Article generation method and device, equipment and storage medium

Abstract

The invention provides an article generation method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence, in particular to the technical field of large models, agents and text processing. According to the specific implementation scheme, the method comprises the steps of obtaining a writing request, and identifying the writing request to obtain a writing requirement; respectively generating a draft outline and available materials according to the writing requirements; adopting the available materials to rewrite the draft outline to obtain a target article outline; and generating a target article according to the target article outline. Through the technical scheme, the article generation quality can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and specifically to the field of large models and text processing technology. Background Art

[0002] With the explosive development of AI big model technology, trending applications and products around big models are constantly attracting attention. Among the many big model application scenarios, intelligent writing is particularly prominent. It is exciting to be able to let the big model generate a generally good first draft of an article by inputting simple instructions. However, after a large number of trials, many intelligent writing products have difficulty in generating long articles, and the generated content often lacks depth and substance. Therefore, an efficient method for generating long articles is urgently needed. Summary of the invention

[0003] The present invention provides an article generation method, device, equipment and storage medium.

[0004] According to one aspect of the present disclosure, a method for generating an article is provided, the method comprising:

[0005] Obtaining a writing request, identifying the writing request, and obtaining writing requirements;

[0006] Generate draft outlines and available materials according to the writing requirements;

[0007] Rewriting the draft outline using the available materials to obtain a target article outline;

[0008] Generate a target article according to the target article outline.

[0009] According to another aspect of the present disclosure, there is provided an article generating device, the device comprising:

[0010] A writing requirement determination module, used to obtain a writing request, identify the writing request, and obtain a writing requirement;

[0011] An outline and material generation module, used to generate a draft outline and available materials according to the writing requirements;

[0012] A target outline determination module, used to rewrite the draft outline using the available materials to obtain a target article outline;

[0013] The target article generation module is used to generate the target article according to the target article outline.

[0014] According to another aspect of the present disclosure, there is provided an electronic device, the electronic device comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the article generation method described in any embodiment of the present disclosure.

[0018] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to execute the article generation method described in any embodiment of the present disclosure.

[0019] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the article generating method described in any embodiment of the present disclosure is implemented.

[0020] According to the technology disclosed in the present invention, the quality of article generation can be improved.

[0021] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0023] Figure 1 is a flow chart of an article generation method provided according to an embodiment of the present disclosure;

[0024] Figure 2 is a flowchart of another article generation method provided according to an embodiment of the present disclosure;

[0025] Figure 3 is a flowchart of another article generation method provided according to an embodiment of the present disclosure;

[0026] Figure 4 is a flowchart of another article generation method provided according to an embodiment of the present disclosure;

[0027] Figure 5 is a structural schematic diagram of an article generating device provided according to an embodiment of the present disclosure;

[0028] Figure 6 It is a block diagram of an electronic device used to implement the article generation method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0029] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] In addition, it should be noted that the collection, storage, use, processing, transmission, provision and disclosure of writing request-related data involved in the technical solution of the present invention are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0032] Figure 1 It is a flow chart of an article generation method provided according to an embodiment of the present disclosure. The method is applicable to situations in which an article is generated, and is particularly applicable to situations in which long articles are generated in more professional writing scenarios such as bidding documents, architectural plans, and research reports. Specifically, it is applicable to situations in which the text generated by a large model is not long enough and has low quality. The method can be executed by an article generation device, which can be implemented in software and / or hardware, and can be integrated into an electronic device that carries the article generation function, such as a server. It should be noted that the article generation method of the present disclosure can be executed by an intelligent agent based on a large model, wherein an intelligent agent refers to a computer program based on a large language model, which has the ability to plan and think, the ability to remember, the ability to use tool functions, and can autonomously complete a given task.

[0033] like Figure 1 As shown, the article generation method of this embodiment may include:

[0034] S101, obtaining a writing request, identifying the writing request, and obtaining writing requirements.

[0035] In this embodiment, the so-called writing request refers to a request for identifying writing requirements. The so-called writing requirements refer to requirements identified based on the writing request, which may include subject, style, word count, type, etc.

[0036] Specifically, the writing request input by the demander can be obtained, and the intelligent agent uses the big model intention recognition, that is, through the requirement recognition big model writing request to identify and obtain the writing requirement. For example, the writing request Query = "Help me write a market research on the big model, 10,000 words", through the requirement recognition big model can automatically identify the topic = "big model market research", the number of words = "about 10,000 words", the type = "market research", and the style = "formal".

