Document information display method and device, equipment and storage medium
By using mind maps and generative language models in the application to automatically categorize and store new documents, the problem of low document management efficiency in multi-document applications is solved, achieving rapid categorization and efficient human-computer interaction.
Patent Information
- Application Number
- CN202410405013.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-07
- Publication Date
- 2025-10-21
AI Technical Summary
In multi-document applications, as the number of documents increases, the list of documents becomes more numerous and hierarchical, requiring users to perform a lot of interactive operations to locate the document's storage location, resulting in decreased document management efficiency and low human-computer interaction efficiency.
By displaying mind maps in the application, the content of newly added documents is analyzed using a generative language model, automatically determining their node positions in the existing mind maps, and adding document information at those positions, thus achieving rapid categorization and storage.
It improves document management efficiency, reduces user interaction, enhances human-computer interaction efficiency, and enables newly added documents to be quickly categorized into appropriate locations without requiring users to manually select storage locations.
Smart Images
Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] The present application relates to the field of human-computer interaction, and in particular to a method, apparatus, device, and storage medium for displaying document information. Background Art
[0002] In multi-document applications, users can create, store, and manage documents, organizing them into organized categories for easy search and query. For example, in note-taking applications, users can create, edit, and manage notes to organize and record information from their daily lives and work, improving efficiency and productivity.
[0003] In related art, multiple documents are managed using a list structure, that is, documents are organized and categorized in a hierarchical manner. For example, different folders are created on the application homepage to categorize documents; subfolders can be nested within folders to further subdivide documents, and so on.
[0004] However, as the number of documents increases, the number of list items required to manage them increases, and the hierarchy deepens. This requires a lot of interaction for users to locate the document storage location, making it difficult to quickly categorize documents. This reduces the efficiency of document management and the efficiency of human-computer interaction, which is a problem that needs to be solved. Summary of the Invention
[0005] The embodiments of the present application provide a method, apparatus, device, and storage medium for displaying document information, which can provide a method for quickly classifying and storing newly added documents in an application, thereby improving document management efficiency and human-computer interaction efficiency. The technical solution is as follows:
[0006] According to one aspect of the present application, a method for displaying document information is provided, the method comprising:
[0007] Displaying a user interface of the application, wherein the user interface includes a first mind map, and nodes in the first mind map are used to display document information of at least one existing document;
[0008] In response to an add operation on a target document in the application, obtaining document content of the target document;
[0009] Document information of the target document is added and displayed on a target node in the first mind map. The node position of the target node in the first mind map is determined by the application based on document content analysis of the target document.
[0010] According to another aspect of the present application, a device for displaying document information is provided, the device comprising:
[0011] a display module, configured to display a user interface of the application, wherein the user interface includes a first mind map, and nodes in the first mind map are configured to display document information of at least one existing document;
[0012] an acquisition module, configured to acquire the document content of the target document in response to an add operation on the target document in the application;
[0013] The display module is used to add and display document information of the target document on the target node in the first mind map, and the node position of the target node in the first mind map is determined by the application based on the document content analysis of the target document.
[0014] According to another aspect of the present application, a computer device is provided, comprising a processor and a memory, wherein the memory stores at least one program, and the at least one program is loaded and executed by the processor to implement the document information display method as described above.
[0015] According to another aspect of the present application, a computer-readable storage medium is provided, in which at least one program is stored. The at least one program is loaded and executed by the processor to implement the document information display method as described above.
[0016] According to another aspect of the present application, a computer program product is provided, wherein at least one program segment is stored in the computer program product, and the at least one program segment is loaded and executed by the processor to implement the document information display method as described above.
[0017] The beneficial effects of the technical solutions provided in the embodiments of the present application include at least:
[0018] The application analyzes newly added documents, automatically determines the node position of the newly added documents in the mind map based on existing documents, and quickly classifies and stores the newly added documents, thereby improving document management efficiency. It also eliminates the need for users to manually select the storage location of newly added documents, reducing user interaction operations and helping to improve human-computer interaction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.
[0020] Figure 1It is a conceptual diagram of the mind map;
[0021] Figure 2 This is a flowchart of an interface diagram of a method for displaying document information provided by an exemplary embodiment of the present application;
[0022] Figure 3 is a structural block diagram of a computer system provided by an exemplary embodiment of the present application;
[0023] Figure 4 is a flowchart of a method for displaying document information provided by an exemplary embodiment of the present application;
[0024] Figure 5 is a flowchart of a method for displaying document information provided by an exemplary embodiment of the present application;
[0025] Figure 6 This is a schematic diagram of an interface of a method for displaying document information provided by an exemplary embodiment of the present application;
[0026] Figure 7 is a schematic diagram of a mind map provided by an exemplary embodiment of the present application;
[0027] Figure 8 This is a schematic diagram of an interface of a mind map optimization method provided by an exemplary embodiment of the present application;
[0028] Figure 9 is a flowchart of a mind map optimization method provided by an exemplary embodiment of the present application;
[0029] Figure 10 This is a schematic diagram of an interface for adding a target document provided by an exemplary embodiment of the present application;
[0030] Figure 11 This is a schematic diagram of an interface for adding a target document provided by an exemplary embodiment of the present application;
[0031] Figure 12 is a flowchart of a method for displaying document information provided by an exemplary embodiment of the present application;
[0032] Figure 13 This is a structural block diagram of a device for displaying document information provided by an exemplary embodiment of the present application;
[0033] Figure 14 It is a block diagram of a computer device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION
[0034] In order to make the objectives, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.
[0035] First, the terms that may appear in this application are explained.
[0036] Mind Map: A mind map is a graphical tool used to represent the relationships and hierarchical structure between information such as ideas, concepts, words, or tasks. A mind map places a central idea or concept at the center, and then branches out related ideas, subtopics, tasks, or other related information, creating a diagram that resembles a tree or brain cell structure.
[0037] Figure 1 A schematic diagram of a mind map is shown. Figure 1 Node 0 in the mind map is the central theme, also known as the core concept, central idea, or root node. It is the core concept of the mind map. The mind map has three branches, indicated by nodes 1, 2, and 3, respectively. Node 1 has two sub-branches, indicated by nodes 4 and 5. Node 1 can be called the parent node of nodes 4 and 5, and nodes 4 and 5 can be called child nodes of node 1. Nodes with the same parent node are siblings, meaning that nodes 4 and 5 are siblings.
