A document processing method, system, device and product
By using parameter text processing agents and fault text processing agents to perform structured processing on parameter data and fault description data in documents, the problem of difficult and efficient conversion of complex data type documents in traditional methods is solved, and efficient and reliable structured processing is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SUNGROW (SHANGHAI) CO LTD
- Filing Date
- 2024-12-10
- Publication Date
- 2026-06-12
AI Technical Summary
Traditional document processing methods struggle to quickly and effectively transform documents containing complex data types into structured documents, especially when dealing with product parameters and semi-structured table information, resulting in poor efficiency and effectiveness.
A parameter text processing agent and a fault text processing agent are used to perform structured processing on the parameter data and fault description data in the text to be processed, respectively. Through document parsing, text recognition, data extraction and transformation, structured parameter data and structured fault description data are generated.
It improves the efficiency and quality of structured processing of documents containing complex data types, ensures the consistency and reliability of data processing, and has good scalability and flexibility.
Smart Images
Figure CN122197824A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data processing technology, and in particular to a document processing method, system, device and product. Background Technology
[0002] The research, development, operation and maintenance of new energy products have generated a large number of documents, which contain important information such as product design parameters and troubleshooting methods.
[0003] Traditional document processing methods mainly include manual processing, deep learning algorithms, and document parsing tools. These methods are generally suitable for simple documents with uniform formatting. However, when faced with documents containing complex data types such as product parameters and semi-structured table information, their efficiency and effectiveness are often insufficient, making it difficult to achieve fast and effective structured conversion. Summary of the Invention
[0004] This application provides a document processing method, system, device, and product to solve the problem that traditional document processing methods struggle to achieve fast and effective structured conversion when dealing with documents containing complex data types.
[0005] In a first aspect, embodiments of this application provide a document processing method, including:
[0006] The document to be processed is parsed to obtain the text to be processed;
[0007] The parameter text processing agent is invoked to perform structured processing on the parameter data in the text to be processed, thereby obtaining structured parameter data;
[0008] The fault text processing agent is invoked to perform structured processing on the fault description data in the text to be processed, thereby obtaining structured fault description data.
[0009] Secondly, embodiments of this application provide a document processing system, including:
[0010] The document parsing module is used to parse the document to be processed and obtain the text to be processed.
[0011] The parameter structuring module is used to call the parameter text processing agent to perform structuring processing on the parameter data in the text to be processed, and obtain structured parameter data;
[0012] The fault structuring module is used to call the fault text processing agent to perform structuring processing on the fault description data in the text to be processed, and obtain structured fault description data.
[0013] Thirdly, embodiments of this application provide an electronic device, the electronic device comprising:
[0014] At least one processor; and
[0015] A memory communicatively connected to the at least one processor; wherein,
[0016] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the document processing method described in any embodiment of this application.
[0017] Fourthly, embodiments of this application provide a computer program product including a computer program, which, when executed by a processor, implements the document processing method described in any embodiment of this application.
[0018] The technical solution of this application embodiment involves parsing the document to be processed to obtain the text to be processed; calling a parameter text processing agent to perform structured processing on the parameter text in the document to be processed to obtain structured parameter data; and calling a fault text processing agent to perform structured processing on the fault description text in the document to be processed to obtain structured fault description data. By utilizing specially designed agents to perform structured processing on the parameter text and fault description text in the document to be processed separately, a highly automated and intelligent processing flow is achieved. This solves the problem that traditional document processing methods struggle to achieve fast and effective structured conversion when dealing with documents containing complex data types, thus improving the processing efficiency and quality of structured processing for documents containing complex data types.
[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description
[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0021] Figure 1 A flowchart illustrating a document processing method provided in Embodiment 1 of this application;
[0022] Figure 2 This is a schematic diagram of the structure of a document processing system provided in Embodiment 1 of this application;
[0023] Figure 3This is a schematic diagram of the structure of a document processing system provided in Embodiment 2 of this application;
[0024] Figure 4 A schematic diagram of the structure of an electronic device for implementing the document processing method of this application embodiment. Detailed Implementation
[0025] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.
[0026] 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.
[0027] Example 1
[0028] Figure 1 This is a flowchart illustrating a document processing method provided in Embodiment 1 of this application. This embodiment is applicable to the structured processing of documents containing complex data types. The method can be executed by a document processing system, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:
[0029] S110. Parse the document to be processed to obtain the text to be processed.
[0030] In this context, "documents to be processed" can be understood as documents that require structured processing. These documents can contain complex data types, such as fault case analysis reports, product operation manuals, and user manuals. Such documents typically include product parameters and semi-structured table information. Supported document formats for documents to be processed include, but are not limited to, PDF, Word, Excel, PPT, TXT, or images. "Text to be processed" can be understood as text that requires structured processing.
[0031] The text to be processed can specifically include fault description information and parameter information. Fault description information is typically used to describe various fault phenomena encountered by the product in actual use, and may contain a certain degree of ambiguity, uncertainty, and subjectivity. Fault description information can include key information such as fault definition data, fault performance data, fault diagnosis data, and fault handling data. Common fault description information can include anomalies, faults, failures, damage, inability, fault codes, errors, inability to start, and system crashes. Parameter information typically refers to the parameters of the product in its running or dormant state, and may include specific values, specifications, performance indicators, etc., related to the product. It has standardized and normalized characteristics and is usually a quantitative indicator. Common parameter information can include: voltage, power, capacity, frequency, input, output, size, weight, speed, and material.
