Mind mapping generation method and device based on large model and electronic equipment

Abstract

The invention provides a mind map generation method and device based on a large model and electronic equipment, and relates to the field of artificial intelligence, in particular to the technical field of large models and natural language processing. Determining N types of reference elements; according to the N types of reference elements, generation data of the mind map is determined, the generation data comprises nodes and node information of the nodes, and the node information carries node filling data obtained by the large model based on the reference elements; generating data of the mind map are transmitted to a front end, a target mind map is generated at the front end, at least one piece of node information of the target mind map is filled with chart information and supports any type and number of reference elements, the mind map is generated, and at least one piece of node information of the target mind map is filled with chart information and supports any type and number of reference elements. The visualization degree of the target mind map is improved, the generated target mind map better meets user requirements, and the user experience is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence, specifically to the field of large models and natural language processing technology, and in particular to a method, device and electronic device for generating a mind map based on a large model. Background Art

[0002] In related technologies, when generating a mind map based on files, the type and quantity of the files are often limited, and the content of the nodes in the output mind map is mostly a general sentence, resulting in the generated mind map being unable to meet user needs. Summary of the Invention

[0003] The present disclosure provides a method for generating a mind map based on a large model, an apparatus for generating a mind map based on a large model, an electronic device, a storage medium, and a computer program product.

[0004] According to a first aspect of the present disclosure, a method for generating a mind map based on a large model is provided, comprising: determining N types of reference elements; determining generation data of the mind map based on the N types of reference elements, wherein the generation data comprises nodes and node information of the nodes, and the node information carries fill data of the nodes obtained by the large model based on the reference elements; transmitting the generation data of the mind map to a front end to generate a target mind map at the front end, wherein at least one of the node information of the target mind map is filled with the diagram information.

[0005] According to a second aspect of the present disclosure, a method for generating a mind map based on a large model is provided, comprising: determining N types of reference elements; sending the N types of reference elements to a backend, wherein the reference elements are used to generate generation data of a mind map, and the generation data comprises nodes and node information of the nodes, wherein the node information carries fill data of the nodes obtained by the large model based on the reference elements; receiving the generation data of the mind map transmitted by the backend; generating a target mind map according to the generation data of the mind map, wherein at least one of the node information of the target mind map is filled with the diagram information.

[0006] According to a third aspect of the present disclosure, a device for generating a mind map based on a large model is provided, comprising: a first determination module for determining N types of reference elements; a second determination module for determining generation data of a mind map based on the N types of reference elements, wherein the generation data comprises nodes and node information of the nodes, and the node information carries fill data of the nodes obtained by the large model based on the reference elements; a transmission module for transmitting the generation data of the mind map to a front end to generate a target mind map at the front end, wherein at least one of the node information of the target mind map is filled with the chart information.

[0007] According to a fourth aspect of the present disclosure, a device for generating a mind map based on a large model is provided, comprising: a determination module for determining N types of reference elements; a sending module for sending the N types of reference elements to a back end, wherein the reference elements are used to generate generation data of a mind map, and the generation data comprises nodes and node information of the nodes, wherein the node information carries fill data of the nodes obtained by the large model based on the reference elements; a receiving module for receiving the generation data of the mind map transmitted by the back end; a generation module for generating a target mind map according to the generation data of the mind map, wherein at least one of the node information of the target mind map is filled with the diagram information.

[0008] According to a fifth aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable 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 method for generating a mind map based on a large model described in the first aspect of the present disclosure or the method for generating a mind map based on a large model described in the second aspect.

[0009] According to a sixth 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 the computer to execute the method for generating a mind map based on a large model described in the first aspect of the present disclosure or the method for generating a mind map based on a large model described in the second aspect.

[0010] According to a seventh aspect of the present disclosure, a computer program product is provided, comprising a computer program, characterized in that when the computer program is executed by a processor, it implements the method for generating a mind map based on a large model according to the first aspect of the present disclosure or the method for generating a mind map based on a large model according to the second aspect.

[0011] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0013] Figure 1 Schematic diagram of a flow chart of a method for generating a mind map based on a large model according to an embodiment of the present disclosure;

[0014] Figure 2 Schematic diagram of a flow chart of a method for generating a mind map based on a large model according to an embodiment of the present disclosure;

[0015] Figure 3 Schematic diagram of a flow chart of a method for generating a mind map based on a large model according to an embodiment of the present disclosure;

[0016] Figure 4 Schematic diagram of a flow chart of a method for generating a mind map based on a large model according to an embodiment of the present disclosure;

[0017] Figure 5 Schematic diagram of a flow chart of a method for generating a mind map based on a large model according to an embodiment of the present disclosure;

[0018] Figure 6 A schematic diagram of a mind map according to an embodiment of the present disclosure;

[0019] Figure 7 Schematic diagram of the structure of a device for generating a mind map based on a large model according to an embodiment of the present disclosure;

[0020] Figure 8 Schematic diagram of the structure of a device for generating a mind map based on a large model according to an embodiment of the present disclosure;

[0021] Figure 9 A schematic block diagram of an electronic device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0022] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize 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.

[0023] The following is a brief description of the technical field involved in the solution of the present disclosure:

[0024] Artificial Intelligence (AI) is the study of how computers can simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). This discipline encompasses both hardware and software technologies. AI hardware technologies generally include computer vision, speech recognition, natural language processing, as well as deep learning / learning, big data processing, and knowledge graphs.

[0025] Natural Language Processing (NLP) is a key research area in the field of artificial intelligence. It integrates knowledge from multiple disciplines, including linguistics, computer science, machine learning, mathematics, and cognitive psychology. It is an interdisciplinary subject that integrates computer science, artificial intelligence, and linguistics. It encompasses two main areas: natural language understanding and natural language generation. Its research covers multiple levels, including characters, words, phrases, sentences, paragraphs, and texts. It serves as a bridge between machine language and human language. It aims to enable machines to understand, interpret, and generate human language, enabling effective communication between humans and machines and enabling computers to perform tasks such as language translation, sentiment analysis, and text summarization.

[0026] A large model refers to a machine learning model with a large parameter scale and complexity. It requires a large amount of computing resources and storage space to train and store, and often requires distributed computing and special hardware acceleration technology. Large models have stronger generalization and expression capabilities.

