Electric power financial data processing method and system

Abstract

The invention discloses an electric power financial data processing method and device, relates to the technical field of data processing, and mainly aims to solve the problem of poor processing efficiency of existing electric power financial data. Comprising the steps of obtaining financial keywords and financial data in response to a financial data processing request; calling financial processing nodes based on the financial keywords, and performing examination and approval evaluation on the financial data based on a large language model corresponding to the financial processing nodes to obtain an examination and approval evaluation result; when the examination and approval evaluation result is an abnormal evaluation result, sending the financial data to an examination and approval node; and when the examination and approval evaluation result is a normal evaluation result, calling a financial component to process the financial data.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and apparatus for processing power financial data. Background Technology

[0002] With the advent of the big data era, the power system is gradually introducing big data processing technology to process various business data. For example, in the power grid financial operation system, big data processing technology is used to replace manual processing in order to improve the accuracy and efficiency of financial data.

[0003] Currently, the processing of financial data in existing power grid systems typically requires multi-process and multi-dimensional manual approval. This greatly increases the complexity of the financial data processing flow and labor costs, thereby reducing the efficiency of power financial data processing and introducing potential errors. Summary of the Invention

[0004] In view of this, this application provides a method and apparatus for processing power financial data, the main purpose of which is to solve the problem of poor processing efficiency of existing power financial data.

[0005] According to one aspect of this application, a method for processing power financial data is provided, comprising: In response to a financial data processing request, retrieve financial keywords and financial data; Based on the financial keywords, the financial processing nodes are retrieved, and the financial data is approved and evaluated based on the large language model corresponding to the financial processing nodes to obtain the approval and evaluation results. If the approval assessment result is an abnormal assessment result, the financial data will be sent to the approval node; If the approval assessment result is a normal assessment result, the financial component will be retrieved to process the financial data.

[0006] Furthermore, before obtaining the approval evaluation result by performing an approval evaluation on the financial data based on the large language model corresponding to the financial processing node, the method further includes: Obtain language training samples and construct a large language model, wherein the language training samples include financial text samples carrying approval tags; The large language model is trained based on the language training samples to obtain a large language model that has completed training.

[0007] Furthermore, the method also includes: The financial keywords in the language training samples are parsed, and the financial keywords and the large language model trained using the financial keywords are bound to the constructed financial nodes.

[0008] Furthermore, after sending the financial data to the approval node, the method further includes: Effective financial information is extracted from the financial data based on an information extraction model; Based on the aforementioned valid financial information, an approval report containing the approval assessment results, approval time, and approval personnel is generated according to a preset time forecast, and the approval report is submitted.

[0009] According to another aspect of this application, a power financial data processing system is provided, comprising: The retrieval module is used to respond to financial data processing requests and retrieve financial keywords and financial data; The question-and-answer module is used to retrieve the financial processing node based on the financial keywords, and to evaluate the financial data based on the large language model corresponding to the financial processing node to obtain the evaluation result. If the evaluation result is an abnormal evaluation result, the financial data is sent to the approval node. If the evaluation result is a normal evaluation result, the financial component is retrieved to process the financial data.

[0010] Furthermore, the question-answering module is also used to acquire language training samples and construct a large language model, wherein the language training samples include financial text samples carrying approval tags; and to train the large language model based on the language training samples to obtain a large language model that has completed model training.

[0011] Furthermore, the question-answering module is also used to parse the financial keywords in the language training samples and bind the financial keywords and the large language model trained using the financial keywords to the constructed financial nodes.

[0012] Furthermore, the system also includes: An extraction module is used to extract valid financial information from the financial data based on an information extraction model; The report generation module is used to combine the aforementioned valid financial information to generate an approval report containing the approval evaluation results, approval time, and approver information according to a preset time prediction, and then submit the approval report to the higher authorities.

[0013] According to another aspect of this application, a storage medium is provided that stores at least one executable instruction, which causes a processor to perform operations corresponding to the above-described method for processing power financial data.

[0014] According to another aspect of this application, a terminal is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the above-described method for processing power financial data.