[0037] It should be noted that in the present disclosure, a large model refers to a machine learning model or a deep learning model with a large parameter scale and complexity. The so-called requirement recognition large model refers to a large model used for writing requirement recognition.

[0038] It should be further explained that, after receiving a writing request, the agent can automatically plan and process to generate the target article.

[0039] S102, generate a draft outline and available materials according to writing requirements.

[0040] In this embodiment, the draft outline refers to the article outline initially generated based on the writing request. The available materials refer to the high-quality writing materials collected based on the writing request.

[0041] An optional way to generate a draft outline according to writing requirements is that the intelligent agent can search the text library based on the writing requirements, obtain related title content, combine the related title content, and obtain the draft outline. Alternatively, the intelligent agent can input the writing requirements into the big model based on the big model and output the draft outline.

[0042] Another optional way to generate usable materials according to writing requirements is that the intelligent agent uses the writing requirements as an index to retrieve related text content from a related text library as usable materials.

[0043] S103, using available materials, rewrite the draft outline to obtain the target article outline.

[0044] In this embodiment, the target article outline refers to the outline ultimately used for generating the target article.

[0045] In an optional way, the agent can determine the content associated with the draft outline from the available materials, and then expand the draft outline into chapters to obtain the target article outline. For example, the draft outline includes Chapter 1, Chapter 2, and Chapter 3. For example, for Chapter 1, the agent can determine the content associated with the chapter title corresponding to Chapter 1 from the available materials, generate Chapter 1.1, Chapter 1.2, and so on, and obtain the target article outline.

[0046] S104, generating a target article according to the target article outline.

[0047] In this embodiment, the target article refers to a long article that is finally generated, for example, a long article whose word count exceeds a set word count value, wherein the set word count value may be a value of more than 10,000 words.

[0048] In an optional manner, the agent can obtain the target article based on the article generation model and the target article outline, wherein the article generation model can be a large model.

[0049] The technical solution provided by the embodiment of the present disclosure obtains a writing request, identifies the writing request, obtains the writing requirements, and then generates a draft outline and available materials according to the writing requirements, and then rewrites the draft outline using the available materials to obtain the target article outline, and finally generates the target article according to the target article outline. The above technical solution generates the article after re-updating the draft outline to obtain the target article outline, so that the final target article outline has significant improvements in terms of content authenticity, logical integrity, novelty, etc., thereby making the obtained target article of higher quality.

[0050] Based on the above embodiment, as an optional method of the present disclosure, after generating the target article according to the target article outline, the method further includes: adding footnotes and / or references to the target article.

[0051] Specifically, the intelligent agent can add footnotes to the content of the target article and / or add references based on the materials used in the article generation process through the large model.

[0052] It is understandable that by adding footnotes and references to the target article, the article can be traced back to its source, thereby improving the credibility of the article.

[0053] Based on the above embodiment, as another optional method of the present disclosure, after generating the target article according to the target article outline, the method further includes: generating pictures and / or charts for the target article.

[0054] Specifically, the agent can generate pictures and / or charts for the chapter contents of the article chapter by chapter through the big model. For example, it can be to perform statistical comparative analysis on the chapter contents through the big model and draw charts such as line graphs, etc.; and / or, identify the names of objects in the chapter contents through the big model and configure pictures for the object names based on the image search method.

[0055] It is understandable that by making pictures and diagrams for the target article, it makes the article more beautiful and easier to understand.

[0056] Figure 2 1 is a flowchart of another article generation method provided according to an embodiment of the present disclosure. Based on the above embodiment, this embodiment further optimizes "generating available materials according to writing requirements" and provides an optional solution. Figure 2 As shown, the article generation method provided in this embodiment includes:

[0057] S201, obtaining a writing request, identifying the writing request, and obtaining writing requirements.

[0058] S202, generate a draft outline according to writing requirements.

[0059] S203, performing content retrieval according to writing requirements to obtain target content.

[0060] In this embodiment, the target content refers to excellent articles related to the writing requirements.

[0061] Specifically, the writing requirements can be used as an index to retrieve content from related article websites to obtain the target content. For example, the agent can use the large model to use the writing requirements as an index and use the retrieval tool to retrieve content from related article websites to obtain the target content.

[0062] S204, generating editors with different perspectives according to the target content.

[0063] Specifically, the intelligent agent uses the big model for role-playing, that is, the big model is used to generate editors with different perspectives according to the target content; for example, for a certain kind of food, different editors have different focuses on describing the food, such as historians pay more attention to the development and origin of the food, while gourmets pay more attention to the taste and cooking methods of the food.