[0027] A method for generating a mind map based on a large model according to an embodiment of the present disclosure is described below with reference to the accompanying drawings.

[0028] Figure 1 It is a flowchart of a method for generating a mind map based on a large model according to an embodiment of the present disclosure.

[0029] It should be noted that the execution subject of the method for generating a mind map based on a large model in this embodiment is a device for generating a mind map based on a large model, and the device for generating a mind map based on a large model can be a hardware device or software in a hardware device, etc. The hardware device may be a terminal device, a server, etc.

[0030] like Figure 1 As shown, the method for generating a mind map based on a large model proposed in this embodiment includes the following steps:

[0031] S101. Determine N types of reference elements.

[0032] In the embodiment of the present disclosure, N types of reference elements may be determined at the front end and sent to the back end to determine the N types of reference elements.

[0033] It should be noted that the present disclosure does not limit the types of reference elements.

[0034] Optionally, the types of reference elements include but are not limited to txt, word, pdf, ppt, jpg, png, jpeg, webp, heic, tiff, audio, video, web page link, etc.

[0035] Optionally, the request information sent by the front end may be received and parsed to obtain an element identifier (Identity document, ID for short) of the reference element in the request information, and the reference element may be determined based on the element ID.

[0036] It should be noted that when the reference elements are more complex, in order to improve the fluency of the mind map generation process, remote mapping can be performed based on the element ID to determine the reference elements.

[0037] Optionally, after the element ID is obtained, element matching and reading operations are performed in the canvas data based on the element ID to determine N types of reference elements.

[0038] S102. Determine generated data of a mind map based on N types of reference elements, wherein the generated data includes nodes and node information of the nodes, and the node information carries fill data of the nodes obtained by the large model based on the reference elements.

[0039] In the embodiment of the present disclosure, the request information sent by the front end can be received and parsed to obtain the requirement information of the mind map in the request information, and the generation data of the mind map can be determined based on the requirement information of the mind map and N types of reference elements.

[0040] The mind map generation data is used to generate the mind map.

[0041] Optionally, the generation prompt words of the mind map can be determined based on the demand information of the mind map, and cross-modal understanding of N types of reference elements can be performed based on multiple intelligent agents to obtain multimodal structured parsing results. The generation data of the mind map can be determined based on the generation prompt words and the structured parsing results.

[0042] It should be noted that the present disclosure does not limit the type of data filled in the node. Optionally, the type of data filled in includes but is not limited to text and / or translated text in different languages corresponding to the text, charts, web pages, web page links, etc.

[0043] S103: Transmitting the generated data of the mind map to the front end to generate a target mind map at the front end, wherein at least one node information of the target mind map is filled with chart information.

[0044] It should be noted that in related technologies, the nodes of mind maps are often filled with simple text content, but in the target mind map generated on the front end, the nodes can be filled with various types of information such as pictures and tables, which can better meet user needs and be more consistent with the original content of the reference elements, thereby improving the visualization of the target mind map.

[0045] In the embodiment of the present disclosure, after acquiring the generated data of the mind map, the generated data of the mind map is transmitted to the front end. Accordingly, after receiving the generated data of the mind map, the front end can generate a target mind map at the front end.

[0046] According to the method for generating a mind map based on a large model in an embodiment of the present disclosure, by determining N types of reference elements, the generation data of the mind map is determined according to the N types of reference elements, wherein the generation data includes nodes and node information of the nodes, and the node information carries the fill data of the nodes obtained by the large model based on the reference elements, and the generation data of the mind map is transmitted to the front end to generate a target mind map at the front end, wherein at least one node information of the target mind map is filled with chart information. Therefore, the present disclosure supports any type and number of reference elements to generate a mind map, and at least one node information of the target mind map is filled with chart information, thereby improving the visualization of the target mind map, and the generated target mind map is more in line with user needs, thereby improving the user experience.

[0047] Figure 2 It is a flowchart of a method for generating a mind map based on a large model according to an embodiment of the present disclosure.

[0048] like Figure 2 As shown, the method for generating a mind map based on a large model proposed in this embodiment includes the following steps:

[0049] S101 “determining N types of reference elements” in the above embodiment may specifically include S201 and S202 .

[0050] S201: Receive request information sent by the front end, where the request information includes an element identification ID of a reference element.

[0051] In the embodiment of the present disclosure, the request information sent by the front end is received, and the request information is parsed to obtain the element identification ID of the reference element in the request information.

[0052] S202: Determine the reference element according to the element ID.

[0053] In the embodiment of the present disclosure, after the element ID is obtained, remote mapping can be performed in the canvas data based on the element ID to determine N types of reference elements.

[0054] Optionally, a mapping relationship table between canvas data and element IDs may be queried, and based on the element ID, element matching and reading may be performed by querying the mapping relationship table to determine target data in the canvas data, and the target data may be determined as a reference element.

[0055] For example, if the element IDs are "ABC" and "123", based on the element ID "ABC", the elements are matched and read by querying the mapping relationship table, and the target data in the canvas data is determined to be "Picture 2", and "Picture 2" is determined as the reference element. Based on the element ID "123", the elements are matched and read by querying the mapping relationship table, and the target data in the canvas data is determined to be "PPT1", and "PPT1" is determined as the reference element.

[0056] In the above embodiment, S102 “determining the generated data of the mind map according to N types of reference elements” may specifically include S203 and S204.

[0057] S203. Obtain demand information for the mind map.

[0058] In the disclosed embodiment, the request information sent by the front end can be received, and the request information can be parsed to obtain the mind map generation requirements in the request information, and the mind map generation requirements can be semantically understood to obtain the mind map requirement information.

[0059] S204: Determine the generated data of the mind map according to the demand information and N types of reference elements.

[0060] In the disclosed embodiment, the generation prompt words of the mind map can be determined based on the demand information of the mind map, and cross-modal understanding of N types of reference elements can be performed based on multiple intelligent agents to obtain multimodal structured parsing results. The generation data of the mind map can be determined based on the generation prompt words and the structured parsing results.

[0061] Optionally, multiple pre-trained intelligent agents can be obtained, such as text understanding intelligent agents, image understanding intelligent agents, audio and video understanding intelligent agents, etc. According to N types of reference elements, a reference element type routing matrix is constructed. Based on the reference element type routing matrix, corresponding intelligent agents are matched. Based on multiple intelligent agents, cross-modal understanding of N types of reference elements is performed to obtain multimodal structured parsing results.

