Technical document writing method, device and system based on intelligent agent

Abstract

The invention discloses a technical document writing method, device and system based on intelligent agents, and the method comprises the steps: based on a preset knowledge base, a first intelligent agent is used for giving at least one question according to a theme input by a user and receiving the reply of the user, and the question is used for guiding the user to define a research direction; based on the knowledge base, the second agent is used for giving a research thought according to the question and the reply of the user, and the research thought comprises one or more of a scene, a question, a limiting condition, significance, connotation and a research method involved in a technical document; based on the knowledge base, the third agent is used for determining chapter titles of the technical document according to the research thought; based on the knowledge base, the fourth agent is used for writing the technical document according to the chapter title and the research thought. According to the technical scheme, the problem that illusion exists in document generation in the prior art can be solved.

Description

TECHNICAL FIELD

[0001] The present document relates to the technical field of artificial intelligence based on agents, and in particular to an agent-based technical document writing method, device and system. BACKGROUND

[0002] In the face of explosive growth and rapid iteration of knowledge, traditional manual writing of technical documents is difficult to guarantee comprehensiveness and timeliness, and the quality of the documents is strongly related to the knowledge reserve of the writer, and it is difficult to maintain stable and high-quality output.

[0003] To solve the above problems, more and more existing technologies apply large language models such as Deepseek and Qwen to document writing tasks.

[0004] However, due to the lack of sufficient professional knowledge in general large models, the hallucination problem is obvious when generating technical documents, and the reliability of the generated results is low, especially when it comes to technical indicators. On the other hand, although LLM (Large Language Model) can generate the text content required by technical documents, it cannot directly generate system architecture diagrams, flowcharts and other pictures that assist in understanding the technology. In actual application, it is often necessary to first describe the picture in code format through LLM, and then access other drawing platforms to obtain visual pictures. SUMMARY

[0005] In view of the above scheme, the present application aims to provide an agent-based technical document writing method, device and system to solve at least one of the above problems.

[0006] In a first aspect, one or more embodiments of the present application provide an agent-based technical document writing method, the agent including a first agent, a second agent, a third agent and a fourth agent, the method comprising the following steps: Based on a preset knowledge base, the first agent is configured to give at least one question according to a topic input by a user, and receive the user's reply, the question being used to guide the user to clarify the research direction; Based on the knowledge base, the second agent is configured to give a research idea according to the question and the user's reply, the research idea including one or more of the scene, the problem, the limitation, the significance, the connotation and the research method involved in the technical document; Based on the knowledge base, the third agent is configured to determine a chapter title of the technical document according to the research idea; Based on the knowledge base, the fourth agent is configured to write the technical document according to the chapter title and the research idea.

[0007] Further, the knowledge base includes an internal knowledge base and an external knowledge base; The internal knowledge base is used to store the company's core technologies, R&D ideas, and technological achievements. The external knowledge base is used to store network information collected based on knowledge data from the internal knowledge base.

[0008] Furthermore, the technical documentation includes: Based on the knowledge base and the research approach described above, write the article's opening and closing paragraphs respectively. Based on the knowledge base, and according to the research ideas and the chapter titles, the text corresponding to each chapter title is written.

[0009] Furthermore, the text corresponding to the chapter titles is written, including: Extract keywords from the chapter titles; Based on the keywords, target text data is selected from the knowledge base in a sequential manner; Based on the research approach and target text data, write the text corresponding to the chapter titles.

[0010] Furthermore, the method also includes: Based on a preset fifth intelligent agent, the document format of the technical document is optimized according to a preset format.

[0011] Furthermore, the method also includes: Based on a pre-set sixth intelligent agent, the system receives the user's input of the chart requirements. Generate a chart in a preset format based on the chart requirements information; Based on a pre-set Python script, the chart is transformed into a visual chart; Embed the chart into the technical document.

[0012] Furthermore, the method also includes: Based on a pre-defined seventh intelligent agent, the technical document is optimized according to user needs.

[0013] Secondly, one or more embodiments of this application provide a technical document writing apparatus based on an intelligent agent, comprising: The data acquisition module is used to acquire data to be processed. The extraction module is used to extract keyword vectors from the data to be processed based on the simulated annealing algorithm; The data processing module is used to input the keyword vector into a preset dual-output deep neural network and output a patent list; The feedback module is used to collect user interaction data based on the patent list; when a search command is received from an external source, it adjusts the weight of the keyword vector according to the interaction data and performs the search. Thirdly, embodiments of this application provide a technical document writing system based on intelligent agents, comprising: the technical document writing device described in the second aspect and the knowledge base described in any one of the first aspects.