[0015] By employing the above technical solutions, the technical solutions provided in the embodiments of this application have at least the following advantages: This application provides a method and system for processing power financial data. Compared with the prior art, the embodiments of this application obtain financial keywords and financial data in response to a financial data processing request; based on the financial keywords, a financial processing node is invoked, and the financial data is approved and evaluated based on the large language model corresponding to the financial processing node to obtain an approval and evaluation result; when the approval and evaluation result is an abnormal evaluation result, the financial data is sent to the approval node; when the approval and evaluation result is a normal evaluation result, the financial component is invoked to process the financial data, thereby achieving the purpose of effectively processing power financial data and greatly improving the accuracy of power financial data processing.

[0016] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating a method for processing power financial data according to an embodiment of this application is shown; Figure 2 This invention provides a block diagram of a power financial data processing apparatus according to an embodiment of the present application. Figure 3 A schematic diagram of the structure of a terminal provided in an embodiment of this application is shown. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] This application provides a method for processing power financial data, such as... Figure 1 As shown, the method includes: 101. In response to a financial data processing request, obtain financial keywords and financial data.

[0021] In this embodiment, the power financial data processing method, as the current executing entity, can be a server or a terminal device. In this case, the user triggers a financial data processing request to instruct the current executing entity to obtain financial keywords and financial data. Financial keywords are used to characterize the main terms required in the financial processing process, such as settlement and reimbursement. Financial data refers to all the data required to generate financial changes and can be directly retrieved from the power system; this embodiment does not impose specific limitations on this.

[0022] 102. Retrieve the financial processing node based on the financial keywords, and conduct an approval evaluation on the financial data based on the large language model corresponding to the financial processing node to obtain the approval evaluation result.

[0023] In this embodiment, the current execution end retrieves the financial processing node based on financial keywords. Here, the financial processing node represents the node that performs financial processing through the financial operations system and can be pre-configured; this embodiment does not impose specific limitations. Furthermore, different financial nodes are pre-bound with different Large Language Models (LLMs) to evaluate and approve financial data, obtaining the evaluation results. The Large Language Model (LLM) is a deep learning-based natural language processing model trained on massive amounts of text data, possessing the ability to understand, generate, and reason about text.

[0024] 103a. When the approval assessment result is an abnormal assessment result, the financial data is sent to the approval node.

[0025] In this embodiment of the application, after obtaining the approval evaluation result, if the approval evaluation result is an abnormal evaluation result, it means that online approval cannot be carried out. Therefore, the financial data can be sent to the approval node for manual approval.

[0026] 103b. If the approval assessment result is a normal assessment result, the financial component is retrieved to process the financial data.

[0027] In this embodiment, after obtaining the approval evaluation result, if the evaluation result is normal, it indicates that online approval can be carried out. Therefore, the financial component can be invoked to process the financial data online. The financial component can be a pre-set component with calculation functions, embedding different financial processing algorithms; this embodiment does not impose specific limitations on it.

[0028] In another embodiment of this application, for further definition and explanation, before the step of approving and evaluating the financial data based on the large language model corresponding to the financial processing node and obtaining the approval and evaluation result, the method further includes: Obtain language training samples and build a large language model; The large language model is trained based on the language training samples to obtain a large language model that has completed training.

[0029] To achieve intelligent approval and evaluation, the current execution end first acquires language training samples, which can be historical data recorded in the power financial operation system. These language training samples include financial text samples carrying approval tags. Simultaneously, a large language model (LLM) is constructed. A large language model (LLM) is a natural language processing technique based on deep learning. Trained on massive amounts of text data, it possesses the ability to understand, generate, and reason about text. During training, an unsupervised learning phase can be employed, learning language patterns by predicting words or masked words in the text. Simultaneously, a supervised learning phase uses labeled data to optimize performance for specific tasks. This embodiment of the application does not impose specific limitations.

[0030] In another embodiment of this application, for further definition and explanation, the steps also include: The financial keywords in the language training samples are parsed, and the financial keywords and the large language model trained using the financial keywords are bound to the constructed financial nodes.

[0031] To improve the processing accuracy of power financial data by matching different large language models to different financial nodes, the current execution end first parses the financial keywords in the language training samples. Alternatively, it can perform pre-training based on natural language recognition algorithms before parsing to determine the financial keywords. Then, the financial keywords and the large language model trained using these financial keywords are bound to the constructed financial nodes.