[0032] Fault description information and parameter information are quite distinct. The former focuses on describing problems, phenomena, and faults, while the latter focuses on providing quantitative specifications and technical indicators of the product. Based on their respective characteristics, fault description information and parameter information in the document to be processed can be effectively identified.
[0033] For example, the process of parsing a document to be processed to obtain the text to be processed may include: when parsing the document to be processed, text recognition may be performed on the document to be processed to obtain the document text content, and the document text content may be segmented according to the document title to obtain multiple document text segments, and each document text segment may be parsed to obtain the text to be processed.
[0034] For example, methods for parsing the document to be processed can include using document parsing algorithms, document parsing models, or document parsing tools to parse the document to be processed.
[0035] S120. Call the parameter text processing agent to perform structured processing on the parameter data in the text to be processed, and obtain structured parameter data.
[0036] In this context, parameter data can be understood as data that describes the state of a control device or system based on numerical values. Parameter data can be assigned constant values or other variable values. Structured parameter data can be understood as parameter data stored in a clearly defined format and according to specific rules.
[0037] A Parameter Processing Agent (PP-Agent) can be understood as an agent specifically designed for structuring parameter text. It can identify, extract, process, and transform data within parameter text, converting it from unstructured or semi-structured to structured data for further analysis and application. In this embodiment, the PP-Agent can be a single agent or a group of agents collaborating to process complex content. The agent possesses a degree of autonomy, enabling it to perform tasks independently; it can integrate various tools and plugins, such as calculators and code interpreters, to expand its processing capabilities; and it also possesses powerful Natural Language Processing (NLP) capabilities, allowing it to understand and generate natural language, facilitating interaction with human users and other systems.
[0038] Specifically, the parameter text processing agent is invoked to perform text recognition and extraction on the data to be processed to obtain parameter data, and then performs data transformation on the parameter data to obtain structured parameter data. The structured parameter data can be in the form of JSON, XML, or database tables for easy storage and analysis.
[0039] S130. Call the fault text processing agent to perform structured processing on the fault description data in the text to be processed, and obtain structured fault description data.
[0040] Fault description data can be understood as data used to describe the fault status and fault diagnosis information of a device or system. Structured fault description data can be understood as fault description data stored in a clearly defined format and according to specific rules.
[0041] A Fault Processing Agent (FP-Agent) can be understood as an intelligent agent specifically designed to perform structured processing of fault description text. It can identify, extract, process, and transform data within the fault description text, converting it from unstructured or semi-structured form into structured data for further analysis and application. In this embodiment, the fault processing agent can be a single agent or an intelligent system in which multiple agents collaborate to process complex content.
[0042] Specifically, the fault text processing agent is invoked to perform text recognition and parameter extraction on the text to be processed to obtain fault description data, and then the fault description data is transformed to obtain structured fault description data. The structured fault description data can be in the form of JSON, XML, or database tables for easy storage and analysis.
[0043] The technical solution of this application embodiment involves parsing the document to be processed to obtain the text to be processed; calling a parameter text processing agent to perform structured processing on the parameter data in the text to obtain structured parameter data; and calling a fault text processing agent to perform structured processing on the fault description data in the text to obtain structured fault description data. By utilizing specially designed agents to perform structured processing on the parameter data and fault description data in the document to be processed separately, the processing efficiency and quality of structured processing of documents containing complex data types are improved.
[0044] As an optional embodiment of this application, the parameter text processing agent PP-Agent includes a parameter text recognition agent (PR-Agent) and a parameter data structuring agent (PS-Agent). The parameter text recognition agent can be understood as an intelligent system capable of automatically recognizing and parsing parameter data in parameter text. The parameter data structuring agent can be understood as an intelligent system capable of automatically understanding and structuring parameter data. S120, the step of calling the parameter text processing agent to perform structuring processing on the parameter data in the text to be processed to obtain structured parameter data includes:
[0045] S121. Call the parameter text recognition agent to extract the parameter data contained in the parameter text from the text to be processed.
[0046] Here, parameter text can be understood as text containing parameter data.
[0047] Specifically, the parameter text recognition agent PR-Agent is invoked to identify parameter text containing parameter data in the text to be processed, and the text content contained in the parameter text is extracted to obtain the parameter data.
[0048] S122. Call the parameter data structuring agent to perform structuring processing on the parameter data to obtain structured parameter data.
[0049] Specifically, the parameter data structuring agent PS-Agent is invoked to convert the unstructured or semi-structured parameter data output by the parameter text recognition agent into structured parameter data.
[0050] In an alternative embodiment, the parametric text recognition agent PR-Agent and the parametric data structuring agent PS-Agent can be constructed based on a large model and a prompt message.
[0051] For example, the prompts from the PR-Agent may include: a description of the parameter text recognition task, learning examples, and output format requirements, such as:
[0052] (1) Task Description:
[0053] "You are a parametric text content recognition expert. You will identify parametric information in the given input text. If there is no parametric content, please output "null".
[0054] (2) Learning Example:
[0055] "Input: Product A is a new energy product with very high quality and the ability to generate high economic benefits. The parameter data of Product A are as follows: maximum input voltage 550V, rated input voltage 300V;"
[0056] The statement "Product A is a new energy product with very high quality and can generate high benefits" is an introduction to Product A and not product parameter information, therefore it should be discarded. The statement "Product A's parameter data is as follows: maximum input voltage 550V, rated input voltage 300V" is product parameter information and meets the requirements.