[0038] A Generative Language Model (GLM) is a machine learning model used to generate language data. It can be applied to a variety of tasks, including code generation, machine translation, question-answering systems, and text summarization. Given an input sentence, the GLM generates a corresponding response sentence based on the input sentence.
[0039] Large Language Model (LLM): This is a type of generative language model used to process and understand natural language text. It is trained on large amounts of text data and can perform tasks such as text summarization, translation, question-answering, and code generation. LLMs are based on deep neural networks and are language models whose model parameters exceed a first threshold, or whose training data exceeds a second threshold, or whose computational complexity exceeds a third threshold. These thresholds are all in the hundreds of billions.
[0040] Generative Pre-trained Transformer (GPT): This is a generative language model based on the Transformer architecture, used to generate natural language text. The key feature of GPT models is that they are pre-trained on a large scale to learn the grammar, semantics, and context of natural language. These models can be used for text generation, text classification, question-answering systems, language understanding, and various natural language processing tasks.
[0041] Retrieval Augmented Generation (RAG): is a natural language processing technique that combines information retrieval and generative language models. It combines an information retrieval component with a text generation model. This method enhances the quality and relevance of the generative model's output by first retrieving relevant information from a large document database and then feeding this information as context into the generative model.
[0042] Figure 2 Taking the note application as an example, an interface diagram of a method for displaying document information provided by an exemplary embodiment of the present application is shown. Figure 2 As shown, the user interface 10 of the note application includes a note editing area 11 and a mind map area 12. Users edit notes through the note editing area 11, for example, by directly entering text in the note preview area 13, or performing audio input or speech-to-text operations through the language input control 14, or inserting images through the image insertion control 15, or selecting files to be inserted into the note through the file insertion control 16, etc. The mind map area 12 displays a mind map, which is used to display the document information of existing documents (i.e., existing notes) in the note application. It uses node 17 as the central theme and displays the document information of existing documents through branches and child nodes. For example, taking the branch where node 18 is located as an example, node 18 has two child nodes, namely node 19 and node 20; node 19 and node 20 respectively display the document information of the corresponding note, and node 18 displays the document information that summarizes the above two notes. It should be noted that node 18 is also called the parent node of node 19 and node 20, and node 19 and node 20 are called sibling nodes.
[0043] After the user finishes editing a note, they publish it by clicking the send control 21. A pop-up window 22 appears in the note editing area 11, informing the user that the note application is analyzing and categorizing the note's contents. After the note application completes the analysis, it determines that the note's contents are an expansion of node 23. Therefore, a child node 24 of node 23 is added to the mind map in the mind map area 12. This child node 24 displays the document information for the newly added note.
[0044] Figure 3 The block diagram of a computer system provided by an exemplary embodiment of the present application is shown. The computer system includes: a terminal 120 and a server 140.
[0045] Terminal 120 can be at least one of a smartphone, a game console, a desktop computer, a tablet computer, an e-book reader, an MP3 (Moving Picture Experts Group Audio Layer III) player or an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop computer, an intelligent robot, and a self-service payment device. Terminal 120 has installed and runs an application that supports document management. For example, the application can be a file management application, a note-taking application, a mind mapping application, a system application, a search engine application, a collaboration and sharing application, a social application, a programming application, and the like.
[0046] In an illustrative example, a human-computer interaction (HCI) interface for document management is displayed on the terminal 120 , and a user creates, edits, stores, and manages documents through the HCI interface.
[0047] In another illustrative example, the terminal 120 displays an application running interface, which displays a mind map determined based on an existing document, and the mind map is used to display document information of the existing document; the user adds a document on the interface, and the terminal displays a mind map with the document information of the added document added.
[0048] The terminal 120 is connected to the server 140 via a wireless network or a wired network.
[0049] Server 140 includes at least one of a single server, multiple servers, a cloud computing platform, and a virtualization center. Server 140 provides backend services for clients. Optionally, server 140 performs primary computing tasks, while the client performs secondary computing tasks; alternatively, server 140 performs secondary computing tasks, while the client performs primary computing tasks; alternatively, server 140 and the client utilize a distributed computing architecture for collaborative computing.
[0050] In an illustrative example, server 140 includes a processor, a database, and a user-oriented input / output interface (I / O interface). The processor is used to load instructions stored in server 140, process data from the database and trigger services, and run the generative language model. The database is used to store data related to document management.
[0051] Those skilled in the art will appreciate that the number of the terminals 120 and servers 140 may be greater or lesser. For example, there may be only one terminal 120 or server 140, or there may be dozens, hundreds, or even more terminals 120 or servers 140. The embodiments of the present application do not limit the number or device types of the terminals or servers.
[0052] Figure 4 A flowchart of a method for displaying document information provided by an exemplary embodiment of the present application is shown. Figure 3 The method is described by way of example using a terminal or server.
[0053] Step 220: Display the user interface of the application;
[0054] The application described in this embodiment is a software tool that provides document management capabilities, such as creating, storing, managing, tracking, protecting, and distributing documents. This application typically provides a variety of features to support file organization, search, version control, and collaboration. This application can be a cloud-based service or locally installed software.
[0055] For example, the application can be a note-taking application, and the documents in the note-taking application are notes. Users can quickly record, organize, and retrieve various forms of notes in the note-taking application; the note-taking application can provide users with note management functions such as storage, classification, search, labeling, and sharing of notes.
[0056] Exemplarily, a user interface of an application is displayed on a terminal; the user interface includes a first mind map, wherein nodes in the first mind map are used to display document information of at least one existing document. Specifically, the first mind map is a mind map for displaying document information of existing documents in the application; that is, the first mind map presents the document information of the existing documents in the form of a mind map based on the document content of the existing documents and in accordance with logical relationships.
[0057] Optionally, the structural information of the first mind map includes node labels and node content of each node, and the node content includes at least one of a node title, a node keyword, and a node sub-layer name. Optionally, the structural information of the first mind map is stored in JSON format.
