Graph generation method, device, equipment and medium based on multi-round dialogue

Abstract

The present application provides a graph generation method, device, equipment and medium based on multi-round dialogues, which relates to the field of graph technology and is applied to a server. The method includes: receiving an N-round dialogue message input by a user terminal for an N-round dialogue; querying a database and performing data preprocessing according to the N-round dialogue message to obtain graph data to be processed corresponding to the N-round dialogue; creating a graph to be processed according to the graph data to be processed; compressing and storing the graph to be processed to obtain compressed graph data, wherein the stored compressed graph data is used to respond to multi-round dialogue messages input by the user terminal in multiple dialogues after the N-round dialogue; sending the compressed graph data to the user terminal so that the user terminal can decompress and reconstruct the compressed graph data to obtain a reconstructed graph, and displaying the reconstructed graph.

Description

Technical Field

[0001] The present application relates to the field of graph technology, and in particular to a graph generation method, device, equipment and medium based on multi-round dialogue. Background Art

[0002] A graph is a data structure used to represent complex relationships between entities (such as people, places, and things). It constructs a network structure through nodes (representing entities) and edges (representing relationships), so that the intrinsic connections between data can be intuitively displayed. Graphs are widely used in many fields, such as social network analysis, knowledge graph construction, bioinformatics, etc. In graph image display, graphs are used to organize and display large amounts of data so that users can understand and analyze them more easily.

[0003] A multi-round conversation refers to a series of interactions between a user and a system, each of which is based on the content of the previous one or several previous interactions. In a multi-round conversation, the system needs to understand the user's intentions, remember historical conversation information, and generate appropriate responses based on this information.

[0004] In recent years, the graph rendering solution strategy applied to multi-round conversations has been implemented by integrating a variety of technologies and methods. It is mainly to solve the problem of displaying a large amount of complex data relationships to users when processing graph data. Often, excessive data volume will affect performance and occupy service resources.

[0005] Combining multiple rounds of conversations with graph data rendering, each round of conversation corresponds to the display of a large amount of graph data. In actual applications, if the amount of graph data is too large, it will cause pressure on the browser. In severe cases, it may cause the browser to freeze, affecting the user terminal experience. Summary of the invention

[0006] In view of this, the embodiments of the present application provide a method, apparatus, device and medium for generating a graph based on multi-round conversations, so as to overcome the above problems or at least partially solve the above problems.

[0007] A first aspect of an embodiment of the present application provides a graph generation method based on multi-round dialogues, which is applied to a server, and the method includes: Receiving an N-th round of dialogue message input by a user terminal for the N-th round of dialogue; According to the Nth round of dialogue message, query the database and perform data preprocessing to obtain the to-be-processed graph data corresponding to the Nth round of dialogue; Creating a graph to be processed according to the graph data to be processed; Compressing and storing the graph to be processed to obtain compressed graph data, wherein the stored compressed graph data is used to respond to multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; The compressed graph data is sent to the user terminal, so that the user terminal decompresses and reconstructs the compressed graph data to obtain a reconstructed graph, and displays the reconstructed graph.

[0008] Optionally, the method further comprises: receiving an N+1th round of dialogue message input by the user terminal for the N+1th round of dialogue; Acquire from the database a plurality of compressed graph data corresponding to the first N rounds of dialogue before the N+1th round of dialogue; According to the N+1th round of dialogue message, query the database to obtain the collected data corresponding to the N+1th round of dialogue message, wherein the collected data corresponding to the N+1th round of dialogue message is obtained by querying the message content in the N+1th round of dialogue message that is different from the first N rounds of dialogue messages corresponding to the first N rounds of dialogue; The to-be-processed graph data corresponding to the N+1th round of dialogues are obtained according to the plurality of compressed graph data corresponding to the first N rounds of dialogues and the collected data corresponding to the messages of the N+1th round of dialogues.

[0009] Optionally, compressing and storing the to-be-processed graph to obtain compressed graph data includes: Removing redundant information between spectra in the spectrum to be processed; By means of inter-band compression, the data dimension and redundancy of the to-be-processed graph are reduced to obtain a preliminarily compressed to-be-processed graph; The preliminarily compressed to-be-processed graph is in-band compressed to reduce the amount of data in the preliminarily compressed to-be-processed graph to obtain the compressed graph data; During the inter-band compression process and the intra-band compression process, sparse coding is performed, and dictionary elements are used to obtain data representation of the graph to be processed.