[0057] Output: {"Parameter Information": "Product A's parameter data is as follows: Maximum input voltage 550V, rated input voltage 300V"}
[0058] (3) Output format requirements: Please output the results of parameter-related information recognition in JSON format.
[0059] The prompts from PS-Agent may include: a description of the parameter data structure processing task, learning examples, and output format requirements, for example:
[0060] (1) Task Description:
[0061] "You are an expert in parametric data structuring, which processes parametric information in a given input text in a structured manner."
[0062] (2) Example of few-shot:
[0063] "Input: The parameter data of product A are as follows: maximum input voltage 550V, rated input voltage 300V; "
[0064] Output: {"Maximum Input Voltage": "550V", "Rated Input Voltage": "300V"}
[0065] (3) Output format requirements: Please output the results of the parameter structure processing in JSON format.
[0066] This embodiment utilizes the specialized processing capabilities of both the parameter text recognition agent and the parameter data structuring agent, along with collaboration among multiple agents, to transform the parameter data contained in the text to be processed into structured parameter data. This not only improves processing efficiency but also reduces the error rate, ensuring the consistency and reliability of data processing. Furthermore, the agent-based processing architecture offers excellent scalability and flexibility.
[0067] As another optional embodiment of this application, the method further includes:
[0068] S141. Call the parameter verification agent to perform data verification on the parameter data and the structured parameter data to obtain the first verification result.
[0069] Among them, the Parameter Verification Agent (PV-Agent) can be understood as an intelligent system that can automatically verify the content and format of the extracted parameters and structured results.
[0070] The first verification result can be understood as the data verification result for both the parameter data and the structured parameter data. The first verification result can be that both the parameter data and the structured parameter data pass the verification, or the first verification result can be that the parameter data fails the verification and / or the structured parameter data fails the verification.
[0071] Specifically, the parameter data output by the parameter text recognition agent and the structured parameter data output by the parameter data structuring agent are input into the parameter verification agent. The parameter verification agent performs data verification on the parameter data and the structured parameter data to obtain the first verification result.
[0072] S142. If the first verification result includes the parameter data failing verification, then generate first optimization information based on the parameter data and send the first optimization information to the parameter text recognition agent, so that the parameter text recognition agent can re-extract the parameter data contained in the parameter text from the text to be processed based on the first optimization information.
[0073] The first optimization information can be understood as information that optimizes the prompt information of the parameter text recognition agent.
[0074] Specifically, if the first verification result indicates that the parameter data verification fails, the parameter verification agent PV-Agent generates first optimization information based on the parameter data and sends the first optimization information to the parameter text recognition agent, so that the parameter text recognition agent PR-Agent is optimized based on the first optimization information, and the parameter data contained in the parameter text is extracted from the text to be processed again based on the optimized parameter text recognition agent PR-Agent.
[0075] S143. If the first verification result includes the structured parameter data failing verification, then second optimization information is generated based on the structured parameter data, and the second optimization information is sent to the parameter data structured intelligent agent, so that the parameter data structured intelligent agent re-structures the parameter data based on the second optimization information to obtain structured parameter data.
[0076] The second optimization information can be understood as information that optimizes the prompt information of the parameter data structured intelligent agent.
[0077] Specifically, if the first verification result indicates that the structured parameter data verification fails, the parameter verification agent PV-Agent generates second optimization information based on the structured parameter data, sends the second optimization information to the parameter data structured agent PS-Agent, so that the parameter data structured agent PS-Agent is optimized based on the second optimization information, and re-structures the parameter data based on the optimized parameter data structured agent PS-Agent to obtain structured parameter data.
[0078] It should be noted that if the first verification result indicates that only the parameter data verification fails, while the structured parameter data verification passes, the parameter text recognition agent PR-Agent can be optimized, but the parameter data structured agent PS-Agent should not be optimized. However, after re-extracting the parameter data using the optimized parameter text recognition agent PR-Agent, the unoptimized parameter data structured agent PS-Agent still needs to be called to re-structure the parameter data to obtain new structured parameter data.
[0079] If the first verification result indicates that only the structured parameter data verification fails, while the parameter data verification passes, the parameter data structured agent PS-Agent can be optimized, but the parameter text recognition agent PR-Agent cannot be optimized. Alternatively, the parameter data can be restructured directly using the optimized parameter data structured agent PS-Agent to obtain new structured parameter data without regenerating the parameter data.
[0080] For example, the prompts for the optimized parameter text recognition agent (PR-Agent) may include: a description of the parameter text recognition task, learning examples, improvement requirements, and output format requirements, such as:
[0081] (1) Task Description:
[0082] "You are a parametric text content recognition expert. You will identify parametric information in the given input text. If there is no parametric content, please output "null".
[0083] (2) Learning Example:
[0084] Input: Product A is a new energy product with very high quality and the ability to generate high economic benefits. Product A's specifications are as follows: Maximum input voltage 550V, rated input voltage 300V.
[0085] The statement "Product A is a new energy product with very high quality and can generate high benefits" is an introduction to Product A and not product parameter information, therefore it should be discarded. The statement "Product A's parameter data is as follows: maximum input voltage 550V, rated input voltage 300V" is product parameter information and meets the requirements.