[0058] In some embodiments, the node numbering convention is "N" + "level 1 node" + "level 2 node" + "level 3 node" + ..., and so on, where "N" represents the root node. That is, the node numbering includes information about the node's level, parent node, child node, and sibling nodes. For example, node N-2 is the parent node of node N-2-1; for another example, node N-3-2 and node N-3-3 are sibling nodes.
[0059] In some embodiments, each child node at the lowest level in the first mind map is used to display document information of an existing document, and each parent node is used to display comprehensive document information of at least one existing document corresponding to all its child nodes; for example, Figure 2 Node 19 and node 20 in respectively display the document information of a corresponding existing document, and its parent node 18 displays the comprehensive document information of the two existing documents corresponding to node 19 and node 20.
[0060] In some embodiments, document information for an existing document corresponding to a node in the first mind map is displayed in the form of a node card. Optionally, the node card also displays the document's creation time, last edit time, or last open time. Optionally, the node card also displays the number of images, links, files, audio, or video included in the document. Optionally, a user can click a node card to jump to the display interface of the existing document corresponding to the node.
[0061] For example, Figure 2 The node card of node 19 in the example corresponds to a document about "XYZ Company Website Design". The node card displays the title of the document, part of the document content, creation time, and the number of pictures / links / audio contained in the document. Users can jump to view the document by clicking on the node card. For another example, Figure 2 The document corresponding to node 18 in the figure is generated based on the documents "XYZ Company Website Design" and "ABC Company Website Design" corresponding to its child nodes, which include the document information of the two documents; the node card displays the title, abstract and the number of pictures / links / videos contained in the document, which are generated by integrating the document information of the two documents. Users can jump to view the document by clicking on the node card, that is, view the comprehensive information of the two documents corresponding to nodes 19 and 20 respectively.
[0062] A document is a file containing information such as text, charts, and data. A document can be created within an application (e.g., a notebook or quick note in a note-taking application), imported from outside the application, or collected into the application from a third-party platform. Documents can be in plain text (.txt), rich text format (.rtf), Word documents (.doc, .docx), Excel spreadsheets (.xls, .xlsx), PDF documents (.pdf), web page formats (.html), e-book formats (.epub, .mobi), and more. Each document in an application has a different storage location.
[0063] Document information includes at least one of a document title, a document summary, and document keywords. The document title can be a user-defined title when adding a document, or a title automatically generated by an application based on the document content. The document summary can be a user-defined summary when adding a document, or a summary of all or part of the document content, such as the first 20 words of the document, or a summary of the document content automatically generated by an application based on the document content. The document keywords can be user-defined keywords when adding a document, or keywords automatically generated by an application based on the document content.
[0064] In some embodiments, based on the document content of an existing document, a generative language model is used to generate document information of the existing document; that is, the document content of the existing document is input into the generative language model to generate at least one of the document title, document summary, and document keywords of the existing document, which is then displayed in a node of the first mind map.
[0065] It should be noted that this application does not limit the content of the document information of the existing documents displayed in the nodes in the first mind map, the display method of the document information of the existing documents in the nodes, and the scope of the document information displayed in each node.
[0066] Step 240: In response to the adding operation of the target document in the application, obtain the document content of the target document;
[0067] The target document is a new document to be added to the application. This can be a document that the user has created within the application; a document that has been imported from another application; or a document that the user has saved to the application from a webpage or third-party application.
[0068] Exemplarily, in response to a target document creation operation in an application, the document content of the created target document is obtained; in response to an import operation on a target document in an application, the document content of the imported target document is obtained; in response to a target document collection operation in an application, the document content of the collected target document is obtained.
[0069] Step 260: Add document information of the target document to the target node in the first mind map.
[0070] The target node is used to display the document information of the newly added target document. The target node can be an existing node in the first mind map, or a newly added node in the first mind map. The node position of the target node in the first mind map is determined by the application based on the document content analysis of the target document.
[0071] In some embodiments, a generative language model is used to determine the node position of the target node in the first mind map. That is, the node position of the target node in the first mind map is determined based on the document content of the target document and the structural information of the first mind map using a first generative language model. The structural information of the first mind map includes the node label and node content of each node, and the node content includes at least one of a node title, a node keyword, and a node sub-layer name.
[0072] In some embodiments, the document content of the target document, the structural information of the first mind map, and the first prompt sentence are input into the first generative language model to determine the node position of the target node in the first mind map; the first prompt sentence is used to instruct the generative language model to determine the node position of the target node in the first mind map.
[0073] The first generative language model is a generative language model used to determine the node position of the target node in the first mind map based on the document content of the target document. The first generative language model can be a public cloud-deployed generative language model, such as a GPT model or a Large Language Model MetaArtificial Intelligence (LLaMA) model; or the first generative language model can be a privately deployed generative language model.
[0074] Exemplarily, based on the analysis of the application program, determining the node position of the target node in the first mind map includes at least one of the following situations:
[0075] In the case where the document content of the target document is a continuation of the node content of the first node in the first mind map, additionally displaying the document information of the target document at the first node;
[0076] When the document content of the target document is the expansion of the node content of the second node in the first mind map, the document information of the target document is displayed at a newly added child node of the second node;
[0077] If the document content of the target document has a similar theme to the node content of the third node in the first mind map but has different content, the document information of the target document is displayed at a newly added node at the same level as the third node;
[0078] If the document content of the target document is not related to the node content of any node in the first mind map, display the document information of the target document at the newly added node;
[0079] The node content includes at least one of a node title, a node keyword, and a node sub-layer name.
[0080] In some embodiments, based on the document content of the target document, the document information of the target document is generated based on a third generative language model; wherein the document information of the target document includes at least one of a document title, a document summary, and document keywords. That is, the document information of the target document displayed in the first mind map can be customized by the user; or, the document information of the target document displayed in the first mind map can also be generated by the third generative language model based on the document content of the target document; or, only the document information not defined by the user is generated using the third generative language model. The third generative language model is a model for generating document information; the third generative language model can be the same model as the first generative language model, or a different model from the first generative language model. This application does not impose any restrictions on this.
[0081] In some embodiments, if the first mind map meets the optimization conditions, a second optimized mind map is displayed on the user interface of the application. The second mind map is a mind map optimized based on the second generative language model. The second generative language model is a generative language model used to optimize the mind map. The second generative language model and the first generative language model can be different or the same model. The second generative language model and the third generative language model can be different or the same model.