[0032] In another embodiment of this application, for further definition and explanation, after the step of sending the financial data to the approval node, the method further includes: Effective financial information is extracted from the financial data based on an information extraction model; Based on the aforementioned valid financial information, an approval report containing the approval assessment results, approval time, and approval personnel is generated according to a preset time forecast, and the approval report is submitted.

[0033] To achieve accurate financial data reporting and avoid invalid reporting, the current execution end uses an information extraction model to extract valid financial information from the financial data. Information extraction (IE) is one of the core tasks of Natural Language Processing (NLP), aiming to automatically identify, extract, and organize structured information from unstructured or semi-structured text. Information extraction mainly includes three sub-tasks: Named Entity Recognition (NER): identifying and classifying entities (such as names, locations, and organizations) in the text; Relation Extraction (RE): extracting semantic relationships between entities (such as "employed" and "located in"); and Event Extraction (EE): identifying events and their elements (such as trigger words and arguments). In this embodiment, the preferred information extraction model is PP-UIE (PaddlePaddle Universal Information Extraction) to uniformly handle multiple information extraction tasks (such as named entity recognition, relation extraction, and event extraction). Furthermore, by combining effective financial information, an approval report containing the approval evaluation results, approval time, and approval personnel is generated according to a preset time forecast, and the approval report is submitted. For example, the approval report is generated and submitted once a week or once a month. This application embodiment does not make specific limitations.

[0034] This application provides a method for processing power financial data. Compared with the prior art, the embodiments of this application, in response to a financial data processing request, obtain financial keywords and financial data; based on the financial keywords, call a financial processing node, and perform approval and evaluation on the financial data based on the large language model corresponding to the financial processing node to obtain an approval and evaluation result; when the approval and evaluation result is an abnormal evaluation result, the financial data is sent to the approval node; when the approval and evaluation result is a normal evaluation result, the financial component is called to process the financial data, thereby achieving the purpose of effectively processing power financial data and greatly improving the accuracy of power financial data processing.

[0035] Furthermore, as a response to the above Figure 1 The implementation of the method shown in this application provides a power financial data processing system, such as... Figure 2 As shown, the system includes: The retrieval module 21 is used to retrieve financial keywords and financial data in response to financial data processing requests; The question-and-answer module 22 is used to retrieve the financial processing node based on the financial keywords, and to evaluate the financial data based on the large language model corresponding to the financial processing node to obtain the evaluation result; when the evaluation result is an abnormal evaluation result, the financial data is sent to the approval node; when the evaluation result is a normal evaluation result, the financial component is retrieved to process the financial data.

[0036] Furthermore, the question-answering module is also used to acquire language training samples and construct a large language model, wherein the language training samples include financial text samples carrying approval tags; and to train the large language model based on the language training samples to obtain a large language model that has completed model training.

[0037] Furthermore, the question-answering module is also used to parse the financial keywords in the language training samples and bind the financial keywords and the large language model trained using the financial keywords to the constructed financial nodes.

[0038] Furthermore, the system also includes: An extraction module is used to extract valid financial information from the financial data based on an information extraction model; The report generation module is used to combine the aforementioned valid financial information to generate an approval report containing the approval evaluation results, approval time, and approver information according to a preset time prediction, and then submit the approval report to the higher authorities.

[0039] This application provides a power financial data processing system. Compared with the prior art, the embodiments of this application, in response to a financial data processing request, obtain financial keywords and financial data; based on the financial keywords, call the financial processing node, and perform approval and evaluation on the financial data based on the large language model corresponding to the financial processing node to obtain an approval and evaluation result; when the approval and evaluation result is an abnormal evaluation result, the financial data is sent to the approval node; when the approval and evaluation result is a normal evaluation result, the financial component is called to process the financial data, thereby achieving the purpose of effectively processing power financial data and greatly improving the accuracy of power financial data processing.

[0040] According to one embodiment of this application, a storage medium is provided, the storage medium storing at least one executable instruction, the computer-executable instruction being able to execute the power financial data processing method in any of the above method embodiments.

[0041] Figure 3 The diagram shows a structural schematic of a terminal according to one embodiment of the present application. The specific embodiments of the present application do not limit the specific implementation of the terminal.