[0086] Output: {"Parameter Information": "Product A's parameter data is as follows: Maximum input voltage 550V, rated input voltage 300V"}
[0087] (3) Output format requirements: Please output the results of parameter-related information recognition in JSON format.
[0088] (4) Improvement requirements:
[0089] "Original Text: The original text content to be processed"
[0090] Parameter text recognition results: Initial results output by PR-Agent
[0091] Reasoning for the evaluation: The scoring criteria provided by PV-Agent
[0092] Improvement requirements: The PV-Agent provides requirements for improving the output results. For example, regarding content improvements: the output should remove irrelevant content from the original text or add relevant content; or regarding format improvements: the output should strictly adhere to the JSON format {} and not be written in curly braces such as {}}.
[0093] The optimized parameter data structuring agent PS-Agent's prompts may include: a description of the parameter data structuring processing task, learning examples, improvement requirements, and output format requirements, for example:
[0094] (1) Task Description:
[0095] "You are an expert in parametric data structuring, which processes parameter-related information in a given input text in a structured manner."
[0096] (2) Learning Example:
[0097] "Input: The parameter data of product A are as follows: maximum input voltage 550V, rated input voltage 300V; "
[0098] Output: {"Maximum Input Voltage": "550V", "Rated Input Voltage": "300V", ...}
[0099] (3) Output format requirements: Please output the results of the parameter structure processing in JSON format.
[0100] (4) Improvement requirements:
[0101] Input text: The text content to be processed
[0102] Parameter structuring results: Initial results output by PS-Agent
[0103] Reasoning for the evaluation: The scoring reasons given by PV-Agent
[0104] Improvement requirements: The PV-Agent provides requirements for improving the output results. For example, regarding content improvements: the output should remove irrelevant content from the text being processed or add relevant content; or regarding format improvements: the output should strictly adhere to the JSON format {} and not be written in curly braces such as {}}.
[0105] This embodiment uses a parameter verification agent to check the content and format of parameter data and structured parameter data. Through multiple iterations of verification and feedback mechanisms, it can accurately locate problems and provide optimization suggestions to optimize the processing flow after discovering them, so that the processing results of each step meet the expected standards, thereby improving the accuracy and reliability of the structured parameter data processing results.
[0106] As an optional embodiment of this application, the fault text processing intelligent agent FP-Agent includes a fault description text recognition intelligent agent (FR-Agent) and a fault description data structuring intelligent agent (FS-Agent). The fault description text recognition intelligent agent can be understood as an intelligent system capable of automatically recognizing and parsing fault description data in fault description text. The fault description data structuring intelligent agent can be understood as an intelligent system capable of automatically understanding and structuring fault description data. S130, the step of calling the fault text processing intelligent agent to perform structuring processing on the fault description data in the text to be processed to obtain structured fault description data includes:
[0107] S131. Call the fault description text recognition agent to extract the fault description data contained in the fault description text from the text to be processed.
[0108] The fault description text can be understood as text containing fault description data.
[0109] Specifically, the fault description text recognition intelligent agent FR-Agent is invoked to identify fault description text containing fault description data in the text to be processed, and the text content contained in the fault description text is extracted to obtain the fault description data.
[0110] S132. Call the fault description data structured intelligent agent to perform structured processing on the fault description data to obtain structured fault description data.
[0111] Specifically, the fault description data structuring agent FS-Agent is invoked to convert the unstructured or semi-structured fault description data output by the fault description text recognition agent into structured fault description data.
[0112] In an alternative embodiment, the fault description text recognition agent FR-Agent and the fault description data structuring agent FS-Agent can be constructed based on a large model and a prompt message.
[0113] For example, the prompts from FR-Agent may include: a description of the fault text recognition task, learning examples, and output format requirements, such as:
[0114] (1) Task Description:
[0115] "You are a fault text content recognition expert. You will identify fault diagnosis related information in the given input text. If there is no fault diagnosis content, please output "null".
[0116] (2) Learning Example:
[0117] Input: Product A is a new energy product with very high quality and can generate high economic benefits. The fault diagnosis data of Product A is as follows: Fault Name Fault Cause Troubleshooting Method Overheating Fault Internal fan malfunction Replace with a new fan.
[0118] The statement "Product A is a new energy product with very high quality and can generate high benefits" is an introduction to Product A and not product fault diagnosis information, therefore it should be discarded. However, the statement "Product A's fault diagnosis data is as follows: Fault Name, Fault Cause, Solution: Overheating, Internal Fan Malfunction, Replacement Fan" is product fault diagnosis information and meets the requirements.
[0119] Output: {"Fault Diagnosis Information": "Fault diagnosis data for Product A is as follows: Fault Name: Overheating; Fault Cause: Internal fan malfunction; Solution: Replace with a new fan"}
[0120] (3) Output format requirements: Please output the results of fault diagnosis related information recognition in JSON format.
[0121] The prompts provided by FS-Agent may include: a description of the fault diagnosis data structure processing task, learning examples, and output format requirements, for example:
[0122] (1) Task Description:
[0123] "You are an expert in structuring fault diagnosis data, and you are responsible for structuring fault diagnosis-related information in a given input text."
[0124] (2) Learning Example:
[0125] Input: The fault diagnosis data for product A is as follows: Fault Name: Overheating; Fault Cause: Internal fan malfunction; Solution: Replace the fan…
[0126] Output: {"Fault Name": "Overheating Fault", "Fault Cause": "Internal Fan Malfunction", "Solution": "Replace with a new fan"}
[0127] (3) Output format requirements: Please output the results of the structured processing of fault diagnosis in JSON format.