[0062] It should be noted that by constructing a reference element type routing matrix, intelligent adaptation of various types of intelligent agents can be supported. By performing cross-modal understanding of N types of reference elements, unified semantic understanding of heterogeneous data can be achieved. For example, semantic unit segmentation of text and feature extraction of images, audio and video streams can be performed to obtain multimodal structured parsing results.

[0063] In the embodiment of the present disclosure, the scheduling path of the intelligent agent can be determined based on the generated prompt words and the structured parsing results. According to the scheduling path, the intelligent agent is called, and based on the intelligent agent, multimodal extraction is performed on N types of reference elements to obtain key data related to the node, and the filling data of the node is obtained based on the key data related to the node.

[0064] Optionally, if there are shortcut instructions, the fusion analysis of generated prompt words and shortcut instructions can be achieved based on deep semantic understanding technology, and a domain-specific language (DSL) with weight parameters can be generated through a constraint satisfaction algorithm. A graph neural network is used to build a scheduling task dependency model for the intelligent agent. Based on the DSL with weight parameters, real-time load and the characteristics of the structured parsing results (complexity / priority / resource requirements), the scheduling path of the intelligent agent is determined, and the intelligent agent is called according to the scheduling path. This can support flexible reconstruction of the scheduling path to improve the efficiency of determining the generated data for the mind map.

[0065] Optionally, based on the called intelligent agent, multimodal extraction of N types of reference elements is performed in parallel through parallel processing, and a unified knowledge representation is generated based on the knowledge fusion layer to obtain key data related to the node and provide a standardized data interface for downstream modules.

[0066] Optionally, in the process of multi-cross-modal understanding of N types of reference elements, a streaming processing mechanism can be supported to support real-time incremental updates.

[0067] In the embodiment of the present disclosure, the key data of the node is screened and adaptively laid out according to the content density of the key data to obtain the filling data of the node.

[0068] For example, the information entropy weight of key data can be dynamically calculated through feature space projection technology, and the content density of key data can be determined based on the information entropy weight. According to the content density of key data, the key data of the node can be screened and adaptively laid out to obtain the filling data of the node.

[0069] For example, based on the content density of key data, the size, contrast and other visual presentation strategies of the node's filling data and the interaction mode of the node's filling data can be intelligently adjusted.

[0070] It should be noted that based on the above process, a mind map generation system can be constructed, which includes at least: a cross-modal content understanding module (perception layer), a dynamic task scheduling agent module (decision-making layer), a multimodal execution engine, a mind map generation data module (structured content assembler) and a continuous learning feedback module. Among them, the cross-modal content understanding module is used to perform cross-modal understanding of N types of reference elements based on multiple agents; the dynamic task scheduling agent module is used to determine the scheduling path of the agent; the multimodal execution engine is used to perform multimodal extraction of N types of reference files based on the agent; the mind map generation data module is used to generate mind map generation data to generate a mind map, and the continuous learning feedback module is used to optimize the system.

[0071] Optionally, for the continuous learning feedback module, after the target mind map is generated based on the generated data of the mind map, a triplet feedback learning framework of "original output, user correction, and difference vector" can be obtained. By establishing a high-dimensional feedback feature space, a comparative learning method is used to extract user preference features, and explicit feedback (user behavior data) and implicit feedback (user editing data) are combined. Explicit feedback includes but is not limited to: user downloading, liking, and disliking behavior data of the target mind map; implicit feedback includes but is not limited to: user editing, modification, and other data of the target mind map. An editing trajectory heat map is constructed based on the user editing data, and the user behavior data and the editing trajectory heat map are multimodally fused. The above-mentioned mind map generation system is dynamically updated through an incremental model optimization algorithm to ensure the generation of a more accurate mind map.

[0072] In the above embodiment, S103 “transmitting the generated data of the mind map to the front end” may specifically include S205 and S208.

[0073] S205: Determine the dependency relationship between nodes based on the node information.

[0074] Optionally, the causal relationship and hierarchical relationship between nodes may be determined from the node information, so as to determine the dependency relationship between the nodes based on the causal relationship and hierarchical relationship.

[0075] S206. Based on the dependency relationship, the generated data of the mind map is streamed to the front end.

[0076] Among them, streaming transmission usually adopts the incremental transmission method, that is, only the changed data is transmitted.

[0077] In an embodiment of the present disclosure, the node ID of the node to be updated can be determined based on the dependency relationship, and the first node information sent by the node and the second node information to be sent can be determined based on the node ID. The incremental node information of the node can be determined based on the first node information and the second node information. Based on the incremental node information, a data frame can be generated and sent to the front end.

[0078] Optionally, the node ID of the updated node may be determined, and the node ID of the node to be updated may be determined based on the inter-node dependency and the node ID of the updated node.

[0079] It should be noted that in order to increase the speed of generating the target mind map, after obtaining the first node information and the second node information, the first node information and the second node information can be compared to obtain the difference information between the first node information and the second node information, and the difference information can be used as the incremental node information of the node.

[0080] For example, if the first node information is node 1, node 2, and node 3, and the second node information is node 1, node 2, node 3, and node 4, then the incremental node information of the node is node 4.

[0081] It should be noted that after obtaining the incremental node information, a data frame can be generated based on the incremental node information, that is, the data frame contains the incremental node information, and the data frame is streamed to the front end without the need for full transmission, which improves the transmission rate of the mind map generation data, ensures the mind map generation rate, and thus can reduce the user's waiting time.

[0082] In the embodiment of the present disclosure, after the generated data of the mind map is streamed to the front end, the canvas data fed back after each rendering by the front end is received, and the canvas data is transcoded and stored.

[0083] Among them, the data frame is used to generate the target mind map.

[0084] S207: Generate a target mind map at the front end, wherein at least one node information of the target mind map is filled with chart information.

[0085] According to the method for generating a mind map based on a large model in an embodiment of the present disclosure, a request information sent by a front end is received, the request information includes an element identification ID of a reference element, the reference element is determined according to the element ID, the requirement information of the mind map is obtained, the generation data of the mind map is determined according to the requirement information and N types of reference elements, the dependency relationship between the nodes is determined according to the node information, and the generation data of the mind map is streamed to the front end based on the dependency relationship, and a target mind map is generated at the front end, wherein at least one node information of the target mind map is filled with chart information. Therefore, the present disclosure determines the reference element according to the element ID. When the reference element is relatively large, the space occupied by the data can be reduced through remote mapping. By streaming the generation data of the mind map to the front end, the transmission rate of the generation data of the mind map is improved, the jamming problem in the generation process of the target mind map is solved, the efficiency and fluency of generating the target mind map are improved, and the waiting time of the user is reduced, thereby improving the user experience.

