is subjected to the target data extraction operation to obtain the corresponding target financial data block. Based on the above content, before the financial data block is extracted, the reliability of the target matching parameters obtained is higher due to the sequential operation of table information extraction, key information mining and feature representation update, thereby ensuring the reliability of financial data extraction.

[0135] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An enterprise financial management method for extracting financial data, characterized in that: include: When the financial data extraction instruction information is obtained, performing a representative information extraction operation on the data extraction identification information in the financial data extraction instruction information to output a representative information extraction result corresponding to the data extraction identification information; Performing key information mining operations on each network member in the representative information extraction result and the optimized data block identification network to output key information feature representations corresponding to the representative information extraction result and initial member information feature representations corresponding to the network members; Analyzing the importance characterization parameters corresponding to the network members based on the identification correlation information between the data block identification information corresponding to the network members; and updating the initial member information feature representation according to the importance characterization parameters to form an updated member information feature representation corresponding to the network members; Determine matching characterization parameters between the key information feature representation and the updated member information feature representation, and perform interval mapping operation on the matching characterization parameters to output information matching parameters between the representative information extraction result and the relationship network members; According to the information matching parameter, at least one first data block identification information is determined from the data block identification information combination; and, based on the mapping relationship representation data corresponding to the first data block identification information, at least one first financial data block to be processed is determined from the financial data block cluster to be processed, and the identification reliability parameter corresponding to the first financial data block to be processed is multiplied by the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information, wherein the mapping relationship representation data is used to reflect the mapping information between the hash function value of the financial data block and the data block identification information; According to the target matching parameter, performing a target data extraction operation on the at least one first financial data block to be processed to obtain a corresponding target financial data block; The enterprise financial management method further includes: Determine a cluster of financial data blocks to be processed, and use at least one financial data analysis network to perform multiple levels of financial data analysis operations on the financial data blocks to be processed in the cluster of financial data blocks to be processed, so as to form multiple pieces of identification information of each financial data block to be processed and an identification reliability parameter corresponding to each piece of identification information of the financial data block to be processed; According to the identification reliability parameter corresponding to the identification information of the data block to be processed, determining the screening data block identification information corresponding to the financial data block to be processed from the plurality of identification information of the data blocks to be processed; Determining, from the configuration data block identification information cluster, configuration data block identification information having correlation with the screening data block identification information as the relevant data block identification information; Combining the relevant data block identification information and the screening data block identification information to form a data block identification information combination corresponding to the financial data block cluster to be processed; Determining identification correlation information between the data block identification information from the data block identification information combination, wherein the identification correlation information includes content correlation information and upper-lower level correlation information; Based on the upper-lower level correlation information, with the data block identification information as a member of the relationship network, determining an initial data block identification relationship network; Based on the content relevance information, optimizing the initial data block identifier relationship network to form a corresponding optimized data block identifier relationship network; The step of optimizing the initial data block identifier relationship network based on the content relevance information to form a corresponding optimized data block identifier relationship network includes: Determine, from the initial data block identification relationship network, the relationship network members corresponding to the data block identification information having content correlation information, to output at least one relationship network member combination; According to the content relevance information corresponding to the relationship network member combination, the initial data block identifier relationship network is optimized to form an intermediate data block identifier relationship network; Selecting data block identification information having content correlation information with data block identification information corresponding to network members on the intermediate data block identification relationship network from the data block identification information combination to form a corresponding optimized data block identification relationship network; The step of selecting data block identification information having content correlation information with data block identification information corresponding to network members on the intermediate data block identification relationship network from the data block identification information combination to form a corresponding optimized data block identification relationship network comprises: Selecting data block identification information having content correlation information with data block identification information corresponding to a network member on the intermediate data block identification relationship network from the data block identification information combination to form a relevant data block identification information combination; Analyzing the relationship network distribution information of the data block identification information based on the content correlation information corresponding to each data block identification information in the combination of related data block identification information; According to the relationship network distribution information, the data block identification information in the combination of related data block identification information is marked as a relationship network member, and is distributed to the intermediate data block identification relationship network to form a corresponding optimized data block identification relationship network; After the step of determining the screening data block identification information corresponding to the to-be-processed financial data block from the plurality of to-be-processed data block identification information based on the identification reliability parameter corresponding to the to-be-processed data block identification information, the enterprise financial management method further comprises: A hash function mapping operation is performed on the financial data block to be processed to output a hash function value corresponding to the financial data block to be processed; and based on the hash function value and the data block substantial information corresponding to the financial data block to be processed, data block representation information corresponding to the financial data block to be processed is determined; and, based on the screening data block identification information, the financial data block to be processed is clustered, and, based on the obtained clustering result and the hash function value corresponding to the financial data block to be processed, a mapping relationship representation data between the hash function value and the screening data block identification information is determined; The step of determining at least one first financial data block to be processed from the cluster of financial data blocks to be processed based on the mapping relationship representation data corresponding to the first data block identification information, and performing a multiplication operation on the identification reliability parameter corresponding to the first financial data block to be processed and the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information comprises: Based on the mapping relationship representation data, determine the target hash function value corresponding to the first data block identification information from the hash function value; determine the target data block representation information corresponding to the target hash function value from the data block representation information; based on the target data block representation information, select at least one first financial data block to be processed from the financial data block cluster to be processed corresponding to the data block identification information combination; and multiply the identification reliability parameter corresponding to the first financial data block to be processed by the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information.