[0128] This embodiment leverages the specialized processing capabilities of both the fault description text recognition agent and the fault description data structuring agent, along with collaboration among multiple agents, to transform the fault description data contained in the text to be processed into structured fault description data. This not only improves processing efficiency but also reduces the error rate, ensuring the consistency and reliability of data processing. Furthermore, the agent-based processing architecture offers excellent scalability and flexibility.
[0129] As another optional embodiment of this application, the method further includes:
[0130] S151. Call the fault description verification agent to perform data verification on the fault description data and the structured fault description data to obtain a second verification result.
[0131] Among them, the Fault Verification Agent (PV-Agent) can be understood as an intelligent system that can automatically verify the content and format of the extracted and structured results of fault description data.
[0132] The second verification result can be understood as the data verification result for both the fault description data and the structured fault description data. The second verification result can be that both the fault description data and the structured fault description data pass the verification, or the first verification result can be that the fault description data fails the verification and / or the structured fault description data fails the verification.
[0133] Specifically, the fault description data output by the fault description text recognition agent and the structured fault description data output by the fault description data structuring agent are input into the fault description verification agent. The fault description verification agent performs data verification on the fault description data and the structured fault description data to obtain a second verification result.
[0134] S152. If the second verification result includes the failure of the fault description data verification, then generate third optimization information based on the fault description data, and send the third optimization information to the fault description text recognition agent, so that the fault description text recognition agent can re-extract the fault description data contained in the fault description text from the text to be processed based on the third optimization information.
[0135] The third optimization information can be understood as information that optimizes the prompts from the fault description text recognition agent.
[0136] Specifically, if the second verification result indicates that the fault description data verification fails, the fault description verification agent PV-Agent generates third optimization information based on the fault description data and sends the third optimization information to the fault description text recognition agent FR-Agent, so that the fault description text recognition agent FR-Agent is optimized based on the third optimization information, and the fault description data contained in the fault description text is extracted from the text to be processed again based on the optimized fault description text recognition agent FR-Agent.
[0137] S153. If the second verification result includes the structured fault description data failing verification, then a fourth optimization information is generated based on the structured fault description data, and the fourth optimization information is sent to the fault description data structured intelligent agent, so that the fault description data structured intelligent agent re-structures the fault description data based on the fourth optimization information to obtain structured fault description data.
[0138] The fourth optimization information can be understood as information that optimizes the prompting information of the fault description data structured intelligent agent.
[0139] Specifically, if the second verification result indicates that the structured fault description data verification fails, the parameter verification agent FV-Agent generates fourth optimization information based on the structured fault description data and sends the fourth optimization information to the fault description data structured agent FS-Agent, so that the fault description data structured agent FS-Agent is optimized based on the fourth optimization information. Based on the optimized fault description data structured agent FS-Agent, the fault description data is re-structured to obtain structured fault description data.
[0140] It should be noted that if the second verification result indicates that only the fault description data fails verification, while the structured fault description data passes verification, the fault description text recognition agent FR-Agent can be optimized, but the fault description data structured agent FS-Agent should not be optimized. However, after re-extracting the fault description data using the optimized fault description text recognition agent FR-Agent, it is still necessary to call the unoptimized fault description data structured agent FS-Agent to re-structure the fault description data to obtain new structured fault description data.
[0141] If the second verification result indicates that only the structured fault description data fails verification, while the fault description data passes verification, the fault description data structured intelligent agent FS-Agent can be optimized, but the fault description text recognition intelligent agent FR-Agent cannot be optimized. Alternatively, the fault description data can be restructured directly using the optimized fault description data structured intelligent agent FS-Agent to obtain new structured fault description data without regenerating the fault description data.
[0142] For example, the prompts provided by the optimized fault description text recognition agent FR-Agent may include: a description of the fault description text recognition task, learning examples, improvement requirements, and output format requirements, such as:
[0143] (1) Task Description:
[0144] "You are a fault text content recognition expert. You will identify fault diagnosis related information in the given input text. If there is no fault diagnosis content, please output "null".
[0145] (2) Learning Example:
[0146] "Input: Product A is a new energy product with very high quality and can generate high benefits. The fault diagnosis data of Product A is as follows: Fault Name Fault Cause Troubleshooting Method Overheating Fault Internal fan malfunction Replace with a new fan;
[0147] The statement "Product A is a new energy product with very high quality and can generate high benefits" is an introduction to Product A and not product fault diagnosis information, therefore it should be discarded. However, the statement "Product A's fault diagnosis data is as follows: Fault Name, Fault Cause, Solution: Overheating, Internal Fan Malfunction, Replacement Fan" is product fault diagnosis information and meets the requirements.
[0148] Output: {"Fault Diagnosis Information": "Fault diagnosis data for Product A is as follows: Fault Name: Overheating; Fault Cause: Internal fan malfunction; Solution: Replace with a new fan"}
[0149] (3) Output format requirements: Please output the results of fault diagnosis related information recognition in JSON format.
[0150] (4) The improvement requirements are as follows:
[0151] "Original Text: The original text content to be processed"
[0152] Fault diagnosis text recognition results: Initial results output by FR-Agent
[0153] Reasoning for the rating: DV-Agent's rating criteria
[0154] Improvement requirements: The DV-Agent provides requirements for improving the output results. For example, regarding content improvements: the output should remove irrelevant content from the original text or add relevant content; or regarding format improvements: the output should strictly adhere to the JSON format {} and not be written in curly braces such as {}}.