[0064] S205, the editor asks questions in a circular manner from his / her perspective to obtain writing problems.

[0065] In this embodiment, the writing questions refer to relevant questions corresponding to the perspective.

[0066] Specifically, the intelligent agent can perform role-playing through a large model. For each editor, the intelligent agent can conduct a circular question-and-answer session from the editor's perspective to obtain writing problems.

[0067] S206, split and rewrite the writing questions, and perform content retrieval and summary to obtain usable materials.

[0068] The so-called available materials refer to the materials used to update the draft outline, which are presented in the form of a question-and-answer format.

[0069] Specifically, the intelligent agent can split and rewrite writing problems through the big model, retrieve and summarize content from relevant content websites, and obtain usable materials.

[0070] S207, using available materials, rewrite the draft outline to obtain the target article outline.

[0071] It should be noted that the intelligent agent can ensure a good and coherent writing idea by first generating a draft outline through the big model, but the outlines generated between the big models are often too empty. Therefore, the intelligent agent adopts the idea of ​​role-playing of the big model, allowing the big model to use retrieval tools to ask and answer questions based on writing requirements to collect high-quality writing materials, that is, usable materials, and then use the available materials combined with the draft outline to generate the final target article outline, which significantly improves the target article outline in terms of content authenticity, logical integrity, novelty and other aspects.

[0072] S208, generating a target article according to the target article outline.

[0073] The technical solution provided by the disclosed embodiment generates a draft outline according to writing requirements, performs content retrieval according to the writing requirements to obtain target content, generates editors with different perspectives according to the target content, uses the editors to ask questions in a loop from their perspectives to obtain writing problems, splits and rewrites the writing problems, performs content retrieval and summary to obtain available materials, uses the available materials to rewrite the draft outline to obtain the target article outline, and then generates the target article according to the target article outline. The above technical solution can obtain high-quality writing materials by collecting and summarizing relevant content based on writing requirements through role-playing, thereby providing support for the generation of the final article outline.

[0074] Figure 3 1 is a flowchart of another article generation method provided according to an embodiment of the present disclosure. Based on the above embodiment, this embodiment further optimizes "generating a target article according to the target article outline" and provides an optional implementation scheme. Figure 3 As shown, the article generation method provided in this embodiment includes:

[0075] S301, obtaining a writing request, identifying the writing request, and obtaining writing requirements.

[0076] S302, generate a draft outline and available materials according to writing requirements.

[0077] S303, using available materials, rewrite the draft outline to obtain the target article outline.

[0078] S304, extracting the structure of the target article outline to obtain the outline hierarchical structure.

[0079] In this embodiment, the outline hierarchical structure refers to the hierarchical structure of the target article outline, including chapter nodes, such as a tree structure; wherein the root node is the article title, the second-level node is the second-level title, and so on.

[0080] Specifically, the intelligent agent can extract the structure of the target article outline according to the outline catalog of the target article outline to obtain the outline hierarchical structure.

[0081] S305, generating an article according to the outline hierarchical structure to obtain a target article.

[0082] Specifically, the intelligent agent can traverse the chapter nodes in the outline hierarchy through the large model, generate the chapter content corresponding to each chapter node in parallel, and obtain the target article.

[0083] The technical solution provided by the embodiment of the present disclosure obtains a writing request, identifies the writing request, obtains the writing requirements, and then generates a draft outline and available materials according to the writing requirements, and then rewrites the draft outline using the available materials to obtain the target article outline, and finally extracts the structure of the target article outline to obtain the outline hierarchy structure, and generates the article according to the outline hierarchy structure to obtain the target article. The above technical solution generates the article through the outline hierarchy structure, which can ensure that the outline can be followed during the article generation process, thereby ensuring the quality of the article generation.

[0084] On the basis of the above embodiment, as an optional method of the present disclosure, an article is generated according to the outline hierarchy structure to obtain a target article, including: writing chapters from top to bottom according to the outline hierarchy structure to generate a first draft of the article; checking the first draft of the article to obtain the target article.

[0085] Among them, the first draft of the article refers to the article content generated for the first time based on the outline hierarchical structure.

[0086] Specifically, the intelligent agent can adopt a large model, traverse the outline hierarchy, and generate each chapter article in parallel according to each chapter node from top to bottom. For example, it can traverse each root chapter node in parallel, and generate the corresponding chapter content of the second-level chapter nodes under each root chapter node in parallel to obtain a first draft of the article. After that, the first draft of the article can be checked, such as format check, word count check, etc., to obtain the target article.

