Prompt generation method and device based on atlas, electronic equipment and storage medium

Abstract

The invention discloses a graph-based prompt generation method and device, electronic equipment and a storage medium, and can be applied to the technical field of graphs. According to the method, the basic graph is constructed according to the basic feature information of the target object, the first tag corresponding to the basic graph is generated, the plurality of level graphs are constructed according to the associated feature information of the target object, the second tag of each level graph is generated, and the first tag and the second tag are adjusted to obtain the third tag; inputting the basic atlas, the plurality of level atlases and the third label into the preset large model, and outputting to obtain a target prompt, so that the demand and the target of the target object can be known by utilizing the atlas construction process, and the output precision of the preset large model is improved by combining the adjusted label; and thus, the generated prompt quality and the prompt generation efficiency can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of graph technology, and in particular to a graph-based prompt generation method and device, electronic device and storage medium. Background Art

[0002] In the related technologies, prompts are used in more and more scenarios in the collaborative office field. At present, the prompt generation process in the collaborative office field mainly relies on predefined templates and manual operations, which require the target object to manually input, select and modify. In the case of limited training capabilities and insufficient internal data, the prompts generated by relying on predefined templates and manual operations are of low quality and low generation efficiency.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a graph-based prompt generation method and device, electronic device and storage medium, which can improve the quality of generated prompts and the efficiency of prompt generation.

[0005] To achieve the above object, one aspect of an embodiment of the present application proposes a method for generating a prompt based on a graph, the method comprising the following steps:

[0006] Obtain basic feature information of the target object;

[0007] Constructing a basic graph according to the basic feature information and generating a first label corresponding to the basic graph;

[0008] Acquire the associated feature information of the target object;

[0009] Constructing a plurality of level maps according to the associated feature information and generating a second label for each level map, wherein the plurality of level maps include a primary map, a secondary map, a tertiary map, and a quaternary map;

[0010] Performing hierarchical adjustment on the fourth label to obtain a third label, wherein the fourth label includes the first label and the second label;

[0011] The basic map, the plurality of level maps and the third label are input into a preset large model to generate a target prompt.

[0012] In some embodiments, constructing a basic map according to the basic feature information includes:

[0013] Performing vectorization processing on the basic feature information to obtain vector information;

[0014] The basic map is constructed with the vector information of the current target object as the center.

[0015] In some embodiments, constructing several level graphs according to the associated feature information includes:

[0016] Constructing the primary graph according to the first interaction information in the associated feature information;

[0017] Constructing the secondary graph according to the related object information in the associated feature information;

[0018] Constructing the three-level graph according to the second interaction information in the associated feature information;

[0019] The four-level graph is constructed according to the feedback information in the associated feature information.

[0020] In some embodiments, constructing the primary graph according to the first interaction information in the associated feature information includes:

[0021] Acquire document operation information, first project collaboration information or task update information in the first interaction information;

[0022] Taking the document operation information, the first project collaboration information or the task update information as a first node;

[0023] Analyzing a first association relationship between the document operation information, the first project collaboration information or the task update information and the target object;

[0024] The primary graph is constructed according to the first node and the association relationship.

[0025] In some embodiments, constructing the secondary graph according to the related object information in the associated feature information includes:

[0026] Acquire basic information of the related object, communication information, second project collaboration information or shared resource information in the related object information;

[0027] Taking the related object basic information, the communication information, the second project collaboration information or the shared resource information as a second node;

[0028] Analyzing a second association relationship between the related object basic information, the communication information, the second project collaboration information or the shared resource information and the target object;

[0029] The secondary graph is constructed according to the second node and the second association relationship.

[0030] In some embodiments, constructing the three-level graph according to the second interaction information in the associated feature information includes:

[0031] Acquire input habit information and input content adjustment information in the second interaction information;

[0032] Determine the object's intention and motivation and the object's behavior path according to the input habit information and the input content adjustment information;

[0033] The object's intention and motivation are used as the third node;

[0034] Using the object behavior path as a connecting edge;

[0035] The three-level graph is constructed according to the third node and the connecting edge.

[0036] In some embodiments, the step of performing hierarchical adjustment on the fourth label to obtain the third label includes:

[0037] Get key indicators at each level;

[0038] Associating the key indicator with the fourth tag;

[0039] According to the associated key indicators, a judgment matrix is ​​constructed based on the hierarchical analysis method;

[0040] Determining the relative importance of different to-be-processed maps according to the judgment matrix, wherein the to-be-processed maps include the basic map, the primary map, the secondary map, the tertiary map, and the quaternary map;

[0041] The fourth label is adjusted according to the relative importance to obtain the third label.