[0155] For example, the prompts for the optimized fault description data structuring agent FS-Agent may include: a description of the fault description data structuring task, learning examples, improvement requirements, and output format requirements, such as:
[0156] (1) Task Description:
[0157] "You are an expert in structuring fault diagnosis data, and you are responsible for structuring the fault description information in the given input text."
[0158] (2) Learning Example:
[0159] "Input: The fault diagnosis data for product A is as follows: Fault Name Fault Cause Solution Overheating Fault Internal fan malfunction Replace with a new fan;
[0160] Output: {"Fault Name": "Overheating Fault", "Fault Cause": "Internal Fan Malfunction", "Solution": "Replace with a new fan"}
[0161] (3) Output format requirements: Please output the results of the fault diagnosis structured processing in JSON format.
[0162] (4) The improvement requirements are as follows:
[0163] Input text: The text content to be processed
[0164] Fault diagnosis structured processing results: Initial results output by FS-Agent
[0165] Reasoning for the rating: The reasons given by FV-Agent
[0166] Improvement requirements: The FV-Agent provides requirements for improving the output results. For example, regarding content improvements: the output should remove irrelevant content from the text being processed or add relevant content; or regarding format improvements: the output should strictly adhere to the JSON format {} and not be written in curly braces such as {}}.
[0167] This embodiment uses a fault description verification agent to check the content and format of fault description data and structured fault description data. Through multiple iterations of verification and feedback mechanisms, it can accurately locate problems and provide optimization suggestions to optimize the processing flow after discovering them, so that the processing results of each step meet the expected standards, thereby improving the accuracy and reliability of the structured processing results of fault description data.
[0168] In one optional embodiment, the parameter verification agent PV-Agent and the fault description verification agent FV-Agent can be integrated into a common data verification agent (DV-Agent), responsible for verifying the content and format of parameter data, structured parameter data, fault description data, and structured fault description data. The data verification agent DV-Agent distinguishes verification results through identification information.
[0169] For example, the prompt flag indicates the verification result and the agent that needs optimization. The output flag indicates the processing result output by the data verification agent DV-Agent. For example:
[0170] A prompt flag of 00 indicates that the verification failed and prompt optimization information needs to be sent to the parameter text recognition agent.
[0171] A prompt flag of 01 indicates that the verification failed and prompt optimization information needs to be sent to the fault description text recognition agent.
[0172] A prompt flag of 10 indicates that the validation failed, meaning that the prompt optimization information needs to be sent to the parameter structure processing agent.
[0173] A prompt flag of 11 indicates that the verification failed, meaning that the prompt optimization information needs to be sent to the fault description data structured agent.
[0174] A prompt flag of 20 indicates that the verification failed and that the prompt optimization information needs to be sent to both the parameter recognition agent and the parameter data structuring agent.
[0175] A prompt flag of 21 indicates that the verification failed, meaning that the prompt optimization information needs to be sent to both the fault description text recognition agent and the fault description data structure agent.
[0176] A prompt flag of 3 indicates that the verification has passed and no prompt optimization information needs to be sent.
[0177] An output flag of 0 indicates that the validation failed and the output is empty.
[0178] An output flag of 1 indicates that the validation passed and the output result is structured parameter data.
[0179] An output flag of 2 indicates that the verification passed and the output result is structured fault description data.
[0180] As an optional embodiment of this application, the method further includes: constructing parameter triples based on the structured parameter data and / or constructing fault triples based on the structured fault description data.
[0181] In this context, a triple can be understood as an ordered set of three elements, and its structure is used to describe the relationships between things in the real world. Based on the concept of triples, and extending this with fault information, we obtain parameter triples and fault triples. A parameter triple can be understood as a set of three related elements organized from parameter data. A fault triple can be understood as a set of three related elements organized from fault description data. Specifically, structured parameter data is converted into triples in a knowledge graph to construct parameter triples; structured fault description data is converted into triples in a knowledge graph to construct fault description triples. The process of constructing parameter triples and fault description triples involves identifying entities, attributes, and relationships in the data and organizing them into a triple structure.
[0182] In addition, the constructed parameter triples and fault description triples can be stored in the data storage module, so that structured data can be flexibly and effectively saved and quickly retrieved, making it convenient for users to conduct further analysis and application.
[0183] For example, if the output flag of the data verification agent DV-Agent is 1, the structured parameter data is sent to the parameter triplet construction module to construct the parameter triplet; if the output flag of the data verification agent DV-Agent is 2, the structured fault description data is sent to the fault triplet construction module to construct the fault triplet. The graph schema used by the parameter triplet construction module and the fault triplet construction module can be:
[0184] Entity tags: Param, Fault…
[0185] Relationship tags: Param-hasValue->Value, Fault-hasReason->Reason.
[0186] The data storage module can save parameter triples and fault description triples in file server format such as txt, or in database such as MySQL in string format, or directly in graph database.
[0187] This embodiment converts the structured parameter data and fault description data into triples based on graph entity relationships. This processing method not only facilitates data visualization and correlation analysis but also provides a solid foundation for subsequent data mining and knowledge graph construction.