[0087] It is understandable that by generating articles from top to bottom, it can be ensured that the article generation process follows the outline, making the article generation process overall logical and not fanciful, and then checking the first draft of the article can ensure that the generated article is more accurate.

[0088] Figure 4 1 is a flowchart of another article generation method provided according to an embodiment of the present disclosure. Based on the above embodiment, this embodiment further optimizes "checking the first draft of the article to obtain the target article" and provides an optional implementation scheme. Figure 4 As shown, the article generation method of this embodiment may include:

[0089] S401, obtaining a writing request, identifying the writing request, and obtaining writing requirements.

[0090] S402, generate a draft outline and available materials according to writing requirements.

[0091] S403, using available materials, rewrite the draft outline to obtain the target article outline.

[0092] S404, extracting the structure of the target article outline to obtain the outline hierarchical structure.

[0093] S405, write chapters from top to bottom according to the outline hierarchy to generate the first draft of the article.

[0094] S406, merge and check the chapters in the first draft of the article to obtain the wrong and missing chapters.

[0095] In this embodiment, the erroneous and omitted chapters refer to chapters that are written incorrectly and / or omitted, including erroneous written chapters, omitted chapters, etc.

[0096] Specifically, the intelligent agent can compare the similarity of the chapter content of each chapter in the first draft of the article with the outline of the target article, and obtain the erroneous chapters based on the comparison results. For example, for a certain chapter, if the chapter similarity is 0, then the chapter is the omitted chapter in the erroneous chapters; if the similarity in the comparison result of the chapter is less than the set threshold, then the chapter is an erroneous chapter.

[0097] S407, rewrite the erroneous and missing chapters to obtain the target article.

[0098] Specifically, the intelligent agent can use the large model to regenerate the content of the erroneous and missing chapters in the first draft of the article to obtain the target article.

[0099] It should be noted that in the process of generating the target article, since the big model may write too much, too little or make mistakes, fault tolerance and big model reflection mechanisms are added to merge and check the first draft of the article, so that the article generated by the big model strictly follows the article outline.

[0100] The technical solution provided by the disclosed embodiment generates a draft outline and available materials according to writing requirements, and then rewrites the draft outline using the available materials to obtain a target article outline, and extracts the structure of the target article outline to obtain an outline hierarchy structure, and writes chapters from top to bottom according to the outline hierarchy structure to generate a first draft of the article, and then merges and checks the chapters in the first draft of the article to obtain erroneous and missing chapters, and rewrites the erroneous and missing chapters to obtain the target article. The above technical solution can ensure the accuracy of the target article by checking the first draft of the article for errors and omissions to regenerate the corresponding chapter content, thereby improving the quality of the article.

[0101] On the basis of the above embodiment, as an optional method of the present disclosure, the first draft of the article is checked to obtain the target article, including: checking whether the chapters in the first draft of the article contain specified materials according to the outline hierarchy structure; if so, updating the first draft of the article according to the specified materials to obtain the target article.

[0102] The designated material refers to the special writing material or writing requirement specified by the demander, which can be adopted. The demander refers to the party that has the demand to generate the article.

[0103] Specifically, the intelligent agent can check whether the specified material exists in each secondary chapter node in the outline hierarchy structure. If so, the specified material is used to replace the chapter content in the corresponding chapter node to obtain the target article.

[0104] It is understandable that by adding designated material checks, the personalized needs of the demander for chapters can be met.

[0105] Based on the above embodiment, as an optional method of the present disclosure, the first draft of the article is checked to obtain a target article, including: checking the word count of the first draft of the article from bottom to top according to the outline hierarchy structure to obtain a check result; updating the first draft of the article according to the check result to obtain the target article.

[0106] Specifically, the agent can check the word count from the leaf chapter node from bottom to top according to the outline hierarchy. For each leaf chapter node, if the word count of the chapter content corresponding to the leaf chapter node does not meet the word count requirement, the chapter content of the leaf chapter node is regenerated. For example, writing materials can be added, and the leaf chapter node is regenerated through the big model until the word count requirement is met to obtain the target article. Adding writing materials can allow the big model to output more key words to meet the word count requirement.

[0107] It is understandable that by checking the word count of the first draft of the article from the bottom up, it can be ensured that the resulting article is a long article.