[0042] To achieve the above-mentioned purpose, another aspect of the embodiment of the present application provides a prompt generation device based on a graph, the device comprising:

[0043] The first module is used to obtain basic feature information of the target object;

[0044] The second module is used to construct a basic map according to the basic feature information and generate a first label corresponding to the basic map;

[0045] The third module is used to obtain the associated feature information of the target object;

[0046] A fourth module is used to construct a plurality of level maps according to the associated feature information and generate a second label for each level map, wherein the plurality of level maps include a primary map, a secondary map, a tertiary map and a quaternary map;

[0047] A fifth module is used to perform hierarchical adjustment on the fourth label to obtain a third label, wherein the fourth label includes the first label and the second label;

[0048] The sixth module is used to input the basic map, the several level maps and the third label into the preset large model to generate a target prompt.

[0049] To achieve the above object, another aspect of an embodiment of the present application provides an electronic device, including:

[0050] at least one processor;

[0051] at least one memory for storing at least one program;

[0052] When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0053] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0054] The embodiments of the present application include at least the following beneficial effects: The present application provides a graph-based prompt generation method and device, an electronic device and a storage medium, which constructs a basic graph according to the basic feature information of the target object and generates a first label corresponding to the basic graph, and constructs several level graphs according to the associated feature information of the target object and generates a second label for each level graph, then adjusts the first label and the second label to obtain a third label, and then inputs the basic graph, several level graphs and the third label into a preset large model and outputs the target prompt, so that the needs and goals of the target object can be understood by using the graph construction process, and then the output accuracy of the preset large model can be improved by combining the adjusted labels, thereby effectively improving the generated prompt quality and prompt generation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a flow chart of a method for generating prompts based on a graph provided in an embodiment of the present application;

[0056] Figure 2 It is a flowchart of constructing a basic map according to basic feature information provided in an embodiment of the present application;

[0057] Figure 3 It is a flowchart of constructing several level maps according to the associated feature information provided in the embodiment of the present application;

[0058] Figure 4 It is a flow chart for constructing a primary map provided in the embodiment of the present application;

[0059] Figure 5 It is a flow chart for constructing a secondary map provided in an embodiment of the present application;

[0060] Figure 6 It is a flow chart for constructing a three-level map provided in an embodiment of the present application;

[0061] Figure 7 is a flow chart of performing hierarchical adjustment on the fourth label to obtain the third label provided by an embodiment of the present application;

[0062] Figure 8 is a schematic diagram of a judgment matrix provided in an embodiment of the present application;

[0063] Fig. 9 It is a structural schematic diagram of a graph-based prompt generation device provided in an embodiment of the present application;

[0064] Fig.10 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application, they are only examples of devices and methods consistent with some aspects of the embodiments of the present application.

[0066] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0067] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0068] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0069] Before describing the embodiments of the present application in detail, some nouns and terms involved in the embodiments of the present application are first described. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations:

[0070] Prompt means prompting, which is issuing instructions to the AI ​​model through text or other forms to obtain the desired output. Simply put, Prompt is to tell the AI ​​model what to do by inputting some text or voice, and then the model will generate the corresponding output according to these instructions. In the field of natural language processing (NLP), Prompt is often used to guide the pre-trained large language model (LLM) to generate appropriate answers. For example, in a chat interaction, the subject can provide a question or topic as a prompt and let the model generate the corresponding answer.

[0071] The analytic hierarchy process (AHP) refers to treating a complex multi-objective decision-making problem as a system, decomposing the goal into multiple goals or criteria, and then decomposing it into several levels of multiple indicators (or criteria, constraints). The single ranking (weight) and total ranking of the levels are calculated through the fuzzy quantification method of qualitative indicators as a systematic method for optimizing decision-making with goals (multiple indicators) and multiple plans.

[0072] In the field of collaborative office, the application scenarios of Prompt technology are increasing. It can improve the level of office automation by providing assistance at key functional points, fully understanding the content of the office system, and supporting automated business processing. However, the current Prompt generation method faces the following problems: First, due to the insufficient knowledge reserve and inaccurate expression of the subject in the relevant field, the generated prompts may be misleading or incomplete; second, the current large model fails to integrate the knowledge system of a specific customer system, which may lead to deviations in the generated prompts, and then the subject needs to make corresponding adjustments, thus affecting the efficiency of prompt generation.