[0010] Optionally, querying a database and performing data preprocessing according to the N-th round of dialogue message to obtain the to-be-processed graph data corresponding to the N-th round of dialogue includes: Parsing and verifying the Nth round of dialogue message; According to the N-round dialogue message after the analysis and verification, the original graph data is queried in the database, wherein the original graph data at least includes a plurality of compressed graph data corresponding to the first N-1 rounds of dialogue and a collection message corresponding to the N-round dialogue message, and the collection data corresponding to the N-round dialogue message is obtained by querying the message content in the N-round dialogue message that is different from the first N-1 round dialogue message corresponding to the first N-1 round dialogue; Performing data preprocessing on the original graph data, wherein the data preprocessing includes data cleaning and removing noise, outliers and redundant information; generating preprocessed original graph data by defining entities, attributes and relationships according to the preprocessed original graph data; The preprocessed original graph data is simplified to obtain the graph data to be processed corresponding to the Nth round of dialogue.

[0011] A second aspect of an embodiment of the present application provides a graph generation method based on multi-round dialogues, which is applied to a user terminal, and the method includes: Receive compressed graph data for the Nth round of dialogue message sent by the server, wherein the compressed graph data is the Nth round of dialogue message input by the user terminal for the Nth round of dialogue received by the server; query the database and perform data preprocessing according to the Nth round of dialogue message to obtain the graph data to be processed corresponding to the Nth round of dialogue; create a graph to be processed according to the graph data to be processed; compress and store the graph to be processed, wherein the stored compressed graph data is used to respond to multiple rounds of dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; Processing the compressed graph data according to the decompression algorithm and parameters corresponding to the compression by the server to obtain a reconstructed graph corresponding to the Nth round of dialogue messages; Optionally, the method further comprises: Receive the compressed graph data corresponding to the N+1th round of dialogue message sent by the server; Based on the compressed graph data corresponding to the N+1th round of dialogue message, a reconstructed graph corresponding to the N+1th round of dialogue message is obtained; The reconstructed graph corresponding to the N+1-th round of dialogue message is displayed to the user in a graphical manner, wherein the graph contents that are different from the reconstructed graph corresponding to the N+1-th round of dialogue message and the reconstructed graph corresponding to the N-th round of dialogue message are displayed in a manner that is different from the graph contents of the reconstructed graph corresponding to the N-th round of dialogue message.

[0012] A third aspect of an embodiment of the present application provides a graph generation device based on multi-round dialogues, which is applied to a server, and the device includes: A dialogue message receiving module, used to receive the Nth round of dialogue message input by the user terminal for the Nth round of dialogue; A to-be-processed graph data generating module is used to query a database and perform data preprocessing according to the N-th round of dialogue messages to obtain the to-be-processed graph data corresponding to the N-th round of dialogue; A to-be-processed graph creation module, used to create a to-be-processed graph based on the to-be-processed graph data; A compression module, used for compressing and storing the to-be-processed graph to obtain compressed graph data, wherein the stored compressed graph data is used to respond to multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; A sending module is used to send the compressed graph data to the user terminal, so that the user terminal can decompress and reconstruct the compressed graph data to obtain a reconstructed graph, and display the reconstructed graph.

[0013] A fourth aspect of the embodiments of the present application provides a graph generation device based on multi-round dialogues, which is applied to a user terminal, and the device includes: A compressed graph data receiving module is used to receive compressed graph data for the Nth round of dialogue message sent by the server, wherein the compressed graph data is the Nth round of dialogue message input by the user terminal for the Nth round of dialogue received by the server; according to the Nth round of dialogue message, query the database and perform data preprocessing to obtain the graph data to be processed corresponding to the Nth round of dialogue; according to the graph data to be processed, create a graph to be processed; compress and store the graph to be processed, and the stored compressed graph data is used to respond to the multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; A reconstruction module, used to process the compressed graph data according to the decompression algorithm and parameters corresponding to the compression by the server, to obtain a reconstructed graph corresponding to the Nth round of dialogue messages; The display module is used to display the reconstructed reconstruction map to the user in a graphical manner.