[0014] Thirdly, embodiments of this application provide a storage medium for storing computer-executable instructions, characterized in that, when executed, the computer-executable instructions implement the steps of the agent-based technical document writing method described in any one of the first aspects.

[0015] Compared with the prior art, this application can achieve at least the following technical effects: The beneficial effects of this invention are as follows: through dialogue between the agent and the user, the user can be guided to clarify the research direction, while the agent's logic can be constrained. Both a clear research direction and constrained logic reduce the probability of large models exhibiting illusions. Furthermore, each step is based on a knowledge base, enabling the use of professional knowledge to constrain the agent, further reducing the probability of large models exhibiting illusions. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in one or more embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a method for writing technical documents based on an intelligent agent, provided for one or more embodiments of this application; Figure 2 A schematic diagram of the structure of a technical document writing device based on an intelligent agent, provided for one or more embodiments of this application; Figure 3 A schematic diagram of the structure of a technical document writing system based on an intelligent agent, provided for one or more embodiments of this application; Figure 4 A schematic diagram illustrating the working principle of a technical document writing device based on an intelligent agent, provided for one or more embodiments of this application; Figure 5 This is a schematic diagram illustrating the working principle of a knowledge base provided for one or more embodiments of this application. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this application, the technical solutions in one or more embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on one or more embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the protection scope of this document.

[0019] Example 1 This invention provides a method for writing technical documents based on intelligent agents, wherein the intelligent agents include a first intelligent agent, a second intelligent agent, a third intelligent agent, and a fourth intelligent agent. The method includes the following steps, as follows: Figure 1 As shown: Step 1: Based on a pre-set knowledge base, the first intelligent agent is used to provide at least one question based on the topic input by the user and receive the user's answer.

[0020] In this embodiment, the knowledge base includes an internal knowledge base and an external knowledge base. The internal knowledge base stores the enterprise's core technologies, R&D ideas, and technological achievements. The external knowledge base stores network information collected based on knowledge data from the internal knowledge base. Questions are used to guide users to clarify their research direction. For example, if a user inputs "I want to write a technical report on topic A," the first agent will ask questions based on the knowledge base, such as: What is the meaning of topic A? What are the applicable scenarios for topic A? Further questions will be raised based on the user's answers. The number of questions can be set to a fixed value, or the user can trigger the termination of the questioning process.

[0021] Step 2: Based on the knowledge base, the second intelligent agent provides research ideas based on the questions and user responses.

[0022] In this embodiment, the research approach includes one or more of the following: scenarios, problems, limitations, significance, connotations, and research methods involved in the technical document. For example, a user may only answer one limitation, but a second intelligent agent, based on the user-provided topic and knowledge base, analyzes the background of that topic to determine its connotation and extension. Then, if it discovers that other limitations are needed based on the topic's connotation and extension, it automatically completes the limitations and explains their significance to the analysis report and the problems to be solved. Following this, a research approach is proposed based on the above information.

[0023] Step 3: Based on the knowledge base, the third agent determines the chapter titles of the technical documents according to the research ideas.

[0024] In this embodiment of the application, the chapter titles of the technical document are equivalent to the writing outline of the technical document.

[0025] Step 4: Based on the knowledge base, the fourth agent is used to write technical documents according to the chapter titles and research ideas.

[0026] In this embodiment of the application, the fourth intelligent agent searches for relevant information from the knowledge base according to the research ideas, and then fills in the content under the corresponding chapter title based on the relevant information.

[0027] In this application embodiment, writing technical documents specifically includes: (1) Write the beginning and ending of the article according to the research ideas.

[0028] (2) Based on the research approach and chapter titles, simultaneously write the text corresponding to each chapter title. When writing the text corresponding to the chapter titles, first extract keywords from the chapter titles, then select target text data from the knowledge base based on the keywords, and finally write the corresponding text based on the research approach and target text data. This simultaneous writing method improves document generation efficiency.

[0029] In this embodiment of the application, writing the text corresponding to the chapter title includes: Keywords are extracted from the chapter titles; target text data is selected from the knowledge base based on the keywords; and the text corresponding to the chapter titles is written based on the research approach and the target text data. For example, keywords are generated by combining the topic, subheadings, and question-and-answer information. These keywords are then used to iteratively search the industry-specific knowledge base. After integrating the search results, the text generation capabilities of generative LLM are used to write the document's opening paragraph, chapter content, and closing paragraph, gradually filling in the document framework. The selection of target text data from the knowledge base is performed sequentially to ensure the coherence of the knowledge.