[0073] Although there are some Prompt learning and implementation applications in the field of collaborative office, Prompt generation usually relies on predefined templates and manual operations. This method requires the subject to input, select and modify, which results in low quality and low efficiency of generating prompts when the training capacity is limited and the system's internal data is insufficient.

[0074] In view of this, a graph-based prompt generation method and device, an electronic device and a storage medium are provided in the embodiments of the present application. The present application constructs a basic graph according to the basic feature information of the target object and generates a first label corresponding to the basic graph. At the same time, several level graphs are constructed according to the associated feature information of the target object and a second label for each level graph is generated. The first label and the second label are adjusted to obtain a third label. The basic graph, several level graphs and the third label are input into a preset large model and then output to obtain a target prompt. In this way, the graph construction process can be used to understand the needs and goals of the target object, and the output accuracy of the preset large model can be improved by combining the adjusted labels. Then, the preset large model technology can be used to effectively improve the generated prompt quality and prompt generation efficiency.

[0075] The graph-based prompt generation method provided in the embodiment of the present application relates to the field of graph technology. The graph-based prompt generation method provided in the embodiment of the present application can be applied to a terminal, can also be applied to a server, and can also be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or it can be configured as a server cluster or a distributed system composed of multiple physical servers, and can also be configured to provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms and other basic cloud computing services. The server can also be a node server in the blockchain network; the software can be an application that implements a graph-based prompt generation method, etc., but is not limited to the above forms.

[0076] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0077] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0078] Figure 1 is an optional flowchart of the method for generating prompts based on a graph provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S110 to S160:

[0079] Step S110, obtaining basic feature information of the target object;

[0080] Step S120: construct a basic graph according to the basic feature information and generate a first label corresponding to the basic graph;

[0081] Step S130, obtaining associated feature information of the target object;

[0082] Step S140: constructing a plurality of level graphs according to the associated feature information and generating a second label for each level graph, wherein the plurality of level graphs include a primary graph, a secondary graph, a tertiary graph, and a quaternary graph;

[0083] Step S150: hierarchically adjust the fourth label to obtain a third label, wherein the fourth label includes the first label and the second label;

[0084] Step S160: input the basic map, several level maps and the third label into the preset large model to generate a target prompt.

[0085] It is understandable that the basic feature information may include, but is not limited to, the name, age, account information in the office scene, length of service, position, operating habits, and other information of the target object. The associated feature information may include, but is not limited to, the name, age, account information in the office scene, length of service, position, operating habits, project communication, and other information of the target object's corresponding colleague. The preset large model may be a language model that can perform automatic question-and-answer functions in the field of artificial intelligence. The labels in each graph are used to indicate the category of information corresponding to each node in the graph.

[0086] This embodiment constructs a basic graph and its labels based on the basic feature information of the target object, and constructs several levels of graphs and their labels based on the associated feature information of the target object, so that the target object can clearly understand his or her needs and goals and accurately express his or her intentions. Then, by adjusting the labels in layers and combining them with all the graphs, they are input into the preset large model for prompt generation, thereby utilizing the characteristics of the preset large model to improve the quality of generated prompts and the efficiency of prompt generation.

[0087] In the embodiments of the present application, Figure 2 As shown, the process of constructing a basic map according to basic feature information includes but is not limited to steps S210 to S220:

[0088] Step S210, performing vectorization processing on the basic feature information to obtain vector information;

[0089] Step S220: construct a basic map centered on the vector information of the current target object.

[0090] It is understandable that when the target object enters the office system through the interactive interface, the backend will automatically obtain the basic feature information of the target object, and then convert the basic feature information into vector data, and then generate a personalized basic map corresponding to the target object based on the basic feature information of the target object, focusing on the target object's age, length of service, position, daily operation habits in the collaborative office system, etc. The basic map can cover all the operation behaviors of the target object, including but not limited to the operation behaviors that have been displayed and potential operation behaviors.