[0086] Figure 3 It is a flowchart of a method for generating a mind map based on a large model according to an embodiment of the present disclosure.

[0087] like Figure 3 As shown, the method for generating a mind map based on a large model proposed in this embodiment includes the following steps:

[0088] S301: Determine N types of reference elements.

[0089] It should be noted that the present disclosure does not limit the specific method for determining N types of reference elements, and they can be selected according to actual circumstances.

[0090] Optionally, canvas data may be determined on the front end, and a physical operation, such as a selection operation or a click operation, may be performed on at least one canvas data, and the canvas data on which the physical operation is performed may be used as a reference element.

[0091] For example, N types of files can be uploaded to the front end, each uploaded file can be used as a canvas data, a selection operation can be performed on at least one canvas data on one or more canvases, and the canvas data on which the selection operation is performed can be used as a reference element.

[0092] For example, a frame selection operation may be performed on document a on canvas 1 , and a frame selection operation may be performed on picture S on canvas 2 , with document a and picture S used together as reference elements.

[0093] It should be noted that the present disclosure does not limit the specific method for obtaining N types of files, and a method can be selected according to actual circumstances.

[0094] Optionally, N types of files may be selected from a plurality of types of file libraries; optionally, N types of files may be generated based on artificial intelligence.

[0095] It should be noted that the N types include but are not limited to txt, word, pdf, ppt, jpg, png, jpeg, webp, heic, tiff, audio, video, web link, etc.

[0096] S302. Send N types of reference elements to the backend. The reference elements are used to generate the generated data of the mind map. The generated data includes nodes and node information of the nodes. The node information carries the filling data of the nodes obtained by the large model based on the reference elements.

[0097] In the embodiment of the present disclosure, after N types of reference elements are acquired, the N types of reference elements are sent to the backend.

[0098] S303: Receive the mind map generation data transmitted by the back end.

[0099] In an embodiment of the present disclosure, when N types of reference elements are sent to the back end, the back end can determine the generation data of the mind map based on the N types of reference elements, and transmit the generation data of the mind map to the front end. Accordingly, the front end receives the generation data of the mind map transmitted by the back end.

[0100] Among them, the mind map generation data is used to generate the mind map.

[0101] S304: Generate a target mind map according to the generated data of the mind map, wherein at least one node information of the target mind map is filled with chart information.

[0102] In the disclosed embodiment, the dependency relationship between nodes can be determined based on the node information, and the nodes in the mind map can be rendered based on the dependency relationship between nodes and the node information of the nodes, until the end node rendering is completed to generate the target mind map.

[0103] In the embodiment of the present disclosure, starting from the root node, the nodes of the mind map can be rendered layer by layer based on the dependency relationship, and for the current node, the filling area of the node can be filled according to the text information of the node.

[0104] For example, for the current node, the filling area of the node may be filled according to the text information of the node, and the chart information may be inserted into the filling area according to the chart information of the node.

[0105] According to the method for generating a mind map based on a large model according to an embodiment of the present disclosure, by determining N types of reference elements, N types of reference elements are sent to the back end, the reference elements are used to generate the generated data of the mind map, the generated data includes nodes and node information of the nodes, wherein the node information carries the fill data of the nodes obtained by the large model based on the reference elements, the generated data of the mind map transmitted by the back end is received, the dependency relationship between the nodes is determined according to the node information, and the nodes in the mind map are rendered according to the dependency relationship and the node information of the nodes, until the end node rendering is completed to generate the target mind map. Therefore, the present disclosure can support any type and number of reference elements to generate a mind map, improve the accuracy and visualization of the target mind map, and generate a target mind map that better meets the needs of users.

[0106] like Figure 4 As shown, the method for generating a mind map based on a large model proposed in this embodiment includes the following steps:

[0107] S401: Determine N types of reference elements.

[0108] S302 “sending N types of reference elements to the backend” in the above embodiment may specifically include S402 and S405 .

[0109] S402: Determine the element ID of the reference element.

[0110] Optionally, an element ID generation strategy can be determined to generate a unique ID for each reference element, bind the reference element to the corresponding unique ID, and construct a mapping relationship table between the reference element and the element ID. Based on the reference element, the element ID of the reference element can be determined by querying the mapping relationship table.

[0111] S403. Obtaining the generation requirements of the mind map.

[0112] Optionally, the mind map generation requirement information input by the user at the front end may be obtained, and semantic understanding of the mind map generation requirement information may be performed to obtain the mind map generation requirement.

[0113] S404: Generate request information according to the element ID and generation requirements.

[0114] Optionally, after the element ID and the generation requirement are obtained, Server Sent Events (SSE) request information may be generated according to the element ID and the generation requirement.

[0115] S405: Send request information to the backend.

[0116] In the embodiment of the present disclosure, after the request information is generated, the request information is sent to the backend, wherein the request information includes the generation requirement of the mind map and the element ID of the reference element.

[0117] In the above embodiment, S303 “receiving the mind map generation data streamed from the back end” may specifically include S406 .

[0118] S406: Receive a data frame determined by the backend based on the generated data, where the data frame includes incremental node information of at least one node to be updated.

[0119] In the above embodiment, S304 “generating a target mind map according to the mind map generation data” may specifically include S407 and S408.

[0120] S407: Determine the incremental filling data of the node according to the incremental node information in the data frame and the historical node information of the node.

[0121] In an embodiment of the present disclosure, the node ID of the node is determined, the first filling data associated with the node ID is determined from the incremental node information, the second filling data associated with the node ID is determined from the historical node information, and the filling incremental data of the node is obtained based on the first filling data and the second filling data.

[0122] It should be noted that after obtaining the first filling data and the second filling data, the first filling data and the second filling data can be compared to obtain the difference data between the first filling data and the second filling data, and the difference data can be used as the filling incremental data of the node.