2. The enterprise financial management method for extracting financial data according to claim 1, characterized in that: The step of determining an initial data block identification relationship network based on the upper-lower level correlation information and when the data block identification information is used as a member of the relationship network includes: Determine the data block identification information with upper and lower level correlation information from the data block identification information combination to output the corresponding upper and lower level data block identification information combination; Based on the correlation type of the upper and lower level correlation information, clustering the data block identification information in the upper and lower level data block identification information combination; According to the obtained clustering processing result, the data block identification information in the combination of upper and lower data block identification information is marked as a relationship network member to form an initial data block identification relationship network.

3. The enterprise financial management method for extracting financial data according to claim 1, characterized in that: The steps of determining a financial data block cluster to be processed, and using at least one financial data analysis network to perform multiple levels of financial data analysis operations on the financial data blocks to be processed in the financial data block cluster to be processed, so as to form multiple pieces of identification information of each financial data block to be processed and an identification reliability parameter corresponding to each piece of identification information of the financial data block to be processed, include: Determine a financial data block cluster to be processed; Using at least one financial data analysis network, a key information mining operation is performed on the unprocessed financial data blocks in the unprocessed financial data block cluster to form a data block key information feature representation corresponding to the unprocessed financial data blocks at the data level corresponding to the financial data analysis network; Based on the characteristic representation of the key information of the data block, the identification information of the data block to be processed corresponding to the financial data block to be processed at the data level and the identification reliability parameter corresponding to the identification information of the data block to be processed are analyzed.