[0082] In some embodiments, whether a mind map needs to be optimized is determined by determining whether there are multiple nodes in the mind map whose node contents belong to the same type. That is, if there are at least two nodes in the first mind map whose node contents belong to the same type, the at least two nodes are merged and displayed in the second mind map.
[0083] In some embodiments, a second generative language model is used to determine optimizable nodes in the first mind map. Node labels of the optimizable nodes in the first mind map are determined based on the second generative language model based on structural information of the first mind map; wherein the structural information of the first mind map includes node labels and node content of each node, and the node content includes at least one of a node title, a node keyword, and a node sub-layer name.
[0084] In some embodiments, the structural information of the first mind map and the second prompt are input into a second generative language model to determine the node labels of the optimizable nodes in the first mind map. The second prompt is used to instruct the generative language model to determine the optimizable nodes in the mind map. After determining the node labels of the optimizable nodes in the first mind map, the optimizable nodes are merged and displayed, and a parent node of the optimizable nodes is generated.
[0085] It should be noted that the step of optimizing the first mind map into the second mind map can be performed at different times depending on the actual situation. For example, when the user first opens the application each day, it is determined whether there are nodes in the first mind map that can be optimized; for another example, each time the user opens the application, it is determined whether there are nodes in the first mind map that can be optimized; for another example, each time the user adds a new document, it is determined whether there are nodes in the first mind map that can be optimized, and so on. This application does not impose any restrictions on this.
[0086] In some embodiments, users search for document content or ask questions in the application; the application uses vector-based semantic search and RAG technology to replace traditional keyword search, that is, the keywords used by users for search do not need to be exactly the same as the original text, which can help users better find the required document content or get better answers, thereby improving the efficiency of document use.
[0087] In summary, the method provided in the embodiment of the present application displays a first mind map including document information of existing documents. After a new document is added, the document information of the newly added document is added to the corresponding node of the first mind map. This allows the newly added document to be quickly classified into the appropriate location. The user can edit the document quickly and stress-free without having to consider where the document should be stored, thereby improving the efficiency of document use in the application. This method can improve document management efficiency, especially when there are many documents, the user cannot remember all the document contents, and it is difficult to quickly classify the documents.
[0088] In addition, newly added documents can be automatically classified and archived, eliminating the need for users to perform multiple interactive operations between levels to classify documents, thereby improving the efficiency of human-computer interaction.
[0089] In addition, using mind mapping to organize documents makes the documents more structured and systematic, the document content is clearer and easier to remember, and users do not need to search back and forth for files in multi-level list folders, further improving document management and usage efficiency.
[0090] Figure 5 A flowchart of a method for displaying document information provided by an exemplary embodiment of the present application is shown. Figure 3 The terminal or server shown is executed as an example. In this embodiment, the application is a note application as an example for explanation, that is, the document in this embodiment refers to the note. This embodiment is explained by optimizing the mind map when the user opens the note application for the first time every day. The method includes:
[0091] Step 320: Display the user interface of the note application;
[0092] In this embodiment, a note-taking application is used as an example. That is, a user interface of the note-taking application is displayed on the terminal. The user interface of the note-taking application includes a first mind map, and nodes in the first mind map are used to display document information of at least one existing document (i.e., an existing note).
[0093] In some embodiments, the user interface of the note-taking application includes a note editing area and a mind map area. Figure 2 As shown, the user interface of the note-taking application includes a note editing area 11 and a mind map area 12. Users can create and edit notes in the note editing area 11, while the mind map area 12 is used to display a first mind map containing document information of existing documents. The first mind map is based on the document content of existing documents and is determined according to logical relationships. By displaying the document information of existing documents in the form of a mind map, existing documents can be clearly organized and listed, making it easier for users to quickly find the required documents based on the mind map.
[0094] In some embodiments, the nodes in the first mind map correspond to existing documents; the user can jump to the editing interface of the existing document corresponding to the node by clicking on the node. Figure 6 As shown, the first mind map 50 includes multiple nodes, and each node displays the document information of the existing document corresponding to the node in the form of a node card; when the user clicks the node card 51, the user interface jumps to display the editing interface of the existing document corresponding to the node card 51, and the user can perform editing, sharing and other operations on the editing interface.
[0095] In some embodiments, a first mind map is generated based on the document content of an existing document using a generative language model. The generative language model may be the same as at least one of the first, second, and third generative language models; or the generative language model may be different from the first, second, and third generative language models. The generative language model generates keywords for each document based on the document content of the existing document, and generates the first mind map through comparison and matching.
[0096] In some embodiments, the structural information of the first mind map includes the node label and node content of each node, and the node content includes at least one of the node title, node keyword, and node sub-layer name. Optionally, the structural information of the first mind map is generated based on a generative language model; for example, the structural information of the first mind map is generated based on a fourth generative language model; that is, based on the document content of an existing document, at least one of the title, keywords, and sub-layer name of the existing document is generated based on the fourth generative language model. The first generative language model, the second generative language model, the third generative language model, and the fourth generative language model can be different models, or at least two of them can be the same model; this is not limited in this application.
[0097] After obtaining the structural information of the first mind map, it is stored in a JSON structure. The example is as follows:
[0098] {"title":"My Notes","sublayer":[
[0099] {"title":"Application Development Progress Update","nodeid":"n-1","keywords":["Note App","Login Module","User Interface Optimization","Data Encryption","Performance Testing","User Feedback","Push Notification","Stress Testing","Personalized Recommendation","Project Management"],"sublayer":[]},
[0100] {"title":"RAG technology learning summary","nodeid":"n-2","keywords":["RAG technology","natural language processing","text generation","dense vector retrieval","pre-trained language model","question answering system","article writing","generation quality","NLP task","model architecture"],"sublayer":[
[0101] {"title":"RAG Model Architecture Analysis","nodeid":"n-2-1","keywords":["RAG Model","Natural Language Processing","Retrieval","Generator","Transformer","Dense Vector Retrieval","Seq2Seq","BART","T5","End-to-End Training"],"sublayer":[]},
[0102] {"title":"PRACTICAL APPLICATION OF RAG TECHNOLOGY","nodeid":"n-2-2","keywords":["RAG technology","actual project application","information retrieval","text generation","question-answering system","content recommendation","automatic summarization","dataset preparation","retrieval optimization","generator adjustment"],"sublayer":[]}
[0103] ]},
[0104] {"title":"Beijing New Year Travelogue","nodeid":"n-3","keywords":["Beijing","New Year Holiday","The Palace Museum","The Great Wall","Nanluoguxiang","National Centre for the Performing Arts","Peking Opera","Festive Atmosphere","Ancient and Modern","Historical Sedimentation"],"sublayer":[]}
[0105] ]}
[0106] Here, title refers to the node title, keywords refers to the node keywords, sublayer refers to the node sublayer name (for example, a paragraph subheading in a document), and nodeid refers to the node ID. The node ID naming rule is n + "level 1 node" + "level 2 node" + "level 3 node" + ..., and so on.