[0091] Specifically, the basic map represents the relationship network formed with the basic characteristic information of the target object as the core. Among them, the basic characteristic information of this embodiment includes but is not limited to name, contact information, role positioning, job description, business direction, records of personal habits, template-related content, etc. It can be understood that roles and positions are the basis of the map, which are used to clarify the responsibilities and expectations of each member in the team. Business direction is the key area to guide the work of members, and it should be closely aligned with the goals and strategies of the organization. Records of personal habits are essential for team collaboration and task coordination, such as working hours, communication preferences, and decision-making styles. Template-related content Template-related content helps to standardize workflows and improve efficiency, such as document templates, report formats, and project management tools.

[0092] This embodiment obtains the basic data, configuration data, age, position, role, habits and other multi-dimensional information of the object, and pre-processes, cleans, standardizes, integrates and formats the multi-dimensional information to ensure the quality and consistency of the data. Then, the attributes such as the age, position, role and habits of the object are encoded as node attributes in the basic graph, and the interactive relationship between objects is represented by edges, thereby constructing a multi-level, interconnected complex network structure as the basic graph.

[0093] In order to expand and describe the basic graph of an individual in detail, this embodiment can also add information such as the target object's skills and expertise, educational background, work experience, and hobbies. Among them, skills and expertise can help the big model understand the core capabilities and suitable types of work of the object. Educational background and work experience provide background information on the object's professional qualifications and career development. Hobbies help the big model understand the team cohesion formed by the object. In addition, the basic graph also needs to consider data protection and privacy to ensure the security of the object's information.

[0094] In the embodiments of the present application, Figure 3 As shown, the process of constructing several level maps according to the associated feature information includes but is not limited to steps S310 to S340:

[0095] Step S310: construct a primary graph according to the first interactive information in the associated feature information;

[0096] Step S320, constructing a secondary graph according to the related object information in the associated feature information;

[0097] Step S330: construct a three-level graph according to the second interaction information in the associated feature information;

[0098] Step S340: construct a four-level graph according to the feedback information in the associated feature information.

[0099] It is understandable that, after collecting all the associated feature information of the target object through the 4-level collection mode, this embodiment can respectively construct a map around the object's recent data, features and operation data with good intimacy with the object, the operation content of the interactive data, and the feedback data of the result content, thereby forming a 4-level map. Exemplarily, the activity records of the object in the collaborative office platform are collected, such as document editing history, meeting participation records, task allocation and completion status and key content, etc.; the interaction frequency between objects is analyzed, such as frequent collaborative team members and the frequency of communication through emails, instant messages or document comments, etc.; the interactive behavior of the object on the collaborative platform is recorded, such as co-editing documents, sharing files, project discussions, task allocation, etc.; and the satisfaction survey, object feedback, and improvement suggestions of the object after using the collaborative platform functions are collected. After constructing different levels of maps based on the collected information, the object can have a clearer understanding of its own needs and goals, and then guide the prompt generation process of the preset large model.

[0100] In the embodiments of the present application, Figure 4 As shown, the process of constructing a primary graph according to the first interactive information in the associated feature information includes but is not limited to steps S410 to S440:

[0101] Step S410: Obtain document operation information, first project collaboration information or task update information in the first interaction information;

[0102] Step S420: taking the document operation information, the first project collaboration information or the task update information as the first node;

[0103] Step S430, analyzing a first association relationship between the document operation information, the first project collaboration information or the task update information and the target object;

[0104] Step S440: construct a primary graph based on the first node and the association relationship.

[0105] It can be understood that the first-level graph is an indicative graph set formed by deeply mining the interaction content between common functions and target objects in the office system. This graph set covers the operational association relationship of the target object when using common functions, and can enhance the relevance and applicability between node information in the first-level graph by integrating contextual environment and functional knowledge. In the first-level graph, objects and operation information will be used as nodes, while object data and operation behavior will be used as edges of the first-level graph to identify object behavior patterns, discover the intrinsic connection between data, and generate labels for information in the first-level graph accordingly.

[0106] Specifically, the first-level graph can be mainly based on analyzing and mining the data relationship in the associated feature information to show the intrinsic connection between the data. It can be understood that this embodiment uses the document operation information, the first project collaboration information or the task update information as the basic node (first node), analyzes the correlation between this information and the target object, and identifies the interaction between the document and the project, the target object and the task (first association relationship), sorts out the relevant paths between the nodes, and connects different basic nodes with edges to obtain a first-level graph. Among them, the document operation information includes but is not limited to the access, editing, and sharing of the document. After obtaining the document operation information, the first project collaboration information and the task update information, this embodiment uses natural language processing technology to automatically extract keywords and concepts from this information, thereby enhancing the classification and interactivity of the data. At the same time, this embodiment can also integrate visualization tools so that the target object can intuitively understand the complex data relationship and clearly understand its own goals and demand information.