[0123] For example, if the node ID is node 2, when the first filling data of node 2 is m1, m2, m3 and m4, and the second filling data is m1, m2 and m3, the difference data between the first filling data and the second filling data is m4, then the filling incremental data of node 2 is m4.

[0124] S408. Fill the nodes with the rich text editor according to the incremental filling data of the nodes until the end node rendering is completed to generate the target mind map.

[0125] It should be noted that after obtaining the incremental fill data of the node, the front-end tool can be called to convert the incremental fill data of the node from Markdown format to delta format, wherein the incremental fill data in delta format is data that can be directly rendered.

[0126] It should be noted that during the rendering process, if a full rendering is performed each time, it will often be slow. According to the incremental filling data of the nodes, the nodes are filled in through the rich text editor until the last node rendering is completed to generate the target mind map, which greatly improves the rendering performance.

[0127] In an embodiment of the present disclosure, update flag information of a node is determined, and the node is filled in the current mind map through a rich text editor according to the update flag information and the incremental fill data of the node.

[0128] Optionally, a data frame determined by the backend based on generated data can be received, wherein the data frame includes update flag information to the node, wherein the update flag information can be understood as an update marking protocol, and the update flag information is used to indicate whether the node is a newly created node or an updated node.

[0129] In an embodiment of the present disclosure, based on the update flag information, the node is determined to be a new node, the parent node of the node is determined, and the node is connected to the parent node of the node to expand the node of the current mind map, and the node is filled in based on the incremental fill data through the rich text editor.

[0130] For example, for node 2, if node 2 is a newly created node and the parent node of node 2 is node 1, node 2 is connected to node 1, the node of the current mind map is expanded, and node 2 is filled in based on the incremental filling data through the rich text editor.

[0131] In an embodiment of the present disclosure, a node is determined to be an updated node based on update flag information, the node is located in the current mind map based on the node ID of the node, and the node is filled in through a rich text editor based on incremental fill data.

[0132] For example, for node 2, if node 2 is an update node, node 2 is located in the current mind map, and node 2 is filled in through the rich text editor based on the incremental filling data.

[0133] In an embodiment of the present disclosure, after a target mind map is generated, an editing operation for the target mind map may be received, and the target mind map may be edited according to the editing operation to obtain a final mind map.

[0134] For example, operations such as modifying and replacing the filling data of the nodes in the target mind map can be performed, and operations such as adding and deleting the nodes in the target mind map can be performed to obtain the final mind map.

[0135] In an embodiment of the present disclosure, after generating a target mind map, an original text positioning operation for a node in the target mind map can be received, and based on the original text positioning operation, the original content in the reference element associated with the node can be determined and displayed.

[0136] It should be noted that by determining the original content in the reference element associated with the node, users can accurately track the original content in the reference element and perform backtracking. Subsequently, based on the original content, the target mind map can be verified or supplemented with information to ensure the accuracy and interpretability of the final mind map.

[0137] According to the method for generating a mind map based on a large model in an embodiment of the present disclosure, by determining N types of reference elements, determining the element ID of the reference element, obtaining the generation requirements of the mind map, generating request information according to the element ID and the generation requirements, sending the request information to the back end, receiving a data frame determined by the back end based on the generated data, wherein the data frame includes incremental node information of at least one node to be updated, determining the incremental fill data of the node according to the incremental node information and the historical node information of the node in the data frame, filling the node through a rich text editor according to the incremental fill data of the node until the end node rendering is completed to generate a target mind map, thereby, the present disclosure fills the node through a rich text editor according to the incremental fill data of the node until the end node rendering is completed to generate a target mind map, avoiding the problem of flashing screen or even white screen caused by frequent overwriting and rendering of large reference elements, improving the efficiency and smoothness of generating the target mind map, and when at least one node information of the target mind map is filled with chart information, the visualization of the target mind map is improved, the content in the target mind map is interpretable, can better meet user needs, and improve the user experience.

[0138] The specific process of the method for generating a mind map based on a large model implemented in the present disclosure is explained below.

[0139] For example, if Figure 5As shown, the business front-end (front-end) selects data on the canvas, uses the selected data as a reference element, determines the element ID of the reference element, obtains the generation requirement of the mind map, generates request information based on the element ID and the generation requirement, and sends the request information (SSE request) to the business back-end (back-end). The back-end receives the SSE request sent by the front-end. The request information includes the element identification ID of the reference element, maps according to the element ID, determines the reference element, and notifies the transcoding back-end to obtain canvas data, document data, etc., that is, obtains the reference element, processes the reference element through the transcoding back-end, and the transcoding back-end returns the transcoded data to the business back-end. The back-end sends a request to the strategy to generate a mind map. Based on the request, the mind map is generated. The strategy obtains the transcoded data from the transcoding back-end, and the strategy sends a message RR to the transcoding back-end indicating that the transcoded data has been received. return", the strategy determines the generated data of the mind map based on the demand information and the transcoded data. The generated data includes nodes, node information of nodes and dependencies between nodes. The node information carries the fill data of the node. The fill data includes at least one of the text information and chart information of the node. The strategy streams the generated data of the mind map to the business backend. The transcoding backend determines the incremental node information of the node based on the generated data of the mind map, and transcodes it into incremental frames according to the incremental node information (markdown). The transcoding backend returns the incremental frame (data frame) to the business backend, and the backend streams the generated data of the mind map to the front end, that is, sends the data frame to the business front end. The mind map editor is based on The incremental node information in the data frame and the historical node information of the node are used to determine the incremental filling data (filling data) of the node. According to the incremental filling data of the node, the node is filled with filling data through the rich text editor until the end node rendering is completed to generate the target mind map. After the data is changed, the mind map editor sends a canvas data change notification to the business front end. After the business front end receives the canvas data change notification, it submits the new canvas data to the business back end for saving. After the business back end saves the canvas data, it returns the saved result to the business front end and submits the new canvas data to the transcoding back end for transcoding the canvas data. After the transcoding back end transcoding process, it returns the transcoding result to the business back end.

[0140] The specific process of the method for generating a mind map based on a large model implemented in the present disclosure is explained below.