[0014] In a fifth aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the graph generation method based on multi-round dialogue as described in the first aspect or the second aspect.

[0015] In a sixth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored, wherein when the computer program is executed by a processor, the graph generation method based on multi-round dialogue as described in the first aspect or the second aspect is implemented.

[0016] In a seventh aspect of an embodiment of the present application, a computer program product is provided, including a computer program, which, when executed by a processor, implements the graph generation method based on multi-round dialogue as described in the first aspect or the second aspect.

[0017] Beneficial effects of this application: The embodiment of the present application provides a graph generation method, device, equipment and medium based on multi-round dialogues, which are applied to a server, and the method includes: receiving an N-th round dialogue message input by a user terminal for an N-th round dialogue; querying a database and performing data preprocessing according to the N-th round dialogue message to obtain graph data to be processed corresponding to the N-th round dialogue; creating a graph to be processed according to the graph data to be processed; compressing and storing the graph to be processed to obtain compressed graph data, and the stored compressed graph data is used to respond to multi-round dialogue messages input by the user terminal in multiple dialogues after the N-th round dialogue; sending the compressed graph data to the user terminal for the user terminal to decompress and reconstruct according to the compressed graph data to obtain a reconstructed graph, and displaying the reconstructed graph.

[0018] Through the technical solution of the present application, by adopting this method of generating graphs based on multi-round dialogues, the server can gradually clarify the user's query intention through multi-round dialogues, and finally generate the data required by the user, reducing unnecessary data processing and transmission, and improving the overall processing efficiency. At the same time, in the process of multi-round dialogues, users can send multi-round dialogue messages to gradually refine their queries according to their needs. The server quickly responds and provides accurate graph data based on the dialogue messages sent by the user, thereby improving the user experience. In addition, before transmitting the graph to the user terminal, the compressed graph data is obtained through compression processing and then transmitted, which reduces the load of data transmission, saves network bandwidth and storage resources, and also speeds up the data transmission speed. And through the decompression process of the user terminal, it is ensured that the user can obtain clear and accurate graph information. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application.

[0020] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the description of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0021] Figure 1It is a flowchart of a method for generating a graph based on multi-round dialogues applied to a server provided in an embodiment of the present application; Figure 2 It is a flowchart of a method for generating a graph based on multi-round conversations applied to a user terminal provided in an embodiment of the present application; Figure 3 It is a flowchart of another method for generating a graph based on multi-round dialogues provided in an embodiment of the present application; Figure 4 It is a schematic diagram of a framework of a graph generation device based on multi-round dialogues applied to a server provided in an embodiment of the present application; Figure 5 It is a schematic diagram of a framework of a graph generation method based on multi-round dialogues applied to a user terminal provided in an embodiment of the present application; Figure 6 It is a schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0022] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other.

[0023] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0024] The technical purpose of this application is mainly to solve the problem of processing rendered graph data through compression algorithm in multiple rounds of conversations, solve performance drawbacks, reduce the pressure on the browser, and realize efficient compression and display of graphs containing complex data relationships in the form of images, so as to improve the operating efficiency of the service, improve the user experience and interaction efficiency, and optimize the use and display of graph data.

[0025] Figure 1 is a flow chart of a method for generating a graph based on multi-round dialogues applied to a server provided in an embodiment of the present application, such as Figure 1 As shown, an embodiment of the present application provides a graph generation method based on multi-round dialogues, which is applied to a server, and the method includes: Step S101, receiving an Nth round of dialogue message input by a user terminal for the Nth round of dialogue; Step S102, querying a database and performing data preprocessing according to the Nth round of dialogue message to obtain graph data to be processed corresponding to the Nth round of dialogue; Step S103, creating a graph to be processed according to the graph data to be processed; Step S104, compressing and storing the to-be-processed graph to obtain compressed graph data, wherein the stored compressed graph data is used to respond to multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; Step S105, sending the compressed graph data to the user terminal, so that the user terminal decompresses and reconstructs the compressed graph data to obtain a reconstructed graph, and displays the reconstructed graph.