[0030] In this embodiment, based on a preset fifth intelligent agent, the document format of the technical document is optimized according to a preset format. The document format includes font, font size, paragraph format, alignment, line spacing, etc.

[0031] In this embodiment, a pre-defined sixth intelligent agent receives user-inputted chart requirements. Based on these requirements, a chart in a pre-defined format is generated. Using a pre-defined Python script, the chart is transformed into a visual representation. The chart is then embedded into a technical document.

[0032] Specifically, the system first understands the user's intent based on the user's input and extracts information such as chart name, chart type, detailed requirements, and optimization requirements. The intelligent agent can then use the aforementioned information to call tools to draw and visualize the chart.

[0033] Next, based on the database and according to the chart name, chart type, and drawing requirements, the LLM function is invoked to generate charts in Markdown or Mermaid format, and the formatted data in the LLM output is extracted. Using pre-connected Flask services and Python scripts, Markdown charts are converted into HTML links, and Mermaid charts are converted into SVG images, thus achieving the visualization of the charts.

[0034] Finally, if the user wants to optimize the generated chart, the agent first re-examines the chart drawing requirements based on the original needs and the user's subsequent optimization requests. Then, it calls the tool again to draw and visualize the chart based on the updated requirements.

[0035] In this embodiment of the application, the technical document is optimized according to user needs based on a preset seventh intelligent agent.

[0036] This method automates the writing of technical documents through the collaborative work of multiple agents. The first agent guides the user to clarify the research direction by posing questions; the second agent generates the research framework; the third agent determines the chapter titles; the fourth agent writes the document content; the fifth agent optimizes the document format; the sixth agent generates and embeds charts and graphs; and the seventh agent performs final optimization based on user needs. This agent-based approach improves the efficiency and quality of technical document writing while reducing manual workload.

[0037] Example 2 This embodiment provides a technical document writing device based on intelligent agents, such as... Figure 2 As shown, it includes a question-and-answer module 201, a research idea generation module 202, a document structure construction module 203, and a chapter writing module 204.

[0038] The question-and-answer module 201, based on a pre-defined knowledge base, provides at least one question based on the user's input topic and receives the user's response. These questions aim to guide the user in clarifying their research direction. For example, when a user inputs the topic "Application of Artificial Intelligence in Medical Diagnosis," the question-and-answer module might ask questions such as: "What type of medical diagnosis do you hope to focus on?" and "What artificial intelligence technologies are you considering using?" These questions help the user further focus their research scope.

[0039] The research idea generation module 202, based on a knowledge base, provides research ideas based on the questions posed by the question-and-answer module and the user's responses. Research ideas may include one or more of the following: scenarios, problems, limitations, significance, implications, and research methods related to the technical documentation. For example, for the above topic, research ideas might include: "Scenario: Rapid diagnosis in the emergency room; Problem: How to improve the diagnostic accuracy of rare diseases; Limitations: Limited medical resources and time; Significance: Reducing misdiagnosis rates and improving treatment outcomes; Implication: Combining deep learning and expert systems; Research Methods: Model training and clinical validation based on large-scale medical datasets."

[0040] The document structure building module 203, based on a knowledge base, determines the chapter titles of the technical document according to the research approach. It analyzes the key elements of the research approach and transforms them into a logically clear document structure. For example, it might generate chapter titles like: "1. Introduction 2. Related Work 3. System Architecture 4. Algorithm Design 5. Experimental Setup 6. Results Analysis 7. Discussion 8. Conclusion". This structure comprehensively covers the research content and conforms to the standard format of academic papers.

[0041] The chapter writing module 204, based on a knowledge base, writes the specific content of the technical document according to the chapter titles and research ideas. It extracts relevant research idea elements from each chapter title and expands them into complete paragraphs. For example, the "System Architecture" chapter might describe in detail the overall framework of the AI ​​diagnostic system, the functions of each module, and the interactions between them. The "Algorithm Design" chapter might discuss in depth the deep learning model structure and training methods used.

[0042] In a preferred embodiment, the question-answering module can also dynamically adjust subsequent questions based on the user's responses. For example, if the user expresses particular interest in neural network technology, the question-answering module may ask more specific questions related to neural networks, such as "What type of neural network do you intend to use?" or "How can the interpretability problem of neural networks in medical diagnosis be solved?" This dynamic adjustment can more accurately capture the user's research intentions.