[0141] For example, the user can select one or more files on the canvas. The file types include but are not limited to txt, word, pdf, ppt, jpg, png, jpeg, webp, heic, tiff, audio, video, web link, etc., use the selected files as reference elements, and click "Generate Mind Map" in the bottom dialog box. The pre-set text "Generate a mind map based on the selected content, the theme is [theme content]" appears in the input box. You can enter the mind map generation requirement information in the input box and send it, for example: the theme content of the mind map. If the user does not enter the theme content and sends it directly, the mind map can be generated for the reference elements in a summary manner by default. Starting from the root node on the front-end canvas, the nodes of the mind map are rendered layer by layer based on the dependency relationship. The end node can display the extracted pictures, charts, etc. flatly until the entire visual mind map is output. After outputting the mind map, you can click the toolbar for further operations to edit the mind map or locate the original text of the nodes in the mind map.

[0142] For example, for a mind map with the theme of "2024 Micro-Short Drama Industry Research Report Analysis", Figure 6 As shown, the root node is "2024 Micro-short Drama Industry Research Report Analysis", sub-node 1 "Micro-short Drama Concept Definition", sub-node 2 "Industry Overview", sub-node 3 "Content and Competition Analysis", sub-node 4 "Industry Chain and Business Model", with sub-node 2 "Industry Overview" as the parent node, including sub-node 5 "Development Environment Analysis", sub-node 6 "Classification and Characteristics" and sub-node 7 "Market Size and Growth", with sub-node 4 "Industry Chain and Business Model" as the parent node, including sub-node 8 Industry Chain, and with sub-node 5 "Development Environment Analysis" as the parent node, including sub-node 9 "Market Environment" and sub-node 10 "Market Driver", and with sub-node 8 Industry Chain as the parent node, including sub-node 10 "Market Size" and sub-node 11 "Growth Momentum".

[0143] In summary, according to the method for generating a mind map based on a large model in the embodiment of the present disclosure, it is possible to refer to any number of multi-type reference elements to generate a mind map or a document summary mind map around the topic required by the user, and it is possible to automatically extract charts to assist users in understanding and memorizing. The mind map has a high degree of visualization, and the cost of understanding complex knowledge is greatly reduced. The generated mind map does not repeat or miss key knowledge points in the reference elements. After the node has a built-in editor that supports rich text, all key information in the reference elements can be quickly and accurately obtained through the generated mind map. By clicking the reference mark in the node, the corresponding position of the reference element can be quickly jumped to, helping users to accurately track the data source of the generated mind map, which is beneficial for users to further verify or supplement information. The generated mind map has many advantages such as clear structure, detailed content, and strong interpretability.

[0144] In the technical solutions disclosed herein, the acquisition, storage, and application of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0145] Corresponding to the methods for generating mind maps based on large models provided in the above-mentioned embodiments, an embodiment of the present disclosure also provides a device for generating mind maps based on large models. Since the device for generating mind maps based on large models provided in the embodiment of the present disclosure corresponds to the methods for generating mind maps based on large models provided in the above-mentioned embodiments, the implementation of the method for generating mind maps based on large models is also applicable to the device for generating mind maps based on large models provided in this embodiment, and will not be described in detail in this embodiment.

[0146] Figure 7 It is a structural diagram of a device for generating a mind map based on a large model according to an embodiment of the present disclosure.

[0147] like Figure 7 As shown, the device 700 for generating a mind map based on a large model includes: a first determination module 710 , a second determination module 720 and a transmission module 730 .

[0148] A first determining module 710 is configured to determine N types of reference elements;

[0149] A second determining module 720 is configured to determine generated data of a mind map based on the N types of reference elements, wherein the generated data includes nodes and node information of the nodes, and the node information carries populated data of the nodes obtained by the macro model based on the reference elements;

[0150] The transmission module 730 is configured to transmit the generated data of the mind map to the front end, so as to generate a target mind map at the front end, wherein at least one node information of the target mind map is filled with the diagram information.

[0151] The second determining module 720 is further configured to: obtain the requirement information of the mind map; and determine the nodes, the node information of the nodes, and the dependency relationships between the nodes based on the requirement information and the N types of reference elements.

[0152] Among them, the second determination module 720 is also used to: determine the generation prompt words of the mind map based on the demand information of the mind map; based on multiple intelligent agents, perform cross-modal understanding of the N types of reference elements to obtain multimodal structured parsing results; and determine the generation data of the mind map based on the generation prompt words and the structured parsing results.

[0153] Among them, the second determination module 720 is also used to: determine the scheduling path of the intelligent agent based on the generated prompt words and the structured parsing results; schedule the intelligent agent according to the scheduling path, and based on the intelligent agent, perform multimodal extraction on the N types of reference elements to obtain key data related to the node; and obtain the fill data of the node based on the key data related to the node.

[0154] The second determining module 720 is further configured to filter and adaptively layout the key data of the node according to the content density of the key data to obtain filling data of the node.

[0155] The first determining module 710 is further configured to: receive request information sent by the front end, the request information including the element identification ID of the reference element; and determine the reference element according to the element ID.

[0156] The transmission module 740 is further configured to: determine the dependency relationship between nodes according to the node information; and stream the generated data of the mind map to the front end based on the dependency relationship.

[0157] Among them, the transmission module 740 is also used to: determine the node ID of the node to be updated based on the dependency relationship; determine the first node information sent by the node and the second node information to be sent according to the node ID; determine the incremental node information of the node according to the first node information and the second node information; generate the data frame according to the incremental node information, and send the data frame to the front end.

[0158] The device 700 is further configured to receive canvas data fed back by the front end after each rendering, and transcode and store the canvas data.

[0159] According to the embodiment of the present disclosure, a device for generating a mind map based on a large model is disclosed. By determining N types of reference elements, the generation data of the mind map is determined according to the N types of reference elements, wherein the generation data includes nodes and node information of the nodes, and the node information carries the fill data of the nodes obtained by the large model based on the reference elements. The generation data of the mind map is transmitted to the front end to generate a target mind map at the front end, wherein at least one of the node information of the target mind map is filled with chart information. Therefore, the present disclosure supports any type and number of reference elements to generate a mind map, and at least one node information of the target mind map is filled with chart information, thereby improving the visualization of the target mind map, and the generated target mind map is more in line with user needs, thereby improving the user experience.

[0160] Corresponding to the methods for generating mind maps based on large models provided in the above-mentioned embodiments, an embodiment of the present disclosure also provides a device for generating mind maps based on large models. Since the device for generating mind maps based on large models provided in the embodiment of the present disclosure corresponds to the methods for generating mind maps based on large models provided in the above-mentioned embodiments, the implementation of the method for generating mind maps based on large models is also applicable to the device for generating mind maps based on large models provided in this embodiment, and will not be described in detail in this embodiment.