[0107] Figure 7 shows a first mind map corresponding to the example of the structural information of the first mind map described above, Figure 7 In the corresponding node, only the node title is shown; optionally, document information such as the node keyword of the node can also be shown in the node.
[0108] In some embodiments, each child node at the lowest level in the first mind map is used to display document information of an existing document, and each parent node is used to display comprehensive document information of at least one existing document corresponding to all its child nodes. Figure 7 For example, node n-2-1 corresponds to a note related to the analysis of the RAG model architecture, and node n-2-2 corresponds to a note related to the practical application of RAG technology. The parent node of these two nodes, node n-2, is used to display the combined document information of the two nodes. In other words, the note corresponding to node n-2 is a summary note determined based on the comprehensive content of the documents of the two nodes. Optionally, the note content of node n-2 is generated based on the document content of the two nodes using a generative language model.
[0109] The document information includes at least one of a document title, a document summary, and document keywords.
[0110] The document title can be a user-defined title when adding a document, or a title automatically generated by an application based on the document content. Optionally, the document title and the node title can be the same or different.
[0111] The document summary can be a summary customized by the user when adding a document; or, the document summary can be all or part of the document content, such as the first 20 words of the document; or, the document summary can be a summary of the document content automatically generated by the application based on the document content.
[0112] The document keyword can be a keyword that the user customizes when adding a document; or, the document keyword can be a keyword that the application automatically generates based on the document content. Optionally, the document keyword and the node keyword can be the same or different.
[0113] In some embodiments, a third generative language model is used to generate document information for an existing document based on its content. Specifically, the document content of the existing document is input into the third generative language model to generate at least one of a document title, a document summary, and document keywords for the existing document, which are then displayed in a node of the first mind map. The third generative language model is used to generate the document information.
[0114] It should be noted that this application does not limit the content of the document information of the existing documents displayed in the nodes in the first mind map, the display method of the document information of the existing documents in the nodes, and the scope of the document information displayed in each node.
[0115] In some embodiments, when a user opens the note-taking application for the first time each day, an optimization judgment is made on the first mind map; that is, a rationality check of the mind map results is run to determine whether there are nodes that can be optimized. If there are nodes that can be optimized, an optimization suggestion is sent to the user. For example, if there are nodes that can be merged in the first mind map, an optimization suggestion is sent to the user. Optionally, when a user opens the note-taking application for the first time each day, the server runs a rationality check of the mind map results. If an optimizeable node is detected, an optimization suggestion is sent to the client. After the client receives the optimization suggestion, a pop-up window is displayed on the terminal.
[0116] like Figure 8As shown, when the user opens the note application for the first time every day, the user interface displays the first mind map 41; when the first mind map 41 meets the optimization conditions, a pop-up window 42 is displayed, which is used to inform the user of the optimizable items in the current mind map. Optionally, the optimizable items in the first mind map 41 are highlighted or bolded; the user can choose to reject the control 43 (the control with "Ignore" written in the figure) or accept the control 44 (the control with "Aggregate" written in the figure). After the user clicks the accept control 44, the first mind map 41 is optimized to obtain the second mind map 45. Figure 8 As shown, the document contents corresponding to nodes 1.1, 1.2 and 1.3 in the first mind map 41 belong to the same type of content and can be merged and clustered; in the optimized second mind map 45, a new parent node (i.e., node 1.1 in the second mind map) is generated for the three nodes, and the three nodes become nodes 1.1.1, 1.1.2 and 1.1.3 in the second mind map respectively.
[0117] In some embodiments, a second generative language model is used to determine the optimizable nodes in the first mind map. That is, based on the structural information of the first mind map, the node labels of the optimizable nodes in the first mind map are determined based on the second generative language model; wherein the structural information of the first mind map includes the node labels and node content of each node, and the node content includes at least one of the node title, node keyword, and node sub-layer name. wherein the second generative language model is a generative language model for optimizing the mind map; the second generative language model and the first generative language model can be different models, or they can also be the same model; the second generative language model and the third generative language model can be different models, or they can also be the same model.
[0118] In some embodiments, the structural information of the first mind map and the second prompt are input into a second generative language model to determine the node labels of the optimizable nodes in the first mind map. The second prompt is used to instruct the generative language model to determine the optimizable nodes in the mind map. After determining the node labels of the optimizable nodes in the first mind map, the optimizable nodes are merged and displayed, and a parent node of the optimizable nodes is generated.
[0119] An example of the second prompt sentence input into the second generative language model is as follows:
[0120] “##Mind Map Optimization##
[0121] Please analyze the logical grouping dimension of the content theme to see if there are any nodes in the current mind map structure that can be optimized. If so, please list these nodes in order. If not, set "Node Value (idArray)" to ["null"].
[0122] ##Notice##
[0123] Please first output the reason for the analysis process (reason). The reason must be at least 20 characters but no more than 60 characters. The output results must strictly follow the following JSON format:
[0124] ##Output Format##
[0125] {
[0126] "reason": "analysis process",
[0127] "idArray":["ID value of node 1","ID value of node 2",...]