[0026] Specifically, the embodiment is described in detail below: First, in step S101, during the interaction between the server and the user terminal, multiple rounds of dialogue will be conducted. Taking the Nth round of dialogue as an example (N≥1), the server first needs to receive input from the user through the user terminal, that is, the Nth round of dialogue message input by the user to the user terminal in the Nth round of dialogue. The user expresses his query needs or intentions by inputting the Nth round of dialogue message. For example, the Nth round of dialogue message input by the user is "I want to know who are the people who are directly and indirectly connected to me in my social network."

[0027] Further, in step S102, after receiving the Nth round of conversation message from the user, the server will query the to-be-processed graph data related to the Nth round of conversation message from the database according to the content of the received Nth round of conversation message, wherein the to-be-processed graph data is the graph data that has been found but has not yet been compressed.

[0028] The query process may also involve data retrieval and preprocessing, with the goal of converting the queried data in the database into a format suitable for subsequent processing. For example, if the user is asking about social network information, the server can retrieve the user's friend list, friends of friends, and other information in the database, and clean and format it to generate graph data to be processed.

[0029] Further, in step S103, the server creates a graph to be processed based on the queried graph data to be processed.

[0030] Further, in step S104, in order to reduce the load of the graph to be processed and improve efficiency, the server will compress the graph to be processed to obtain compressed graph data. And store the compressed graph data in the database. The compressed graph data stored in the database is to respond to more specific or in-depth query requests or questions that the user may raise in other conversations subsequent to the Nth round of conversation. For example, if the user asks about a specific social circle or relationship chain in the subsequent N+1th round of conversation, the server can directly use the compressed graph data stored in the database to achieve fast query.

[0031] During the compression process, PNG image compression algorithms or other compression techniques can be used to ensure that the data in the to-be-processed graph is transmitted to the user terminal quickly while saving network bandwidth and storage resources used during the transmission process. For example, the server may compress the image file of the to-be-processed graph to reduce the file size while maintaining sufficient quality.

[0032] It should be noted that the compressed graph data is stored according to the conversation messages and the corresponding conversation rounds. The compressed graph data of each round of multi-round conversations of each user are stored in the same area to achieve fast search.

[0033] Finally, in step S105, the server sends the compressed graph data to the user terminal, so that after receiving the data, the user terminal can restore the graph before compression by decompressing and reconstructing the graph, so that the user can view and analyze it. For example, the user terminal can display an interactive social network graph, and the user can click to view different contacts and relationships.

[0034] Through the above-mentioned embodiments, by adopting this graph generation method based on multi-round dialogues, the server can gradually clarify the user's query intention through multi-round dialogues, and finally generate the data required by the user, reducing unnecessary data processing and transmission, and improving the overall processing efficiency. At the same time, in the process of multi-round dialogues, users can send multi-round dialogue messages to gradually refine their queries according to their needs. The server quickly responds and provides accurate graph data based on the dialogue messages sent by the user, thereby improving the user experience. In addition, before transmitting the graph to the user terminal, the compressed graph data is obtained through compression processing and then transmitted, which reduces the load of data transmission, saves network bandwidth and storage resources, and also speeds up the data transmission speed. And through the decompression process of the user terminal, it is ensured that the user can obtain clear and accurate graph information.

[0035] Optionally, the method further comprises: receiving an N+1th round of dialogue message input by the user terminal for the N+1th round of dialogue; Acquire from the database a plurality of compressed graph data corresponding to the first N rounds of dialogue before the N+1th round of dialogue; According to the N+1th round of dialogue message, query the database to obtain the collected data corresponding to the N+1th round of dialogue message, wherein the collected data corresponding to the N+1th round of dialogue message is obtained by querying the message content in the N+1th round of dialogue message that is different from the first N rounds of dialogue messages corresponding to the first N rounds of dialogue; The to-be-processed graph data corresponding to the N+1th round of dialogues are obtained according to the plurality of compressed graph data corresponding to the first N rounds of dialogues and the collected data corresponding to the messages of the N+1th round of dialogues.

[0036] Specifically, in one embodiment, a processing process for the N+1th round of dialogue after the Nth round of dialogue is added. In this process, the server can generate new graph data to be processed based on the user's input in the N+1th round of dialogue, combined with previous dialogue history and data, to further clarify the user's query intention.