[0108] Figure 5 This is a schematic diagram of the structure of an article generation device provided according to an embodiment of the present disclosure. The present disclosure is applicable to situations in which articles are generated, and is particularly applicable to situations in which long articles are generated in more professional writing scenarios such as bidding documents, building plans, and research reports. Specifically, it is applicable to situations in which the length of text generated by a large model is insufficient and the quality is low. The device can be implemented in software and / or hardware, and can be integrated into an electronic device that carries the article generation function, such as a server. Figure 5 As shown, the article generating device 500 includes:

[0109] The writing requirement determination module 501 is used to obtain the writing request, identify the writing request, and obtain the writing requirement;

[0110] An outline and material generation module 502 is used to generate a draft outline and available materials according to writing requirements;

[0111] A target outline determination module 503 is used to rewrite the draft outline using available materials to obtain a target article outline;

[0112] The target article generating module 504 is used to generate the target article according to the target article outline.

[0113] The technical solution provided by the embodiment of the present disclosure obtains a writing request, identifies the writing request, obtains the writing requirements, and then generates a draft outline and available materials according to the writing requirements, and then rewrites the draft outline using the available materials to obtain the target article outline, and finally generates the target article according to the target article outline. The above technical solution generates the article after re-updating the draft outline to obtain the target article outline, so that the final target article outline has significant improvements in terms of content authenticity, logical integrity, novelty, etc., thereby making the obtained target article of higher quality.

[0114] Furthermore, the outline and material generation module 502 is specifically used for:

[0115] Conduct content retrieval according to writing requirements to obtain target content;

[0116] Generate editors with different perspectives based on the target content;

[0117] The editor asks questions in a circular manner from his / her perspective to obtain writing questions;

[0118] Split and rewrite writing problems, and conduct content search and summary to obtain usable materials.

[0119] Furthermore, the target article generation module 504 includes:

[0120] A hierarchical structure extraction unit is used to extract the structure of the target article outline to obtain the outline hierarchical structure;

[0121] The target article generation unit is used to generate an article according to the outline hierarchical structure to obtain a target article.

[0122] Furthermore, the target article generation unit includes:

[0123] The article draft generation subunit is used to write chapters from top to bottom according to the outline hierarchy to generate the article draft;

[0124] The target article determination subunit is used to check the first draft of the article to obtain the target article.

[0125] Furthermore, the target article determines subunits, which are specifically used for:

[0126] Merge and check the chapters in the first draft of the article to find out the missing chapters;

[0127] Rewrite the wrong and missing chapters to obtain the target article.

[0128] Furthermore, the target article determines subunits, which are specifically used for:

[0129] According to the outline hierarchy, check whether the chapters in the first draft of the article contain the specified materials;

[0130] If it exists, the first draft of the article is updated according to the specified material to obtain the target article.

[0131] Furthermore, the target article determines subunits, which are specifically used for:

[0132] According to the outline hierarchy, the word count of the target article is checked from bottom to top to obtain the check result;

[0133] The first draft of the article is updated according to the check results to obtain the target article.

[0134] Furthermore, the device also includes a footnote document adding module, which is used to:

[0135] After generating the target article according to the target article outline, add footnotes and / or references to the target article.

[0136] Furthermore, the device also includes a graph generation module, which is used to:

[0137] After generating the target article according to the target article outline, generate pictures and / or charts for the target article.

[0138] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0139] Figure 6 It is a block diagram of an electronic device used to implement the article generation method of the embodiment of the present disclosure. Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement an embodiment of the present disclosure 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 processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0140] like Figure 6 As shown, the electronic device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0141] Multiple components in the electronic device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the electronic device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0142] The computing unit 601 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 601 performs the various methods and processes described above, such as the article generation method. For example, in some embodiments, the article generation method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of the article generation method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform the article generation method by any other appropriate means (e.g., by means of firmware).

[0143] Various implementations 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 chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0144] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0145] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0146] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer 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 can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the 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 acoustic input, voice input, or tactile input).

[0147] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0148] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0149] Artificial intelligence is a discipline that studies how to use computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). It includes both hardware-level and software-level technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing; artificial intelligence software technologies mainly include computer vision technology, speech recognition technology, natural language processing technology, machine learning / deep learning technology, big data processing technology, knowledge graph technology, and other major directions.

[0150] Cloud computing refers to a technology system that uses network access to elastically scalable shared physical or virtual resource pools. Resources can include servers, operating systems, networks, software, applications, and storage devices, and can be deployed and managed on demand and in a self-service manner. Cloud computing technology can provide efficient and powerful data processing capabilities for technical applications such as artificial intelligence and blockchain, as well as model training.