[0128] }”
[0129] Figure 9 The process of optimizing the first mind map into the second mind map is shown. As shown in the figure, the second prompt statement and the structural information of the first mind map are input into the second generative language model. The second generative language model determines the node labels of the nodes that can be optimized in the first mind map by comparing whether the node content of each node belongs to the same type of content and returns the result. For example, if the node labels of the nodes that can be optimized determined by the second generative language model are N-3, N-4, and N-5, then the three nodes are clustered and merged, and a new parent node (node N-3') is generated for the three merged nodes. The three nodes become node N-3'-1, node N-3'-2, and node N-3'-3 in the optimized mind map respectively.
[0130] It should be noted that the step of optimizing the first mind map into the second mind map can be performed at different times depending on the actual situation. In addition to the above-mentioned execution when the user first opens the note application every day, it can also be executed at a fixed time every day, or every time the user opens the note application, or every time the user adds a new document, etc. This application is not limited to this.
[0131] Step 342: In response to the creation operation of the target document in the note application, obtain the document content of the created target document;
[0132] The target document is a new document added to the note-taking application.
[0133] For example, a user creates a target document in a note application, such as Figure 2 As shown, the user creates and edits a note in the note editing area 11 and publishes the note after completing the editing; the note application obtains the document content of the created note.
[0134] Step 344: In response to the import operation of the target document in the note application, obtain the document content of the imported target document;
[0135] Exemplarily, a user imports a target document into a note-taking application. For example, the user receives a document from a social client and imports it into the note-taking application; another example is the user downloads a document from a search engine and sends it to the note-taking application, etc. This application does not limit the method for importing the target document from an external application into the note-taking application. The note-taking application obtains the document content of the imported note.
[0136] In some embodiments, a user subscribes to a folder path; if a new document is added to the folder path, the new document is automatically imported into the note application. For example, if a user subscribes to a folder named "Work," if a new document is added to the "Work" folder, the new document is automatically imported into the note application; if an updated document is added to the "Work" folder, the updated document is automatically imported into the note application and overwrites the original version of the updated document.
[0137] Step 346: In response to the target document collection operation in the note application, obtain document content of the collected target document;
[0138] Exemplarily, the user collects the target document on a third-party platform and adds the target document to the note application.
[0139] For example Figure 10 As shown, a web page 60 is displayed on the user interface. A target document 61 is displayed on the web page 60. The user clicks on a web page plug-in 62 to collect the target document 61. After collecting the target document 61 to the note application, a prompt pop-up window 63 is displayed on the web page 60, which is used to prompt the user that the target document has been collected.
[0140] For example Figure 11 As shown, the target document 64 opened in the social client is displayed on the user interface of the mobile phone. The user collects the target document into the note application through the interface 65 of the note application set in the social client.
[0141] It should be noted that only one of the above steps 342, 344 and 346 needs to be executed, and the execution is performed according to the method of adding the target document in the note application.
[0142] Step 350: Input the document content of the target document, the structural information of the first mind map, and the first prompt sentence into the first generative language model to determine the node position of the target node in the first mind map;
[0143] After obtaining the document content of the target document, the node position of the corresponding target node displaying the target document in the first mind map is determined based on the document content of the target document; that is, the target document is classified according to its document content.
[0144] The target node is used to display the document information of the newly added target document. The target node can be an existing node in the first mind map, or a newly added node in the first mind map. The node position of the target node in the first mind map is determined by the note-taking application based on the document content analysis of the target document.
[0145] Exemplarily, a generative language model is used to determine the node position of the target node in the first mind map. Specifically, the document content of the target document, the structural information of the first mind map, and the first prompt statement are input into the first generative language model to determine the node position of the target node in the first mind map. The structural information of the first mind map includes the node number and node content of each node, and the node content includes at least one of a node title, a node keyword, and a node sublayer name.
[0146] The first generative language model is a generative language model used to determine the node position of the target node in the first mind map based on the document content of the target document. The first generative language model can be a public, cloud-based generative language model, such as a GPT model or an LLaMA model; or the first generative language model can be a privately deployed generative language model.
[0147] The first prompt statement is used to instruct the generative language model to determine the node position of the target node in the first mind map. An example of the first prompt statement is as follows:
[0148] "##position##
[0149] Read and understand the new input content in detail, and compare the new content with the titles and keywords of all nodes in turn. Analyze their similarities in content continuity, contextual coherence, and thematic style. Determine the relationship between the new content and existing nodes, and make the following judgments based on the structure of the current mind map:
[0150] Case 1: If the new content is a continuation of a node's content, is contextually coherent, and has exactly the same theme and style, it can be merged with the node's content. Please output the node's "nodeid" value (i.e., node ID).
[0151] Case 2: If the new content has exactly the same theme and style as a node's content, and has certain similarities in content, and is a specific expansion of the node's content, then a "nodeid" value for a child node is generated and output.
[0152] Case 3: If the new content is similar in theme and style to a node's content, but the content is not directly related, generate a "nodeid" value of a peer node and output it.
[0153] Case 4: If none of the above apply, find a suitable location for the new content in the current mind map, and generate and output a new node's "nodeid" value based on the naming rules of all node "nodeid" values.
[0154] ##Notice##
[0155] Please first output the reason for the analysis process (reason), then output the identifier (id) of the final result based on the analysis content. The output "id" value must conform to the naming rules of all "nodeid" values in the structure information. The reason must be at least 40 characters and no more than 80 characters. The output results must strictly follow the following JSON format.
[0156] ##Output Format##
[0157] {
[0158] "reason":"Analysis process",
[0159] "id":"N-node number-child node number-..."
[0160] }”
[0161] Step 360: Add document information of the target document to the target node in the first mind map.
[0162] Exemplarily, after determining the node position of the target node, adding and displaying document information of the target document at the target node in the first mind map includes at least one of the following situations:
[0163] When the document content of the target document is a continuation of the node content of the first node in the first mind map, the document information of the target document is added to the first node; that is, corresponding to the above situation 1, the document information of the target document is added to the existing first node.
[0164] When the document content of the target document is an expansion of the node content of the second node in the first mind map, the document information of the target document is displayed at a newly added child node of the second node; that is, corresponding to the above-mentioned situation 2, a child node of the existing second node is generated, and the document information of the target document is added to the child node.
[0165] When the document content of the target document has a similar theme to the node content of the third node in the first mind map but has different content, the document information of the target document is displayed at a newly added sibling node of the third node; that is, corresponding to the above-mentioned situation 3, a sibling node of the existing third node is generated, and the document information of the target document is added to be displayed at the sibling node.