[0037] In this process, first of all, during the multi-round dialogue, the server needs to continuously receive input from the user terminal. In the N+1 round of dialogue, the user will provide more intention information or ask new questions as the input of the N+1 round of dialogue, that is, the N+1 round of dialogue message, based on the data content in the reconstructed graph. The N+1 round of dialogue message will serve as the basis for further processing by the server. For example, the user enters in the N+1 round of dialogue: "I also want to see the contact network of these users in a specific industry."

[0038] Furthermore, the server retrieves and obtains the compressed graph data corresponding to all previous rounds of conversations (i.e., the 1st round of conversations to the Nth round of conversations) from the database. These compressed graph data are the generated results of all previous rounds of conversations and have been compressed and stored so that the server can quickly retrieve and use them in the database. For example, the server can retrieve the compressed graph data corresponding to all social networks previously queried by the user in the database.

[0039] Furthermore, the server needs to query the database to obtain new collected data based on the N+1 round of dialogue messages input by the user in the N+1 round of dialogue. These data are for new questions or query requirements raised by the user in the N+1 round of dialogue messages, which are different from the previous dialogue content. However, this part of the different content cannot be queried in the compressed graph data corresponding to all rounds of historical dialogues. Therefore, it is necessary to query the database again.

[0040] For example, in the N+1 round of conversation, the user asks about the contact network of user A in a specific industry, and the server queries other users and contact information related to user A.

[0041] Furthermore, the server combines the compressed graph data of the first N rounds of conversation with the newly acquired collected data in the N+1 round of conversation to generate new graph data to be processed corresponding to the N+1 round of conversation. In this process, the compressed graph data and collected data of the first N rounds of conversation are integrated and updated to reflect the user's latest query needs. For example, the server may combine the previously retrieved social network graph with the newly retrieved other users and contact information related to user A to generate a new graph.

[0042] As an example, suppose the user queries the social network contact graph in the Nth round of conversation, and the server generates and sends the compressed graph data. In the N+1th round of conversation, the user further asks: "I am particularly interested in contacts related to the financial industry." After receiving this new requirement, the server retrieves the graph data of the previous conversation from the database and queries the new data related to the financial industry. The server combines these data to generate a new graph data to be processed, which contains the social network information previously queried by the user and the newly queried financial industry contact information.

[0043] Optionally, the server can also modify and re-store the compressed graph data that has been generated based on feedback messages from previous and subsequent rounds of conversations with the user. For example, when the user expresses dissatisfaction with the content of the reconstructed graph of the Nth round of conversation in the N+1th round of conversation or believes that the query intention or problem is not resolved, the server can modify, partially delete, update or add the compressed graph data of the Nth round of conversation stored in the database based on the feedback content to improve the user experience during multiple rounds of conversations.

[0044] Through the above embodiments, the server can dynamically adjust and update the graph data according to the new input of the user in the subsequent conversation, thereby improving the interactivity and flexibility of the system. The user can gradually refine and adjust the query requirements, and the server can quickly respond to these changes, thereby improving the user experience. By reusing the compressed graph data of the previous conversation, repeated data queries and processing are reduced, and computing resources and storage space are saved. Moreover, during multiple rounds of conversations, the server can ensure that the data obtained by the user in multiple rounds of conversations is consistent and coherent, avoiding data duplication and conflict, and enhancing the practicality and reliability of the system.

[0045] Optionally, compressing and storing the to-be-processed graph to obtain compressed graph data includes: Removing redundant information between spectra in the spectrum to be processed; By means of inter-band compression, the data dimension and redundancy of the to-be-processed graph are reduced to obtain a preliminarily compressed to-be-processed graph; The preliminarily compressed to-be-processed graph is in-band compressed to reduce the amount of data in the preliminarily compressed to-be-processed graph to obtain the compressed graph data; During the inter-band compression process and the intra-band compression process, sparse coding is performed, and dictionary elements are used to obtain data representation of the graph to be processed.

[0046] Specifically, in one embodiment, the server performs specific steps of compressing the graph to be processed, including removing redundant information, inter-band compression, intra-band compression, and application of sparse coding and dictionary elements.