[0107] In the embodiments of the present application, Figure 5 As shown, the process of constructing a secondary graph according to the related object information in the associated feature information includes but is not limited to steps S510 to S540:

[0108] Step S510, obtaining basic information of the related object, communication information, second project collaboration information or shared resource information in the related object information;

[0109] Step S520: taking the basic information of the relevant object, the communication information, the second project collaboration information or the shared resource information as the second node;

[0110] Step S530, analyzing the second association relationship between the basic information, communication information, second project collaboration information or shared resource information of the related object and the target object;

[0111] Step S540: construct a secondary graph according to the second node and the second association relationship.

[0112] It is understandable that the related objects may be colleagues, neighbors, family members, etc. of the target object. Shared resource information may be resource information of the company's intranet, shared information between family members, or shared information of the internal system of the community. After obtaining the basic information, communication information, second project collaboration information, or shared resource information of the related objects, this embodiment can analyze the preferences and work habits between objects by focusing on the basic information of the related objects, the communication frequency in the communication information, the second collaboration project information, and the shared resource information, and then obtain the collaboration mode and proximity, and then construct a secondary graph based on the second association relationship. Specifically, since the information of this application involves data function relationship types, such as project participation, task allocation, and information sharing, the related relationships (second association relationships) are connected to form objects with common goals, and after identifying the proximity between objects, the large model is used to extract keywords, and object classification is constructed, and labels for the corresponding information of each node in the secondary graph are generated according to the classification proximity of objects, system functions, and services, so that the information in the secondary graph can be described according to priority.

[0113] In the embodiments of the present application, Figure 6 As shown, the process of constructing a three-level graph according to the second interactive information in the associated feature information includes but is not limited to steps S610 to S640:

[0114] Step S610: obtaining input habit information or input content adjustment information in the second interaction information;

[0115] Step S620, determining the object's intention and motivation and the object's behavior path according to the input habit information and the input content adjustment information;

[0116] Step S630, taking the object's intention and motivation as the third node;

[0117] Step S640: Use the object behavior path as a connecting edge;

[0118] Step S650: construct a three-level graph based on the third node and the connecting edge.

[0119] It can be understood that this embodiment improves the office system's ability to understand and respond to object behavior, and collects the object's second interaction information (current interaction information and historical interaction information). For example, after obtaining input habit information and content adjustment information, this information is analyzed to obtain the object's intention, motivation and object behavior path in the office system, and provide timely feedback, and then construct a three-level map based on the intention, motivation and object behavior path.

[0120] Specifically, the labels in the three-level graph can be set with corresponding labels after extracting keywords and commonly used commands from these information through a preset large model, and assigning weights to these labels, so that the size of the object demand can be represented by the size of the weight, and the object behavior pattern can be deeply analyzed to predict the possible future operations and needs of the object in the office system. Based on the setting of the three-level graph and its corresponding labels, personalized prompt recommendations can be made, and the generation of content in personalized prompts can be supported according to the user's operating habits and preferences. In addition, this embodiment can also adjust the label priority based on the real-time input interaction information of the object, so that the label can be adapted to the object's intention. Among them, the connection relationship between the nodes in the three-level graph can also be fine-tuned based on the feedback information of the object, so as to effectively improve the accuracy of the information represented by the three-level graph.

[0121] In the embodiment of the present application, the four-level graph can be constructed by using a circular feedback method according to the feedback information of the object on the generated prompt, wherein the labels in the four-level graph can also be screened by template optimization and sorting, and the accuracy of label sorting is improved according to the matching degree between the label and the actual information of the object. It can be understood that the four-level graph is a graph obtained by adjusting the feedback information and satisfaction corresponding to the generated content of the prompt. This embodiment collects the feedback information of the object in real time and adjusts the node connection relationship in the graph in time to optimize the content generated by the prompt, thereby improving the experience of the object in the office system. Specifically, this embodiment takes the generation of dynamic prompt as the core, allowing the graph to self-adjust and optimize according to the object interaction information, so that the feedback information represented by the four-level graph is close to the object's goals and needs, thereby improving the quality of prompt generation. Exemplarily, when the object is particularly satisfied with a certain generated content in the collected information, this embodiment can increase the weight of the satisfactory information in the label, so that the adjusted weight can be used to sort the labels, thereby improving the accuracy of the information represented by the four-level graph.