[0166] When the document content of the target document is not related to the node content of any node in the first mind map, the document information of the target document is displayed at the newly added node; that is, corresponding to the above situation 4, a new node is added and the document information of the target document is displayed at the node, for example, a new node is added starting from the root node and the document information of the target document is displayed at the node.
[0167] The node content includes at least one of a node title, a node keyword, and a node sub-layer name.
[0168] In some embodiments, based on the document content of the target document, the document information of the target document is generated based on a third generative language model; wherein the document information of the target document includes at least one of a document title, a document summary, and document keywords. That is, the document information of the target document displayed in the first mind map can be customized by the user; or, the document information of the target document displayed in the first mind map can also be generated by the third generative language model based on the document content of the target document; or, only the document information not defined by the user is generated using the third generative language model. The third generative language model can be the same model as the first generative language model, or a different model from the first generative language model. This application is not limited to this.
[0169] Figure 12 FIG shows a process of determining a target node based on the document content of a target document. Figure 12 As shown, the target document's content, the structure of the first mind map, and the first prompt are input into the first generative language model. The first generative language model compares the node information of all nodes in the target document and the first mind map and returns a result, namely, the node number of the target node. For example, if the result returned by the first generative language model is node N-5, as shown in the figure, node 31 is added to the first mind map 30 to display the document information of the target document.
[0170] In some embodiments, after adding document information of the target document to the target node in the first mind map, the node information of the target document is obtained based on the generative language model, and the node information is added to the existing structural information of the first mind map to update the structural information of the first mind map for the next classification comparison.
[0171] In summary, the method provided in the embodiment of the present application displays a first mind map including document information of existing documents. After a new document is added, the document information of the newly added document is added to the corresponding node of the first mind map. This allows the newly added document to be quickly classified into the appropriate location. The user can edit the document quickly and stress-free without having to consider where the document should be stored, thereby improving the efficiency of document use in the application. This method can improve document management efficiency, especially when there are many documents, the user cannot remember all the document contents, and it is difficult to quickly classify the documents.
[0172] In addition, newly added documents can be automatically classified and archived, eliminating the need for users to perform multiple interactive operations between levels to classify documents, thereby improving the efficiency of human-computer interaction.
[0173] In addition, using mind mapping to organize documents makes the documents more structured and systematic, the document content is clearer and easier to remember, and users do not need to search back and forth for files in multi-level list folders, further improving document management and usage efficiency.
[0174] In addition, the method provided in the embodiment of the present application optimizes the mind map in the note-taking application software based on the generative language model, clusters and merges the mergeable nodes, so that the mind map can remain concise and clear, further improving the efficiency of conveying document information; users do not need to manually merge or move documents, but can do it automatically, reducing the number of user interactions and further improving the efficiency of human-computer interaction.
[0175] In addition, the method provided in the embodiment of the present application generates structural information of a mind map based on a generative language model, which can quickly obtain key information of a large number of documents and perform matching or analysis based on this, thereby improving document management efficiency.
[0176] In addition, the method provided in the embodiment of the present application generates document information based on a generative language model. The user does not need to input the title or keywords of the document, but can directly generate it based on the document content, further improving the user's efficiency when using the note-taking application.
[0177] Figure 13The following is a block diagram showing a device for displaying document information according to an exemplary embodiment of the present application. The device includes:
[0178] A display module 420 is configured to display a user interface of the application, wherein the user interface includes a first mind map, and nodes in the first mind map are configured to display document information of at least one existing document;
[0179] An acquisition module 440 is configured to acquire the document content of the target document in response to an add operation on the target document in the application program;
[0180] The display module 420 is further configured to add and display document information of the target document at the target node in the first mind map. The node position of the target node in the first mind map is determined by the application based on document content analysis of the target document.
[0181] In one possible embodiment, the display module 420 is used to add and display the document information of the target document at the first node when the document content of the target document is a continuation of the node content of the first node in the first mind map; or, when the document content of the target document is an expansion of the node content of the second node in the first mind map, display the document information of the target document at a newly added child node of the second node; or, when the document content of the target document has a similar theme but different content to the node content of the third node in the first mind map, display the document information of the target document at a newly added node at the same level of the third node; or, when the document content of the target document is not associated with the node content of any node in the first mind map, display the document information of the target document at a newly added node; wherein the node content includes at least one of a node title, a node keyword, and a node sublayer name.
[0182] In one possible implementation, the display module 420 is used to determine the node position of the target node in the first mind map based on the document content of the target document and the structural information of the first mind map, based on the first generative language model; wherein the structural information of the first mind map includes the node label and node content of each node, and the node content includes at least one of a node title, a node keyword, and a node sublayer name.
[0183] In a possible implementation, the display module 420 is used to input the document content of the target document, the structural information of the first mind map, and the first prompt statement into the first generative language model to determine the node position of the target node in the first mind map, and the first prompt statement is used to instruct the generative language model to determine the node position of the target node in the first mind map.
[0184] In a possible implementation, the display module 420 is configured to display an optimized second mind map on the user interface of the application when the first mind map meets the optimization conditions, where the second mind map is a mind map optimized based on a second generative language model.
[0185] In a possible implementation, the display module 420 is configured to, when there are at least two nodes in the first mind map whose node contents belong to the same type, merge and display the at least two nodes in the second mind map.
[0186] In one possible implementation, the display module 420 is used to determine the node labels of the optimizable nodes in the first mind map based on the structural information of the first mind map and the second generative language model; wherein the structural information of the first mind map includes the node labels and node contents of each node, and the node content includes at least one of the node title, node keyword, and node sublayer name.
[0187] In a possible implementation, the display module 420 is used to input the structural information of the first mind map and the second prompt statement into the second generative language model to determine the node labels of the optimizable nodes in the first mind map, and the second prompt statement is used to instruct the generative language model to determine the optimizable nodes in the mind map.
[0188] In one possible embodiment, the acquisition module 440 is used for at least one of the following: in response to a creation operation on the target document in the application, acquiring the document content of the created target document; in response to an import operation on the target document in the application, acquiring the document content of the imported target document; in response to a collection operation on the target document in the application, acquiring the document content of the collected target document.