[0047] First, during the graph compression process, the server needs to identify and remove redundant information between different spectra in the graph. This step is to reduce duplicate content in the data and improve compression efficiency. For example, in a social network graph, if the contact patterns between multiple user groups are similar, the server can identify this pattern and merge it to reduce the amount of data that needs to be stored and transmitted.

[0048] Furthermore, the server reduces the data dimension of the graph through inter-band compression techniques, such as principal component analysis (PCA) transformation. This step aims to remove redundant information between different data bands, reduce the amount of data, and retain key information. For example, if the graph contains multiple related attributes or metrics, the server can use PCA to compress these attributes into several main components that can capture most of the variation information.

[0049] Furthermore, after the inter-band compression, the server further performs in-band compression on the initially compressed graph. This step aims to reduce the amount of information within a single data band and further compress the data. For example, the server can use the PNG image compression algorithm to compress the attributes of each node and edge of the graph to reduce the size of the image file.

[0050] In addition, during the inter-band compression and intra-band compression process, the server performs sparse coding and uses dictionary elements to represent the graph data. Sparse coding improves compression efficiency by representing the signal as a linear combination of a small number of basis functions. Dictionary elements use a set of basis functions to better represent and compress data. For example, the server can learn a dictionary that contains basis functions of multiple graph structures, and then use this dictionary to sparsely represent the graph data, thereby further reducing the amount of data and improving compression efficiency.

[0051] Through the above-mentioned embodiments, the compression efficiency of the graph data is significantly improved by removing redundant information and applying inter-band compression and intra-band compression technology. The inter-band compression technology effectively reduces the dimension of the graph data and reduces the amount of data while retaining key information. The intra-band compression technology further reduces the amount of data, making the graph data more suitable for storage and transmission. The application of sparse coding and dictionary elements improves the quality of data representation, so that the compressed data can more accurately reflect the structure and characteristics of the original graph. The compressed graph data obtained after compression reduces the storage and transmission costs, making the processing and analysis of large-scale graph data more economical and efficient.

[0052] Optionally, querying a database and performing data preprocessing according to the N-th round of dialogue message to obtain the to-be-processed graph data corresponding to the N-th round of dialogue includes: Parsing and verifying the Nth round of dialogue message; According to the N-round dialogue message after the analysis and verification, the original graph data is queried in the database, wherein the original graph data at least includes a plurality of compressed graph data corresponding to the first N-1 rounds of dialogue and a collection message corresponding to the N-round dialogue message, and the collection data corresponding to the N-round dialogue message is obtained by querying the message content in the N-round dialogue message that is different from the first N-1 round dialogue message corresponding to the first N-1 round dialogue; Performing data preprocessing on the original graph data, wherein the data preprocessing includes data cleaning and removing noise, outliers and redundant information; generating preprocessed original graph data by defining entities, attributes and relationships according to the preprocessed original graph data; The preprocessed original graph data is simplified to obtain the graph data to be processed corresponding to the Nth round of dialogue.

[0053] Specifically, in one embodiment, the server can query the database and perform data preprocessing according to the Nth round of dialogue messages to obtain the to-be-processed graph data corresponding to the Nth round of dialogue.

[0054] First, the server needs to parse the Nth round of dialogue message entered by the user in the Nth round of dialogue to understand the user's query intent and verify the validity of the message. This step ensures that the Nth round of dialogue message entered by the user can be correctly understood and processed. For example, the server may check whether the query of the Nth round of dialogue message entered by the user contains the necessary information and whether it conforms to the expected format.

[0055] Furthermore, the server determines the user's query intention based on the Nth round of dialogue messages, and retrieves the original graph data related to the Nth round of dialogue messages from the database. The original graph data includes the compressed graph data corresponding to the previous N rounds of dialogue, and the newly collected data corresponding to the current Nth round of dialogue messages. For example, if the user asks for information on a specific topic in the Nth round of dialogue, the server will retrieve the graph data related to the topic.

[0056] Furthermore, the server preprocesses the retrieved raw graph data, including data cleaning and removing noise, outliers, and redundant information. This step is to ensure the quality of the raw graph data and lay a solid foundation for subsequent graph construction and analysis. For example, the server may remove duplicate nodes or edges, and correct erroneous data.