[0122] It can be seen that the embodiment of the present application records the operation steps of the object in a hierarchical standard manner around the 5-level atlas, performs weight and vector processing on the commonly used content, and processes and records the relevant results in the form of summary labels, so as to achieve the distinction and use of object association relationships by level and weight label, and then obtain the corresponding level of the atlas and its label. Exemplarily, when defining object behaviors at different levels, object behaviors are divided into levels such as basic operations, collaborative activities, management decisions, and innovative contributions. Weights are assigned to the behaviors of each level to reflect the importance of objects in each level to team collaboration and project results. According to the object change value, the corresponding label is identified to monitor the behavior changes of the object in document editing, task completion, meeting participation, etc., and the label is given according to the extracted content theme. For frequently changing user behaviors, such as frequently editing documents or frequently participating in meetings, higher weights can be given, so that the role of the label corresponding information can be determined based on the weight size, so that the label can more accurately express the category of object information in the atlas.

[0123] In the embodiments of the present application, Figure 7 As shown, the process of hierarchically adjusting the fourth label to obtain the third label includes but is not limited to steps S710 to S750:

[0124] Step S710, obtaining key indicators of each level;

[0125] Step S720: Associating the key indicator with the fourth tag;

[0126] Step S730: construct a judgment matrix based on the associated key indicators based on the hierarchical analysis method;

[0127] Step S740: judging the relative importance of different to-be-processed spectra according to the judgment matrix, wherein the to-be-processed spectra include basic spectra, primary spectra, secondary spectra, tertiary spectra and quaternary spectra;

[0128] Step S750: adjust the fourth label according to the relative importance to obtain the third label.

[0129] It is understandable that the hierarchical analysis method of this embodiment can adopt the AHP method, based on which a judgment matrix is ​​generated, in which the key indicator and the fourth label are associated. After obtaining the judgment matrix, this embodiment calculates the relative importance of different to-be-processed graphs to adjust the fourth label, so as to obtain a more accurately positioned third label.

[0130] For example, in the AHP method, the label adjustment is mainly carried out around the hierarchy clarity (C1), label accuracy (C2), relevance and consistency (C3), object usability (C4) and data availability (C5) in the standard layer. For each standard (C1 to C5) in the criterion layer, a judgment matrix is ​​constructed to compare the relative importance of different to-be-processed graphs (P1, P2, P3, P4, P5), and finally the debugging is carried out according to the Figure 8 The judgment matrix is ​​shown. Figure 8 The matrix shown indicates that in terms of hierarchical clarity, P1 is 3 times more important than P2 and 5 times more important than P3; P2 is 2 times more important than P3. For each judgment matrix, this embodiment needs to calculate the weight of each graph to be processed and perform a consistency check to ensure the consistency of the judgment. Among them, the consistency check can be completed by calculating the consistency ratio (CR), and when the CR is less than 0.1, it is considered to be acceptable for consistency. The weight of the graph to be processed under each criterion is obtained, and the comprehensive weight of each graph to be processed is obtained by weighted summation. The preset large model is guided to generate target prompts according to the graph to be processed corresponding to the comprehensive weight, so that the target prompts meet the needs and goals of the object.

[0131] Reference Fig. 9 , the embodiment of the present application provides a prompt generation device based on a graph, the device comprising:

[0132] The first module 910 is used to obtain basic feature information of the target object;

[0133] The second module 920 is used to construct a basic map according to the basic feature information and generate a first label corresponding to the basic map;

[0134] The third module 930 is used to obtain the associated feature information of the target object;

[0135] The fourth module 940 is used to construct a plurality of level maps according to the associated feature information and generate a second label of each level map, wherein the plurality of level maps include a primary map, a secondary map, a tertiary map and a quaternary map;

[0136] A fifth module 950 is configured to perform hierarchical adjustment on the fourth label to obtain a third label, wherein the fourth label includes the first label and the second label;

[0137] The sixth module 960 is used to input the basic map, several level maps and the third label into the preset large model to generate the target prompt.