[0043] In another preferred embodiment, the research idea generation module can provide multiple alternative research ideas for the user to choose from. For example, it may simultaneously offer two research ideas: "image diagnosis based on deep learning" and "medical record analysis combined with natural language processing." The user can choose the most suitable research direction based on their interests and expertise.

[0044] In another preferred embodiment, the document structure building module can generate corresponding chapter structures based on different document types (such as academic papers, technical reports, patent applications, etc.). For example, for a patent application, it may generate chapter headings such as "Technical Field," "Background Art," "Summary of the Invention," "Description of Drawings," and "Detailed Description of the Embodiments." This flexibility allows the device to adapt to the writing needs of different types of technical documents.

[0045] In a preferred embodiment, the chapter writing module can interact with the user in real time during the writing process. It may pause after each chapter is completed to show the user the current content and solicit feedback. The user can offer suggestions for revision or request further elaboration on certain content. This interactive writing approach ensures that the generated document better meets the user's expectations.

[0046] This device significantly improves the efficiency and quality of technical document writing through intelligent question-and-answer guidance, research idea generation, document structure construction, and chapter writing. It not only helps users quickly organize their research ideas but also generates well-structured and content-rich technical documents. This has significant practical value for researchers, engineers, and students, significantly reducing their time and effort spent on document writing, allowing them to focus more on core research work.

[0047] Example 3 This embodiment provides a technical document writing system based on intelligent agents, such as Figure 3 As shown. The system includes an agent-based technical document writing device and a knowledge base.

[0048] In this embodiment, the agent-based technical document writing device constructs an automated agent workflow and accesses a professional knowledge database to generate technical documents based on user-input topics, and supports subsequent optimization and modification of these documents. The device includes a topic question-and-answer submodule, a research idea optimization submodule, a document structure construction submodule, a chapter writing submodule, a document formatting output submodule, and a document optimization submodule. This clearly defined submodule workflow enables the agent to decompose complex and abstract technical document writing tasks into multiple simple, progressively more in-depth subtasks, ensuring both the depth of document writing and the stability of output quality. Its working process is as follows: Figure 4 As shown.

[0049] In this application embodiment, the ability of small models to handle simple repetitive tasks and their low resource consumption are utilized to construct internal and external professional knowledge bases through two methods: data uploading and web information crawling. This enables the systematic integration and efficient utilization of professional knowledge, providing strong support for the writing of technical documents.

[0050] Specifically, regarding the knowledge base construction process, the internal knowledge base is primarily built through data uploading. Small models combined with automated workflows are used to extract, classify, store, maintain, and update information from internal data. This improves construction efficiency and reduces maintenance costs while ensuring data privacy and security. The external knowledge base is primarily built using small models combined with web scraping. Scraped web information is processed through small models to remove duplicate, erroneous, and irrelevant information before being uploaded to an online database, ensuring the data's real-time performance, accuracy, and availability. Furthermore, the use of small models further saves on computing resources while maintaining high performance in information extraction and integration and effective data cleaning.

[0051] When working, a knowledge base can improve the professionalism of documents and images generated by the agent.

[0052] Specifically, after a user inputs the title of a technical document during the writing process, the agent can retrieve knowledge from a knowledge base and use the returned knowledge as a prompt to generate a professional document. The generated charts are customizable to match the document. The user inputs the chart type, name, and the chapter content output by the agent, and then the knowledge base generates a customized professional chart for the user. Based on the knowledge base, the agent can provide suggestions for document writing in the form of charts, so the user does not need to input detailed requirements. For example, a user can request the agent to provide research in the form of a mind map as suggestions for writing a technical document.

[0053] The workflow of a knowledge base is as follows: Figure 5 As shown, the outer user interaction module understands the user's intent based on the user input, extracting information such as chart name, chart type, detailed requirements, and optimization requirements. If the user's intent is for the first chart generation, the agent can call the tool to draw and visualize the chart based on the aforementioned information. If the user's intent is for subsequent optimization after chart generation, the agent first re-examines the chart drawing requirements based on the original requirements and the user's subsequent input optimization requirements, and then calls the tool again to draw and visualize the chart based on the updated requirements. The inner chart drawing tool module is an automated chart generation tool that takes the received chart name, chart type, and drawing requirements as input and outputs charts in code format and visualization form. This sub-module first calls LLM to generate charts in Markdown or Mermaid format and extracts the formatted data from the LLM output. The module integrates a Flask service Python script, which can convert Markdown format charts into HTML links and Mermaid format charts into SVG images, thus achieving the visualization of the charts.