[0057] Furthermore, based on data preprocessing, the server further defines the entities, attributes, and relationships in the original graph data and converts the original graph data into structured graph data. For example, the server may identify users, products, transactions, etc. as entities and define relationships between them such as purchase and attention.

[0058] Finally, the server simplifies the preprocessed original graph data to generate the to-be-processed graph data corresponding to the Nth round of conversation. The simplification process may include operations such as merging similar nodes and deleting unimportant edges to reduce the complexity of the graph. For example, the server may merge all user nodes related to the "financial industry" into one node to simplify the data content in the graph.

[0059] As an example, suppose the user asks in the Nth round of conversation: "I want to know the cooperation between different departments in our company." The server first parses and verifies this query, and then retrieves the graph data related to the cooperation between the company's departments from the database. The server cleans and preprocesses this data, removes errors and redundant information, and defines entities and relationships such as departments, employees, and cooperative projects. Finally, the server simplifies the graph data, such as merging smaller department nodes, to generate a processed graph data that clearly shows the cooperation within the company.

[0060] Through the above embodiments, the Nth round of dialogue messages are parsed and verified to ensure the accuracy and effectiveness of data processing. The data preprocessing step removes noise and outliers, improves the quality of the data, and provides a reliable data basis for subsequent analysis. In addition, by defining entities, attributes, and relationships, and simplifying the processing, the readability and usability of the graph are enhanced, making it easier for users to understand and analyze graph data, and users can see the information they request more intuitively, which improves the user experience. The simplified graph data reduces the complexity of calculation and rendering, and improves the system response speed and efficiency.

[0061] Figure 2 is a flow chart of a method for generating a graph based on multi-round dialogues applied to a user terminal provided by an embodiment of the present application, such as Figure 2 As shown, based on the same inventive concept, another embodiment of the present application also provides a graph generation method based on multi-round dialogues, which is applied to a user terminal, and the method includes: Step S201, receiving compressed graph data for the Nth round of dialogue message sent by the server, wherein the compressed graph data is obtained by the server querying a graph database according to the Nth round of dialogue message input by the user terminal for the Nth round of dialogue, obtaining the graph data to be processed corresponding to the Nth round of dialogue, and creating and storing a graph to be processed according to the graph data to be processed, wherein the stored graph to be processed is used to respond to the multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; and then compressing the graph to be processed; Step S202, processing the compressed graph data according to the decompression algorithm and parameters corresponding to the compression by the server to obtain a reconstructed graph corresponding to the Nth round of dialogue messages; Step S203, presenting the reconstructed atlas to the user in a graphical manner.

[0062] First, in step S201, the user terminal receives compressed graph data from the server in this step. These data are obtained by the server querying the graph database according to the message input by the user terminal in the Nth round of dialogue, and then processed and compressed. This step is a key link in the interaction between the user terminal and the server, ensuring that the user can receive the required graph data. For example, if the user asks for graph information about a certain topic, the server will process this request and send the corresponding compressed graph data.

[0063] Further, in step S202, the user terminal uses the decompression algorithm and parameters corresponding to the compression performed by the server to obtain a corresponding decoding algorithm, and processes the received compressed image data through the decoding algorithm. This ensures that the user terminal can correctly decompress and reconstruct the reconstructed graph. For example, if the server uses a specific PNG compression algorithm, the user terminal needs to use the corresponding decompression algorithm to restore the graph.

[0064] Finally, in step S203, after decompression and reconstruction to obtain the reconstructed graph, the user terminal displays the reconstructed graph to the user in a graphical manner. This step provides an intuitive graphical interface, allowing users to more easily understand and analyze graph data. For example, the user terminal may use a graphics library or visualization tool to display nodes, edges and attributes, and provide interactive functions such as zooming in, zooming out, searching for nodes, etc.

[0065] For example, suppose the user asks about the internal project cooperation network of the company in the Nth round of dialogue. After processing this request, the server compresses the relevant graph data to obtain compressed graph data and sends it to the user terminal. After receiving the compressed graph data, the user terminal uses the corresponding decompression algorithm to process the compressed graph data to obtain the reconstructed graph and displays it to the user in a graphical way on the screen. The user can see the connection between different projects and departments, as well as the intensity of cooperation between them.