[0151] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0152] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for generating an article, comprising: Obtaining a writing request, identifying the writing request, and obtaining writing requirements; Generate draft outlines and available materials according to the writing requirements; Rewriting the draft outline using the available materials to obtain a target article outline; Generate a target article according to the target article outline.

2. The method according to claim 1, wherein: Generate available materials according to the writing requirements, including: Perform content retrieval according to the writing requirements to obtain target content; Generating editors with different perspectives according to the target content; The editor asks questions in a circular manner from his / her perspective to obtain writing questions; The writing problems are split and rewritten, and content retrieval and summary are performed to obtain usable materials.

3. The method according to claim 1, wherein: Generating a target article according to the target article outline includes: Extract the structure of the target article outline to obtain the outline hierarchical structure; An article is generated according to the outline hierarchical structure to obtain a target article.

4. The method according to claim 3, wherein: Generating an article according to the outline hierarchical structure to obtain a target article includes: According to the outline hierarchy, write chapters from top to bottom to generate a first draft of the article; The draft of the article is checked to obtain a target article.

5. The method according to claim 4, wherein: The first draft of the article is checked to obtain a target article, including: Conduct a combined check on the chapters in the first draft of the article to find out the wrong and missing chapters; Rewrite the erroneous and missing chapters to obtain the target article.

6. The method according to claim 4, wherein: The first draft of the article is checked to obtain a target article, including: According to the outline hierarchical structure, checking whether the chapters in the first draft of the article contain designated materials; If it exists, the article draft is updated according to the specified material to obtain the target article.

7. The method according to claim 4, wherein: The first draft of the article is checked to obtain a target article, including: According to the outline hierarchical structure, the word count of the first draft of the article is checked from bottom to top to obtain a check result; The first draft of the article is updated according to the checking result to obtain the target article.

8. The method according to claim 1, wherein: After generating the target article according to the target article outline, the method further includes: Add footnotes and / or references to the target article.

9. The method according to claim 1, wherein: After generating the target article according to the target article outline, the method further includes: Generate pictures and / or diagrams for the target article.

10. An article generating device, comprising: A writing requirement determination module, used to obtain a writing request, identify the writing request, and obtain a writing requirement; An outline and material generation module, used to generate a draft outline and available materials according to the writing requirements; A target outline determination module, used to rewrite the draft outline using the available materials to obtain a target article outline; The target article generation module is used to generate the target article according to the target article outline.

11. The device according to claim 10, wherein: The outline and material generation module is specifically used for: Perform content retrieval according to the writing requirements to obtain target content; Generating editors with different perspectives according to the target content; The editor asks questions in a circular manner from his / her perspective to obtain writing questions; The writing problems are split and rewritten, and content retrieval and summary are performed to obtain usable materials.

12. The device according to claim 10, wherein: The target article generation module includes: A hierarchical structure extraction unit is used to extract the structure of the target article outline to obtain the outline hierarchical structure; The target article generating unit is used to generate an article according to the outline hierarchical structure to obtain a target article.

13. The device according to claim 12, wherein: The target article generating unit comprises: An article draft generation subunit is used to write chapters from top to bottom according to the outline hierarchy structure to generate an article draft; The target article determination subunit is used to check the first draft of the article to obtain the target article.

14. The device according to claim 13, wherein: The target article determines the subunit, which is specifically used for: Conduct a combined check on the chapters in the first draft of the article to find out the wrong and missing chapters; Rewrite the erroneous and missing chapters to obtain the target article.

15. The device according to claim 13, wherein: The target article determines the subunit, which is specifically used for: According to the outline hierarchical structure, checking whether the chapters in the first draft of the article contain designated materials; If it exists, the article draft is updated according to the specified material to obtain the target article.

16. The device according to claim 13, wherein: The target article determines the subunit, which is specifically used for: According to the outline hierarchical structure, the word count of the target article is checked from bottom to top to obtain a check result; The first draft of the article is updated according to the checking result to obtain the target article.

17. The device according to claim 10, wherein: The device also includes a footnote document adding module, which is used to: After generating the target article according to the target article outline, footnotes and / or references are added to the target article.

18. The device according to claim 10, wherein: The device also includes a graph generation module, which is used to: After generating the target article according to the target article outline, a picture and / or a chart is generated for the target article.

19. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the article generation method according to any one of claims 1 to 9.

20. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to enable a computer to execute the article generation method according to any one of claims 1-9.

21. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the article generation method according to any one of claims 1 to 9.

Images