[0054] This application provides a storage medium for storing computer-executable instructions, characterized in that, when executed, the computer-executable instructions implement the steps of the agent-based technical document writing method described in any one of the embodiments.

[0055] It should be noted that the embodiments concerning storage media in this application and the embodiments concerning the agent-based technical document writing method in this application are based on the same inventive concept. Therefore, the specific implementation of this embodiment can be referred to the implementation of the corresponding agent-based technical document writing method described above, and the repeated parts will not be described again.

[0056] The foregoing has described specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0057] In the 1930s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many improvements to the methodology today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that an improvement to the methodology cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0058] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0059] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0060] For ease of description, the above apparatus is described by dividing it into various functional units. Of course, in implementing the embodiments of this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0061] Those skilled in the art will understand that one or more embodiments of this application can be provided as a method, system, or computer program product. Therefore, one or more embodiments of this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0062] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0063] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0064] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0065] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0066] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0067] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0068] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0069] One or more embodiments of this application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. One or more embodiments of this application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can reside in local and remote computer storage media, including storage devices.

[0070] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0071] The above description is merely an embodiment of this document and is not intended to limit the scope of this document. Various modifications and variations can be made to this document by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this document should be included within the scope of the claims of this document.

Claims

1. An agent-based technical document authoring method, characterized by, The intelligent agent includes a first intelligent agent, a second intelligent agent, a third intelligent agent and a fourth intelligent agent, and the method includes the following steps: Based on a preset knowledge base, the first intelligent agent is configured to give at least one question according to a topic input by a user, and receive a reply of the user, and the question is configured to guide the user to clearly define a research direction; Based on the knowledge base, the second intelligent agent is configured to give a research idea according to the question and the reply of the user, and the research idea includes one or more of a scenario, a question, a limitation condition, a significance, a connotation and a research method involved in a technical document; Based on the knowledge base, the third intelligent agent is configured to determine a chapter title of the technical document according to the research idea; Based on the knowledge base, the fourth intelligent agent is configured to write the technical document according to the chapter title and the research idea.

2. The method of claim 1, wherein the knowledge base includes an internal knowledge base and an external knowledge base; The internal knowledge base is configured to store core technology, research ideas and technical achievements within an enterprise; The external knowledge base is configured to store network information collected based on knowledge data in the internal knowledge base.

3. The method of claim 1, wherein the writing of the technical document includes: Based on the knowledge base, an article beginning and an article end are respectively written according to the research idea; Based on the knowledge base, a text corresponding to each chapter title is written according to the research idea and the chapter title.

4. The method of claim 3, wherein the writing of the text corresponding to each chapter title includes: Extracting a keyword from the chapter title; Selecting target text data from the knowledge base in a serial manner according to the keyword; Writing the text corresponding to the chapter title according to the research idea and the target text data. The method further includes: Based on a preset fifth intelligent agent, optimizing a document format of the technical document according to a preset format. The method further includes:

5. The method of claim 1, wherein, Based on a preset sixth intelligent agent, receiving inputted chart requirement information of a user; Generating a chart in a preset format according to the chart requirement information; 6. The method of claim 1, wherein, Based on a preset python script, converting the chart into a visual chart; Embedding the chart into the technical document. The method further includes: Based on a preset seventh intelligent agent, optimizing the technical document according to a user demand. The method includes:

7. The method of claim 1, wherein, A question and answer module configured to give at least one question according to a topic input by a user based on a preset knowledge base, and receive a reply of the user, and the question is configured to guide the user to clearly define a research direction; A research idea generation module configured to give a research idea according to the question and the reply of the user based on the knowledge base, and the research idea includes one or more of a scenario, a question, a limitation condition, a significance, a connotation and a research method involved in a technical document; 8. An agent-based technical document authoring apparatus, characterized by comprising: A document structure construction module configured to determine a chapter title of the technical document according to the research idea based on the knowledge base; A chapter writing module configured to write the technical document according to the chapter title and the research idea based on the knowledge base. The method includes: ​ ​ 9. An agent-based technical document authoring system, comprising: ​ The technical document authoring apparatus of claim 8 and the knowledge base of any one of claims 1-4.

10. A storage medium for storing computer-executable instructions, the computer-executable instructions comprising instructions for: The computer executable instructions, when executed, implement the steps of the method of any one of claims 1-7.

Images