[0138] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0139] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned graph-based prompt generation method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0140] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0141] See also Fig.10 , Fig.10 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0142] The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0143] The memory 1020 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1020 can store an operating system and other applications. When the technical solution provided in the embodiments of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1020, and the processor 1010 calls and executes the graph-based prompt generation method of the embodiment of this application;

[0144] Input / output interface 1030, used to implement information input and output;

[0145] The communication interface 1040 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WI FI, Bluetooth, etc.);

[0146] Bus 1050 , which transmits information between various components of the device (e.g., processor 1010 , memory 1020 , input / output interface 1030 , and communication interface 1040 );

[0147] The processor 1010 , the memory 1020 , the input / output interface 1030 , and the communication interface 1040 are connected to each other in communication within the device via a bus 1050 .

[0148] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the above-mentioned graph-based prompt generation method is implemented.

[0149] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0150] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0151] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0152] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0153] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0154] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0155] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0156] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0157] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0158] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0159] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0160] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0161] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for generating prompts based on a graph, characterized in that: The method comprises the following steps: Obtain basic feature information of the target object; Constructing a basic graph according to the basic feature information and generating a first label corresponding to the basic graph; Acquire the associated feature information of the target object; Constructing a plurality of level maps according to the associated feature information and generating a second label for each level map, wherein the plurality of level maps include a primary map, a secondary map, a tertiary map, and a quaternary map; Performing hierarchical adjustment on the fourth label to obtain a third label, wherein the fourth label includes the first label and the second label; The basic map, the plurality of level maps and the third label are input into a preset large model to generate a target prompt.

2. The method according to claim 1, characterized in that The step of constructing a basic graph according to the basic feature information includes: Performing vectorization processing on the basic feature information to obtain vector information; The basic map is constructed with the vector information of the current target object as the center.

3. The method according to claim 1, characterized in that The constructing of several level graphs according to the associated feature information includes: Constructing the primary graph according to the first interaction information in the associated feature information; Constructing the secondary graph according to the related object information in the associated feature information; Constructing the three-level graph according to the second interaction information in the associated feature information; The four-level graph is constructed according to the feedback information in the associated feature information.

4. The method according to claim 3, characterized in that The step of constructing the primary graph according to the first interaction information in the associated feature information includes: Acquire document operation information, first project collaboration information or task update information in the first interaction information; Taking the document operation information, the first project collaboration information or the task update information as a first node; Analyzing a first association relationship between the document operation information, the first project collaboration information or the task update information and the target object; The primary graph is constructed according to the first node and the association relationship.

5. The method according to claim 3, characterized in that: The step of constructing the secondary graph according to the related object information in the associated feature information includes: Acquire basic information of the related object, communication information, second project collaboration information or shared resource information in the related object information; Taking the related object basic information, the communication information, the second project collaboration information or the shared resource information as a second node; Analyzing a second association relationship between the related object basic information, the communication information, the second project collaboration information or the shared resource information and the target object; The secondary graph is constructed according to the second node and the second association relationship.

6. The method according to claim 3, characterized in that The constructing the three-level graph according to the second interaction information in the associated feature information includes: Acquire input habit information and input content adjustment information in the second interaction information; Determine the object's intention and motivation and the object's behavior path according to the input habit information and the input content adjustment information; The object's intention and motivation are used as the third node; Using the object behavior path as a connecting edge; The three-level graph is constructed according to the third node and the connecting edge.

7. The method according to claim 1, characterized in that The step of performing hierarchical adjustment on the fourth label to obtain the third label includes: Get key indicators at each level; Associating the key indicator with the fourth tag; According to the associated key indicators, a judgment matrix is ​​constructed based on the hierarchical analysis method; Determining the relative importance of different to-be-processed maps according to the judgment matrix, wherein the to-be-processed maps include the basic map, the primary map, the secondary map, the tertiary map, and the quaternary map; The fourth label is adjusted according to the relative importance to obtain the third label.

8. A graph-based prompt generation device, characterized in that: The device comprises: The first module is used to obtain basic feature information of the target object; The second module is used to construct a basic map according to the basic feature information and generate a first label corresponding to the basic map; The third module is used to obtain the associated feature information of the target object; A fourth module is used to construct a plurality of level maps according to the associated feature information and generate a second label for each level map, wherein the plurality of level maps include a primary map, a secondary map, a tertiary map and a quaternary map; A fifth module is used to perform hierarchical adjustment on the fourth label to obtain a third label, wherein the fourth label includes the first label and the second label; The sixth module is used to input the basic map, the several level maps and the third label into the preset large model to generate a target prompt.

9. An electronic device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images