[0188] As an optional embodiment of this application, the system further includes a Language Recognition Agent (LR-Agent) and a Language Translation Agent (T-Agent). The Language Recognition Agent can be understood as an agent capable of automatically recognizing the language type of the input text. The Language Translation Agent can be understood as an agent capable of automatically translating the language type of the text. After parsing the document to be processed and obtaining the text to be processed, the system further includes:
[0189] S161. Call the language recognition agent to identify the language type of the text to be processed.
[0190] S162. If the language type is a second language type other than the first language type, then the language translation agent is invoked to translate the second language type of the text to be processed into the first language type.
[0191] The language type of the text to be processed can be any language in the world or multiple languages. The first language type can be a preset language type, such as Chinese; the second language type can be any one or more language types other than the first language type.
[0192] Specifically, after obtaining the text to be processed, the language recognition agent is invoked to identify the language type of the text. If the language type of the text to be processed is not the preset first language type, the language translation agent is invoked to translate the second language type of the text to the first language type.
[0193] Optionally, the language recognition agent and language translation agent can be built based on large model technology. Using large model technology unifies the text of the documents to be processed into a single language, which not only improves the accuracy of text recognition and translation but also significantly increases processing speed, providing a foundation for improving the efficiency and accuracy of information processing in the documents.
[0194] In a specific example Figure 2This is a schematic diagram of the structure of a document processing system provided in Embodiment 1 of this application, as shown below. Figure 2 As shown, the document processing system includes a document parsing module, a document segmentation module, a text translation module, a text structuring module, a triplet construction module, and a data storage module. The text translation module includes a language recognition agent (LR-Agent) and a language translation agent (T-Agent). The text structuring module includes a parametric text recognition agent (PR-Agent), a parametric data structuring agent (PS-Agent), a fault description text recognition agent (FR-Agent), a fault description data structuring agent (FS-Agent), and a data verification agent (DV-Agent). The data verification agent (DV-Agent) is used to verify and optimize the content and format of the parametric data output by the parametric text recognition agent (PR-Agent), the structured parametric data output by the parametric data structuring agent (PS-Agent), the fault description data output by the fault description text recognition agent (FR-Agent), and the structured fault description data output by the fault description data structuring agent (FS-Agent).
[0195] This application leverages the specialized processing capabilities of each agent and their collaboration to ensure efficient and accurate completion of the entire process, from data extraction to structured processing of the document to be processed. Furthermore, the agent architecture gives the system excellent scalability and flexibility.
[0196] Example 2
[0197] Figure 3 This is a schematic diagram of the structure of a document processing system provided in Embodiment 2 of this application. Figure 3 As shown, the system includes: a document parsing module 310, a parameter structuring module 320, and a fault structuring module 330; wherein,
[0198] The document parsing module 310 is used to parse the document to be processed and obtain the text to be processed.
[0199] The parameter structuring module 320 is used to call the parameter text processing agent to perform structuring processing on the parameter data in the text to be processed, and obtain structured parameter data.
[0200] The fault structuring module 330 is used to call the fault text processing agent to perform structuring processing on the fault description data in the text to be processed, and obtain structured fault description data.
[0201] The technical solution of this application embodiment involves parsing the document to be processed to obtain the text to be processed; calling a parameter text processing agent to perform structured processing on the parameter text in the document to be processed to obtain structured parameter data; and calling a fault text processing agent to perform structured processing on the fault description text in the document to be processed to obtain structured fault description data. By utilizing specially designed agents to perform structured processing on the parameter text and fault description text in the document to be processed separately, a highly automated and intelligent processing flow is achieved. This solves the problem that traditional document processing methods struggle to achieve fast and effective structured conversion when dealing with documents containing complex data types, thus improving the processing efficiency and quality of structured processing for documents containing complex data types.
[0202] Optionally, the parameter text processing agent includes a parameter text recognition agent and a parameter data structuring agent;
[0203] The parameter structuring module 320 is specifically used for:
[0204] The parameter text recognition agent is invoked to extract the parameter data contained in the parameter text from the text to be processed;
[0205] The parameter data structuring agent is invoked to perform structuring processing on the parameter data to obtain structured parameter data.
[0206] Optionally, the system further includes:
[0207] The parameter verification module is used to call the parameter verification agent to perform data verification on the parameter data and the structured parameter data, and obtain a first verification result;
[0208] The first optimization module is configured to generate first optimization information based on the parameter data if the first verification result includes the parameter data failing verification, and send the first optimization information to the parameter text recognition agent so that the parameter text recognition agent can re-extract the parameter data contained in the parameter text from the text to be processed based on the first optimization information.
[0209] The second optimization module is configured to generate second optimization information based on the structured parameter data if the first verification result includes the structured parameter data failing verification, and send the second optimization information to the parameter data structured intelligent agent so that the parameter data structured intelligent agent can re-structure the parameter data based on the second optimization information to obtain structured parameter data.
[0210] Optionally, the fault text processing agent includes a fault description text recognition agent and a fault description data structuring agent.
[0211] The fault structuring module 330 is specifically used for:
[0212] The fault description text recognition agent is invoked to extract the fault description data contained in the fault description text from the text to be processed;
[0213] The fault description data is processed by a fault description data structuring agent to obtain structured fault description data.
[0214] Optionally, a fault description verification agent may also be included;
[0215] The system also includes:
[0216] The fault description verification module is used to call the fault description verification agent to perform data verification on the fault description data and the structured fault description data, and obtain a second verification result.