[0161] Figure 8 It is a structural diagram of a device for generating a mind map based on a large model according to an embodiment of the present disclosure.

[0162] like Figure 8 As shown, the device 800 for generating a mind map based on a large model includes: a determination module 810, a sending module 820, a receiving module 830 and a generation module 840. Among them:

[0163] A determination module 810 is configured to determine N types of reference elements;

[0164] A sending module 820 is configured to send the N types of reference elements to a backend, where the reference elements are used to generate generated data for a mind map, where the generated data includes nodes and node information of the nodes, where the node information carries the node filling data obtained by the macro model based on the reference elements;

[0165] A receiving module 830 is configured to receive the mind map generation data transmitted from a back-end;

[0166] The generating module 840 is configured to generate a target mind map according to the generated data of the mind map, wherein at least one node information of the target mind map is filled with the diagram information.

[0167] Among them, the generation module 840 is also used to: determine the dependency relationship between nodes based on the node information; render the nodes in the mind map based on the dependency relationship and the node information of the nodes, until the end node rendering is completed to generate the target mind map.

[0168] The generation module 840 is further configured to: start from the root node and render the nodes of the mind map layer by layer based on the dependency relationship; and fill the fill area of the current node according to the text information of the node.

[0169] The receiving module 830 is further configured to receive a data frame determined by the backend based on the generated data, wherein the data frame includes incremental node information of at least one node to be updated.

[0170] Among them, the generation module 840 is also used to: determine the incremental fill data of the node based on the incremental node information in the data frame and the historical node information of the node; fill the node through a rich text editor based on the incremental fill data of the node until the end node rendering is completed to generate the target mind map.

[0171] Among them, the generation module 840 is also used to: determine the node ID of the node; determine the first filling data associated with the node ID from the incremental node information; determine the second filling data associated with the node ID from the historical node information; and obtain the filling incremental data of the node based on the first filling data and the second filling data.

[0172] The generating module 840 is further configured to: determine update flag information of the node; and fill the node in the current mind map using a rich text editor according to the update flag information and the incremental fill data of the node.

[0173] Among them, the generation module 840 is also used to: determine that the node is a new node based on the update flag information; determine the parent node of the node, and connect the node to the parent node of the node to expand the node of the current mind map; based on the incremental fill data, fill the node through a rich text editor.

[0174] Among them, the generation module 840 is also used to: determine that the node is an updated node according to the update flag information; locate the node in the current mind map according to the node ID of the node; and fill the node through a rich text editor based on the incremental fill data.

[0175] The sending module 820 is further configured to: determine the element ID of the reference element; obtain the generation requirement of the mind map; generate request information according to the element ID and the generation requirement; and send the request information to the backend.

[0176] The device 800 is further configured to feed back the canvas data after each rendering to the backend.

[0177] After generating the target mind map based on the generated data of the mind map, the device 800 is further used to: receive an editing operation for the target mind map, and edit the target mind map according to the editing operation to obtain a final mind map.

[0178] Wherein, after generating the target mind map according to the generation data of the mind map, the device 800 is also used to: receive an original text positioning operation for a node in the target mind map; determine the original content in the reference element associated with the node according to the original text positioning operation, and display the original content.

[0179] According to the embodiment of the present disclosure, the device for generating a mind map based on a large model determines N types of reference elements and sends N types of reference elements to the back end. The reference elements are used to generate the generated data of the mind map. The generated data includes nodes and node information of the nodes, wherein the node information carries the filling data of the nodes obtained by the large model based on the reference elements. The generated data of the mind map transmitted by the back end is received, and the nodes in the mind map are rendered according to the dependency relationship and the node information of the nodes until the end node rendering is completed to generate the target mind map. Therefore, the present disclosure supports any type and number of reference elements to generate a mind map, and at least one node information of the target mind map

[0180] Filled with chart information, it improves the visualization of the target mind map and improves the accuracy of the generated target mind map.

[0181] 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.

[0182] Figure 9 A schematic block diagram of an example electronic device 900 that can be used to implement embodiments 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 assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0183] like Figure 9 As shown, the device 900 includes a computing unit 901, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903. Various programs and data required for the operation of the device 900 can also be stored in the RAM 903. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0184] Various components in the device 900 are connected to the I / O interface 905, including an input unit 906, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a magnetic disk, an optical disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the device 900 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0185] The computing unit 901 can be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 901 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 that run machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 901 performs the various methods and processes described above, such as the method for generating a mind map based on a large model. For example, in some embodiments, the method for generating a mind map based on a large model can be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the method for generating a mind map based on a large model described above can be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to execute the method for generating a mind map based on a large model in any other appropriate manner (for example, by means of firmware).

[0186] Various embodiments of the systems and techniques described 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), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes 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.

[0187] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0188] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can 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 can 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.

[0189] 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).

[0190] The systems and techniques described herein can 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 can 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), the Internet, and a blockchain network.

[0191] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0192] The present disclosure also provides a computer program product, including a computer program, which, when executed by a processor, implements the method for generating a mind map based on a large model as described above.

[0193] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0194] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A method for generating a mind map based on a large model, wherein: The method comprises: Determine N types of reference elements; Determining generated data of a mind map based on the N types of reference elements, wherein the generated data includes nodes and node information of the nodes, and the node information carries filling data of the nodes obtained by the large model based on the reference elements; The generated data of the mind map is streamed to a front end to generate a target mind map at the front end, wherein at least one node information of the target mind map is filled with diagram information.

2. The method according to claim 1, wherein The determining of nodes of the mind map and node information of the nodes according to the N types of reference elements includes: Obtaining demand information of the mind map; The node and node information of the node are determined according to the requirement information and the N types of reference elements.

3. The method according to claim 2, wherein: The determining, according to the requirement information and the N types of reference elements, the node and the node information of the node includes: Determining prompt words for generating the mind map according to the requirement information of the mind map; Based on multiple intelligent agents, cross-modal understanding of the N types of reference elements is performed to obtain multimodal structured parsing results; The generated data of the mind map is determined according to the generated prompt words and the structured analysis result.