[0042] like Figure 3 As shown, the terminal may include: a processor 302, a communications interface 304, a memory 306, and a communications bus 308.

[0043] The processor 302, communication interface 304, and memory 306 communicate with each other via communication bus 308.

[0044] Communication interface 304 is used to communicate with other network elements such as clients or other servers.

[0045] The processor 302 is used to execute program 310, specifically to execute the relevant steps in the above embodiment of the method for processing electricity financial data.

[0046] Specifically, program 310 may include program code that includes computer operation instructions.

[0047] Processor 302 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. The terminal includes one or more processors, which may be processors of the same type, such as one or more CPUs; or they may be processors of different types, such as one or more CPUs and one or more ASICs.

[0048] Memory 306 is used to store program 310. Memory 306 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0049] Specifically, program 310 can be used to cause processor 302 to perform the following operations: In response to a financial data processing request, retrieve financial keywords and financial data; Based on the financial keywords, the financial processing nodes are retrieved, and the financial data is approved and evaluated based on the large language model corresponding to the financial processing nodes to obtain the approval and evaluation results. If the approval assessment result is an abnormal assessment result, the financial data will be sent to the approval node; If the approval assessment result is a normal assessment result, the financial component will be retrieved to process the financial data.

[0050] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0051] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method of processing electric power financial data, characterized by, The method comprises the following steps: in response to a financial data processing request, obtaining a financial keyword and financial data; based on the financial keyword, calling a financial processing node, and based on a large language model corresponding to the financial processing node, performing approval evaluation on the financial data to obtain an approval evaluation result; when the approval evaluation result is an abnormal evaluation result, the financial data is sent to an approval node; when the approval evaluation result is a normal evaluation result, the financial data is processed by a financial component.

2. The method of claim 1, wherein, Before the step of based on the large language model corresponding to the financial processing node, performing approval evaluation on the financial data to obtain an approval evaluation result, the method further comprises the following steps: obtain a language training sample and construct a large language model, wherein the language training sample comprises a financial text sample carrying an approval label; based on the language training sample, model training is performed on the large language model to obtain a large language model after model training.

3. The method of claim 2, wherein, The method further comprises the following steps: analyze the financial keyword in the language training sample, and bind the financial keyword and the large language model trained by the financial keyword to the constructed financial node.

4. The method of claim 3, wherein, After the step of sending the financial data to the approval node, the method further comprises the following steps: based on an information extraction model, extract financial effective information from the financial data; combined with the financial effective information, generate an approval report containing an approval evaluation result, an approval time and an approval personnel according to a preset time prediction, and report the approval report.

5. A system for processing electric power financial data, characterized by comprising: The system comprises: a retrieval module, configured to obtain a financial keyword and financial data in response to a financial data processing request; a question and answer module, configured to call a financial processing node based on the financial keyword, and perform approval evaluation on the financial data based on a large language model corresponding to the financial processing node to obtain an approval evaluation result; when the approval evaluation result is an abnormal evaluation result, the financial data is sent to an approval node; when the approval evaluation result is a normal evaluation result, the financial data is processed by a financial component.

6. The system of claim 5, wherein the question and answer module is further configured to obtain a language training sample and construct a large language model, wherein the language training sample comprises a financial text sample carrying an approval label; based on the language training sample, model training is performed on the large language model to obtain a large language model after model training.

7. The system of claim 6, wherein the question and answer module is further configured to analyze the financial keyword in the language training sample, and bind the financial keyword and the large language model trained by the financial keyword to the constructed financial node.

8. The system of claim 7, wherein, The system further comprises: an extraction module, configured to extract financial effective information from the financial data based on an information extraction model; a report generation module, configured to generate an approval report containing an approval evaluation result, an approval time and an approval personnel according to a preset time prediction in combination with the financial effective information, and report the approval report.

9. A computer readable storage medium having stored thereon computer programs / instructions, characterized in that, The computer program / instructions are executed by the processor to realize the steps of the method of claim 1.

10. A computer device comprising a memory, a processor, and a computer program stored on the memory, wherein the computer program comprises instructions that, when executed by the processor, cause the processor to perform the method of any one of claims 1-9. The processor executes the computer program to implement the steps of the method of claim 1.

Images