[0217] The third optimization module is used to generate third optimization information based on the fault description data if the second verification result includes the fault description data failing verification, and send the third optimization information to the fault description text recognition intelligent agent so that the fault description text recognition intelligent agent can re-extract the fault description data contained in the fault description text from the text to be processed based on the third optimization information.
[0218] The fourth optimization module is used to generate fourth optimization information based on the structured fault description data if the second verification result includes the structured fault description data failing verification, and send the fourth optimization information to the fault description data structured intelligent agent so that the fault description data structured intelligent agent can re-structure the fault description data based on the fourth optimization information to obtain structured fault description data.
[0219] Optional, also includes:
[0220] The triplet construction module is used to construct parameter triplets based on the structured parameter data and / or construct fault triplets based on the structured fault description data.
[0221] Optional, also includes:
[0222] The language recognition module is used to invoke the language recognition agent to identify the language type of the text to be processed.
[0223] The language translation module is used to call the language translation agent to translate the second language type of the text to be processed into the first language type if the language type is a second language type other than the first language type.
[0224] The document processing system provided in this application can execute the document processing method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects of the method execution.
[0225] Example 3
[0226] Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of this application, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the application described and / or claimed herein.
[0227] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.
[0228] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.
[0229] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as document processing methods.
[0230] In some embodiments, the document processing method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the document processing method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the document processing method by any other suitable means (e.g., by means of firmware).
[0231] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.
[0232] In some embodiments, the document processing method may be implemented as a computer program, which is implicitly included in a computer program product. When executed by a processor, the computer program implements the document processing method of this application. The computer program product can be understood as a software product that primarily implements its solution through a computer program. The computer program used to implement the method of this application may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0233] In the context of this application, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0234] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).
[0235] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.
[0236] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.
[0237] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this application can be achieved, and this is not limited herein.
[0238] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.
Claims
1. A document processing method, characterized in that, include: The document to be processed is parsed to obtain the text to be processed; The parameter text processing agent is invoked to perform structured processing on the parameter data in the text to be processed, thereby obtaining structured parameter data; The fault text processing agent is invoked to perform structured processing on the fault description data in the text to be processed, thereby obtaining structured fault description data.
2. The document processing method according to claim 1, characterized in that, The parameter text processing agent includes a parameter text recognition agent and a parameter data structuring agent; The invoke parameter text processing agent performs structured processing on the parameter data in the text to be processed to obtain structured parameter data, including: The parameter text recognition agent is invoked to extract the parameter data contained in the parameter text from the text to be processed; The parameter data structuring agent is invoked to perform structuring processing on the parameter data to obtain structured parameter data.
3. The document processing method according to claim 2, characterized in that, The method further includes: The parameter verification agent is invoked to perform data verification on the parameter data and the structured parameter data, and a first verification result is obtained; If the first verification result includes the parameter data failing verification, then first optimization information is generated based on the parameter data, and the first optimization information is sent to the parameter text recognition agent, so that the parameter text recognition agent can re-extract the parameter data contained in the parameter text from the text to be processed based on the first optimization information. If the first verification result includes the structured parameter data failing verification, then second optimization information is generated based on the structured parameter data, and the second optimization information is sent to the parameter data structured agent, so that the parameter data structured agent can re-structure the parameter data based on the second optimization information to obtain structured parameter data.
4. The document processing method according to claim 1, characterized in that, The fault text processing agent includes a fault description text recognition agent and a fault description data structuring agent. The process of calling the fault text processing agent to perform structured processing on the fault description data in the text to be processed, thereby obtaining structured fault description data, includes: The fault description text recognition agent is invoked to extract the fault description data contained in the fault description text from the text to be processed; The fault description data is processed by a fault description data structuring agent to obtain structured fault description data.
5. The document processing method according to claim 4, characterized in that, The method further includes: The fault description verification agent is invoked to perform data verification on the fault description data and the structured fault description data to obtain a second verification result; If the second verification result includes the failure of the fault description data verification, then the third optimization information is generated based on the fault description data, and the third optimization information is sent to the fault description text recognition agent, so that the fault description text recognition agent can re-extract the fault description data contained in the fault description text from the text to be processed based on the third optimization information. If the second verification result includes the structured fault description data failing verification, then fourth optimization information is generated based on the structured fault description data, and the fourth optimization information is sent to the fault description data structured intelligent agent, so that the fault description data structured intelligent agent can re-structure the fault description data based on the fourth optimization information to obtain structured fault description data.
6. The document processing method according to claim 1, characterized in that, The method further includes: Parameter triples are constructed based on the structured parameter data and / or fault triples are constructed based on the structured fault description data.
7. The document processing method according to claim 1, characterized in that, After parsing the document to be processed and obtaining the text to be processed, the process also includes: Call upon a language recognition agent to identify the language type of the text to be processed; If the language type is a second language type other than the first language type, then the language translation agent is invoked to translate the second language type of the text to be processed into the first language type.
8. A document processing system, characterized in that, The system includes: The document parsing module is used to parse the document to be processed and obtain the text to be processed. The parameter structuring module is used to call the parameter text processing agent to perform structuring processing on the parameter data in the text to be processed, and obtain structured parameter data. The fault structuring module is used to call the fault text processing agent to perform structuring processing on the fault description data in the text to be processed, and obtain structured fault description data.
9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the document processing method according to any one of claims 1-6.
10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the document processing method according to any one of claims 1-6.