4. The method according to claim 3, wherein: The step of determining the generated data of the mind map according to the generated prompt words and the structured parsing result includes: Determining a scheduling path for the agent based on the generated prompt words and the structured parsing results; According to the scheduling path, the agent is scheduled, and based on the agent, multimodal extraction is performed on the N types of reference elements to obtain key data related to the node; According to the key data related to the node, the filling data of the node is obtained.

5. The method according to claim 4, wherein The step of obtaining the filling data of the node according to the key data related to the node includes: The key data of the node are screened and adaptively laid out according to the content density of the key data to obtain the filling data of the node.

6. The method according to any one of claims 1 to 5, wherein The determining of N types of reference elements includes: Receive the request information sent by the front end, the request information including the element identification ID of the reference element; The reference element is determined according to the element ID.

7. The method according to any one of claims 1 to 5, wherein The transmitting of the generated data of the mind map to the front end includes: Determining dependencies between nodes based on the node information; Based on the dependency relationship, the generated data of the mind map is streamed to the front end.

8. The method according to any one of claims 1 to 5, wherein The transmitting of the generated data of the mind map to the front end includes: Based on the dependency relationship, determining the node ID of the node to be updated; Determining, according to the node ID, first node information that has been sent by the node and second node information to be sent; Determining incremental node information of the node according to the first node information and the second node information; The data frame is generated according to the incremental node information, and the data frame is sent to the front end.

9. The method according to any one of claims 1 to 5, wherein After transmitting the generated data of the mind map to the front end, the method further includes: Receive canvas data fed back by the front end after each rendering, and transcode and store the canvas data.

10. A method for generating a mind map based on a large model, wherein: The method comprises: Determine N types of reference elements; Sending the N types of reference elements to the backend, where the reference elements are used to generate generated data for a mind map, the generated data including nodes and node information of the nodes, wherein the node information carries the filling data of the nodes obtained by the large model based on the reference elements; Receiving the generated data of the mind map transmitted by the back end; A target mind map is generated according to the mind map generation data, wherein at least one node information of the target mind map is filled with the chart information.

11. The method according to claim 10, wherein: Generating a target mind map according to the generated data of the mind map includes: Determining dependencies between nodes based on the node information; Render the nodes in the mind map according to the dependency relationship and the node information of the nodes until the end node rendering is completed to generate the target mind map.

12. The method according to claim 11, wherein Rendering the nodes in the mind map according to the dependency relationship and the node information of the nodes includes: Starting from the root node, the nodes of the mind map are rendered layer by layer based on the dependency relationship; For the current node, the filling area of the node is filled according to the text information of the node.

13. The method according to any one of claims 10 to 12, wherein: The receiving back-end streaming data of the mind map generation data includes: A data frame determined by the backend based on the generated data is received, wherein the data frame includes incremental node information of at least one node to be updated.

14. The method according to claim 13, wherein Generating a target mind map according to the generated data of the mind map includes: Determining incremental filling data for the node based on the incremental node information in the data frame and the historical node information of the node; According to the incremental filling data of the nodes, the nodes are filled in through a rich text editor until the end node rendering is completed to generate the target mind map.

15. The method according to claim 14, wherein The determining, according to the incremental node information in the data frame and the historical node information of the node, the incremental filling data of the node includes: Determining a node ID of the node; Determining first filling data associated with the node ID from the incremental node information; Determining second filling data associated with the node ID from the historical node information; The incremental filling data of the node is obtained according to the first filling data and the second filling data.

16. The method according to claim 15, wherein The step of filling the node according to the incremental filling data of the node includes: Determining update flag information of the node; The node is filled in the current mind map through a rich text editor according to the update mark information and the incremental filling data of the node.

17. The method according to claim 16, wherein Filling the node in the current mind map according to the update flag information and the incremental filling data of the node includes: Determining, based on the update flag information, that the node is a newly created node; Determine the parent node of the node, and connect the node to the parent node of the node to perform node expansion on the current mind map; Based on the incremental fill data, the node is filled in using a rich text editor.

18. The method according to claim 16, wherein The step of filling and rendering the node with data in the current mind map according to the state information and the incremental filling data of the node includes: Determining, according to the update flag information, that the node is an update node; Locating the node in the current mind map according to the node ID of the node; Based on the incremental fill data, the node is filled in using a rich text editor.

19. The method according to any one of claims 10 to 12, wherein: The sending the N types of reference elements to the backend includes: Determining the element ID of the reference element; Obtain the generation requirements of mind maps; Generate request information according to the element ID and the generation requirement; Send the request information to the backend.

20. The method according to any one of claims 10 to 12, wherein The method further comprises: Feedback the canvas data after each rendering to the backend.

21. The method according to any one of claims 10 to 12, wherein After generating the target mind map according to the generated data of the mind map, the method further includes: An editing operation on the target mind map is received, and the target mind map is edited according to the editing operation to obtain a final mind map.

22. The method according to any one of claims 10 to 12, wherein: After generating the target mind map according to the generated data of the mind map, the method further includes: Receiving an original text positioning operation for a node in the target mind map; According to the original text locating operation, the original content in the reference element associated with the node is determined, and the original content is displayed.

23. A device for generating a mind map based on a large model, wherein: The device comprises: A first determining module, configured to determine N types of reference elements; a second determining module, configured to determine generated data of a mind map based on the N types of reference elements, wherein the generated data includes nodes and node information of the nodes, and the node information carries fill data of the nodes obtained by the large model based on the reference elements; A transmission module is used to transmit the generated data of the mind map to the front end to generate a target mind map at the front end, wherein at least one node information of the target mind map is filled with the chart information.

24. A device for generating a mind map based on a large model, wherein: The device comprises: A determination module, used for determining N types of reference elements; A sending module, configured to send the N types of reference elements to a backend, wherein the reference elements are used to generate generation data of a mind map, wherein the generation data includes nodes and node information of the nodes, wherein the node information carries the filling data of the nodes obtained by the macro model based on the reference elements; A receiving module, configured to receive the mind map generation data transmitted from a back-end; A generating module is used to generate a target mind map according to the generation data of the mind map, wherein at least one node information of the target mind map is filled with the chart information.

25. An electronic device, characterized in that: including processor and memory; The processor reads the executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the method according to any one of claims 1 to 9 or claims 10 to 22.

26. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 9 or claims 10 to 22 is implemented.

27. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 9 or claims 10 to 22.

Images