[0189] In a possible implementation, the display module 420 is configured to generate document information of the target document based on the document content of the target document and based on a third generative language model; wherein the document information of the target document includes at least one of a document title, a document summary, and document keywords.
[0190] Figure 14 13 is a schematic diagram illustrating the structure of a computer device according to an exemplary embodiment. The computer device may include a terminal or a server. The computer device 1300 includes a central processing unit (CPU) 1301, a system memory 1304 including a random access memory (RAM) 1302 and a read-only memory (ROM) 1303, and a system bus 1305 connecting the system memory 1304 and the central processing unit 1301. In some embodiments, the computer device 1300 may also include a basic input / output system (I / O system) 1306 for facilitating information transmission between various components within the computer device, and a mass storage device 1307 for storing an operating system 1313, application programs 1314, and other program modules 1315.
[0191] The basic input / output system 1306 includes a display 1308 for displaying information and an input device 1309, such as a mouse or keyboard, for user input. Both the display 1308 and the input device 1309 are connected to the central processing unit 1301 via an input / output controller 1310 connected to the system bus 1305. The basic input / output system 1306 may also include an input / output controller 1310 for receiving and processing input from a variety of other devices, such as a keyboard, mouse, or electronic stylus. Similarly, the input / output controller 1310 also provides output to a display screen, printer, or other types of output devices.
[0192] The mass storage device 1307 is connected to the central processing unit 1301 via a mass storage controller (not shown) connected to the system bus 1305. The mass storage device 1307 and its associated computer-readable medium provide non-volatile storage for the computer device 1300. In other words, the mass storage device 1307 may include computer-readable media (not shown) such as a hard disk or a CD-ROM drive.
[0193] Without loss of generality, the computer device readable medium may include computer device storage media and communication media. Computer device storage media include volatile and non-volatile, removable and non-removable media implemented by any method or technology for storing information such as computer device readable instructions, data structures, program modules or other data. Computer device storage media include RAM, ROM, Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), CD-ROM, Digital Video Disc (DVD) or other optical storage, tape cassettes, magnetic tapes, disk storage or other magnetic storage devices. Of course, those skilled in the art will appreciate that the computer device storage media is not limited to the above-mentioned ones. The above-mentioned system memory 1304 and mass storage device 1307 can be collectively referred to as memory.
[0194] According to various embodiments of the present disclosure, the computer device 1300 may also be connected to a remote computer device on a network such as the Internet for operation. That is, the computer device 1300 may be connected to the network 1311 via the network interface unit 1312 connected to the system bus 1305, or the network interface unit 1312 may be used to connect to other types of networks or remote computer device systems (not shown).
[0195] The memory further includes one or more programs, which are stored in the memory. The central processing unit 1301 implements all or part of the steps of the above-mentioned document information display method by executing the one or more programs.
[0196] The present application provides a computer-readable storage medium, wherein the storage medium stores at least one program, and the at least one program is loaded by the processor and executes the document information display method provided by each of the above method embodiments.
[0197] The present application also provides a computer program product, in which at least one program is stored. The processor loads the at least one program and executes the document information display method provided by each of the above method embodiments.
[0198] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.
[0199] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.
[0200] The above description is merely an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.
Claims
1. A method for displaying document information, characterized in that: The method comprises: Displaying a user interface of the application, wherein the user interface includes a first mind map, and nodes in the first mind map are used to display document information of at least one existing document; In response to an add operation on a target document in the application, obtaining document content of the target document; Document information of the target document is added and displayed on a target node in the first mind map. The node position of the target node in the first mind map is determined by the application based on document content analysis of the target document.
2. The method according to claim 1, characterized in that The adding and displaying document information of the target document on the target node in the first mind map includes: In a case where the document content of the target document is a continuation of the node content of the first node in the first mind map, additionally displaying the document information of the target document at the first node; or In a case where the document content of the target document is the expansion of the node content of the second node in the first mind map, displaying the document information of the target document at a newly added child node of the second node; or If the document content of the target document and the node content of the third node in the first mind map have similar themes but different contents, display the document information of the target document at a newly added node at the same level as the third node; or In a case where the document content of the target document is not associated with the node content of any node in the first mind map, displaying the document information of the target document at a newly added node; The node content includes at least one of a node title, a node keyword, and a node sub-layer name.
3. The method according to claim 2, characterized in that The method further comprises: Based on the document content of the target document and the structural information of the first mind map, determining the node position of the target node in the first mind map based on a first generative language model; The structural information of the first mind map includes node numbers and node contents of each node, and the node contents include at least one of node titles, node keywords, and node sub-layer names.
4. The method according to claim 1, wherein The method further comprises: When the first mind map meets the optimization condition, an optimized second mind map is displayed on the user interface of the application, where the second mind map is a mind map optimized based on a second generative language model.
5. The method according to claim 4, characterized in that The step of displaying an optimized second mind map when the first mind map meets an optimization condition includes: When there are at least two nodes in the first mind map whose node contents belong to the same type, the at least two nodes are merged and displayed in the second mind map.
6. The method according to claim 5, characterized in that The method further comprises: Based on the structural information of the first mind map, determining the node labels of the optimizable nodes in the first mind map based on a second generative language model; The structural information of the first mind map includes node numbers and node contents of each node, and the node contents include at least one of node titles, node keywords, and node sub-layer names.
7. A document information display device, characterized in that: The device comprises: a display module, configured to display a user interface of the application, wherein the user interface includes a first mind map, and nodes in the first mind map are configured to display document information of at least one existing document; an acquisition module, configured to acquire the document content of the target document in response to an add operation on the target document in the application; The display module is used to add and display document information of the target document on the target node in the first mind map, and the node position of the target node in the first mind map is determined by the application based on the document content analysis of the target document.
8. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory stores at least one program, and the at least one program is loaded and executed by the processor to implement the document information display method according to any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that The readable storage medium stores at least one program, and the at least one program is loaded and executed by a processor to implement the document information display method according to any one of claims 1 to 6.
10. A computer program product, characterized in that The computer program product stores at least one program segment, and the at least one program segment is loaded and executed by a processor to implement the document information display method according to any one of claims 1 to 6.