4. An enterprise financial management system for extracting financial data, characterized in that: include: A representative information extraction module is used to perform a representative information extraction operation on the data extraction identification information contained in the financial data extraction indication information when the financial data extraction indication information is obtained, so as to output a representative information extraction result corresponding to the data extraction identification information; A key information mining module, used to perform key information mining operations on each network member in the representative information extraction result and the optimized data block identification network, so as to output a key information feature representation corresponding to the representative information extraction result and an initial member information feature representation corresponding to the network member; A feature representation updating module is used to analyze the importance representation parameters corresponding to the relationship network members based on the identification correlation information between the data block identification information corresponding to the relationship network members; and, based on the importance representation parameters, update the initial member information feature representation to form an updated member information feature representation corresponding to the relationship network members; A matching parameter processing module, used to determine matching characterization parameters between the key information feature representation and the updated member information feature representation, and to perform interval mapping operation on the matching characterization parameters to output information matching parameters between the representative information extraction result and the relationship network members; a matching parameter updating module, configured to determine at least one first data block identification information from a combination of data block identification information according to the information matching parameter; and, based on the mapping relationship representation data corresponding to the first data block identification information, determine at least one first financial data block to be processed from a cluster of financial data blocks to be processed, and perform a multiplication operation on an identification reliability parameter corresponding to the first financial data block to be processed and the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information, wherein the mapping relationship representation data is used to reflect the mapping information between the hash function value of the financial data block and the data block identification information; A data block extraction module, configured to perform a target data extraction operation on the at least one first financial data block to be processed according to the target matching parameter, so as to obtain a corresponding target financial data block; The enterprise financial management system also includes other functional modules for: Determine a cluster of financial data blocks to be processed, and use at least one financial data analysis network to perform multiple levels of financial data analysis operations on the financial data blocks to be processed in the cluster of financial data blocks to be processed, so as to form multiple pieces of identification information of each financial data block to be processed and an identification reliability parameter corresponding to each piece of identification information of the financial data block to be processed; According to the identification reliability parameter corresponding to the identification information of the data block to be processed, determining the screening data block identification information corresponding to the financial data block to be processed from the plurality of identification information of the data blocks to be processed; Determining, from the configuration data block identification information cluster, configuration data block identification information having correlation with the screening data block identification information as the relevant data block identification information; Combining the relevant data block identification information and the screening data block identification information to form a data block identification information combination corresponding to the financial data block cluster to be processed; Determining identification correlation information between the data block identification information from the data block identification information combination, wherein the identification correlation information includes content correlation information and upper-lower level correlation information; Based on the upper-lower level correlation information, with the data block identification information as a member of the relationship network, determining an initial data block identification relationship network; Based on the content relevance information, optimizing the initial data block identifier relationship network to form a corresponding optimized data block identifier relationship network; The optimizing operation of the initial data block identifier relationship network based on the content relevance information to form a corresponding optimized data block identifier relationship network includes: Determine, from the initial data block identification relationship network, the relationship network members corresponding to the data block identification information having content correlation information, to output at least one relationship network member combination; According to the content relevance information corresponding to the relationship network member combination, the initial data block identifier relationship network is optimized to form an intermediate data block identifier relationship network; Selecting data block identification information having content correlation information with data block identification information corresponding to network members on the intermediate data block identification relationship network from the data block identification information combination to form a corresponding optimized data block identification relationship network; The step of selecting data block identification information having content correlation information with data block identification information corresponding to network members on the intermediate data block identification relationship network from the data block identification information combination to form a corresponding optimized data block identification relationship network comprises: Selecting data block identification information having content correlation information with data block identification information corresponding to a network member on the intermediate data block identification relationship network from the data block identification information combination to form a relevant data block identification information combination; Analyzing the relationship network distribution information of the data block identification information based on the content correlation information corresponding to each data block identification information in the combination of related data block identification information; According to the relationship network distribution information, the data block identification information in the combination of related data block identification information is marked as a relationship network member, and is distributed to the intermediate data block identification relationship network to form a corresponding optimized data block identification relationship network; After determining the screening data block identification information corresponding to the financial data block to be processed from the plurality of data block identification information to be processed according to the identification reliability parameter corresponding to the data block identification information to be processed, the financial data block to be processed is subjected to a hash function mapping operation to output a hash function value corresponding to the financial data block to be processed; and, based on the hash function value and data block substantial information corresponding to the financial data block to be processed, data block representation information corresponding to the financial data block to be processed is determined; and, based on the screening data block identification information, the financial data block to be processed is subjected to clustering processing, and, based on the obtained clustering processing result and the hash function value corresponding to the financial data block to be processed, the mapping relationship representation data between the hash function value and the screening data block identification information is determined; The method further comprises: determining at least one first financial data block to be processed from the cluster of financial data blocks to be processed based on the mapping relationship representation data corresponding to the first data block identification information, and performing a multiplication operation on the identification reliability parameter corresponding to the first financial data block to be processed and the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information. Based on the mapping relationship representation data, determine the target hash function value corresponding to the first data block identification information from the hash function value; determine the target data block representation information corresponding to the target hash function value from the data block representation information; based on the target data block representation information, select at least one first financial data block to be processed from the financial data block cluster to be processed corresponding to the data block identification information combination; and multiply the identification reliability parameter corresponding to the first financial data block to be processed by the information matching parameter to form a target matching parameter between the first financial data block to be processed and the data extraction identification information.

5. The enterprise financial management system for extracting financial data according to claim 4, characterized in that: The determining of an initial data block identification relationship network based on the upper-lower level correlation information and when the data block identification information is used as a member of the relationship network includes: Determine the data block identification information with upper and lower level correlation information from the data block identification information combination to output the corresponding upper and lower level data block identification information combination; Based on the correlation type of the upper and lower level correlation information, clustering the data block identification information in the upper and lower level data block identification information combination; According to the obtained clustering processing result, the data block identification information in the combination of upper and lower data block identification information is marked as a relationship network member to form an initial data block identification relationship network.

Images