[0066] Through the above-mentioned embodiments, compressed graph data is used for transmission, which reduces the load of data transmission and improves the efficiency of data transmission. In addition, the compressed graph data is more difficult to be intercepted and interpreted during transmission, which enhances the security of the data. The process of decompressing and reconstructing the graph is fast and efficient, which improves the response speed of the system. By displaying in a visual way, users can intuitively view and analyze the graph data, which improves the user experience and enhances the practicality and reliability of the system.

[0067] Optionally, the method further comprises: Receive the compressed graph data corresponding to the N+1th round of dialogue message sent by the server; Based on the compressed graph data corresponding to the N+1th round of dialogue message, a reconstructed graph corresponding to the N+1th round of dialogue message is obtained; The reconstructed graph corresponding to the N+1-th round of dialogue message is displayed to the user in a graphical manner, wherein the graph contents that are different from the reconstructed graph corresponding to the N+1-th round of dialogue message and the reconstructed graph corresponding to the N-th round of dialogue message are displayed in a manner that is different from the graph contents of the reconstructed graph corresponding to the N-th round of dialogue message.

[0068] Specifically, in one embodiment, the user terminal can add processing for subsequent conversations (N+1th round) based on the above embodiment, including how the user terminal receives, processes and sends compressed graph data for the N+1th round of conversation messages sent by the server, and displays it to the user in a graphical manner.

[0069] First, the user terminal receives the compressed graph data for the N+1 round of conversation from the server. These data are obtained by the server querying the graph database based on the message input by the user terminal in the N+1 round of conversation, and then processed and compressed, ensuring that the user can receive the latest graph data for further analysis and interaction.

[0070] Furthermore, the user terminal uses the decompression algorithm and parameters corresponding to the compression by the server to process the received compressed graph data to reconstruct the reconstructed graph corresponding to the N+1th round of dialogue, ensuring that the user terminal can correctly decompress the data and reconstruct the original graph so that the user can view the latest information.

[0071] Furthermore, the user terminal displays the reconstructed graph data to the user in a graphical manner, so that the user can more easily understand and analyze the graph data. In this process, for the graph content that is different between the reconstructed graph corresponding to the N+1th round of dialogue message and the reconstructed graph corresponding to the Nth round of dialogue message, the user terminal can display this part of the different graph content in a manner that is different from the reconstructed graph corresponding to the Nth round of dialogue message, so that the user can clearly see the new changes and information.

[0072] For example, suppose the user asks about the project cooperation network within the company in the Nth round of dialogue, and then further asks about the detailed information of a specific department in the N+1th round of dialogue. After processing this request, the server compresses the relevant graph data and sends it to the user terminal. After receiving these compressed graph data, the user terminal uses the corresponding decompression algorithm to process these data to obtain the reconstructed graph and displays it to the user in a graphical way on the screen. The user can see that compared with the Nth round of dialogue, the detailed information of the specific department in the graph of the N+1th round of dialogue is highlighted, such as the size, color or style of the node is different to highlight the new changes and information.

[0073] Through the above embodiments, users can instantly receive and view the latest graph data, which improves the real-time nature of data updates. Moreover, by adopting different display methods, users can intuitively see new changes and information. By distinguishing and displaying graph contents of different rounds of conversations, the comparability of data is enhanced, making it easier for users to identify changes, thereby improving user experience and enhancing the practicality and reliability of the system.

[0074] Figure 3 FIG. 1 is a flow chart of another method for generating a graph based on a multi-round dialogue provided by an embodiment of the present application. The interaction process between the user, the user terminal, the server and the database in the multi-round dialogue is described in detail below. Figure 3 shown.

[0075] The workflow of the entire solution can be described as follows: First, the user initiates the Nth round of dialogue and enters a query request in the user terminal. The user terminal extracts the content of the Nth round of dialogue, obtains the Nth round of dialogue message, and sends the Nth round of dialogue message to the server. After receiving the Nth round of dialogue message, the server parses and verifies it, and stores the parsed and verified Nth round of dialogue message in the database, and obtains the storage result returned by the database.

[0076] Furthermore, the server requests to query the database according to the Nth round of dialogue messages, and retrieves the to-be-processed graph data corresponding to the Nth round of dialogue. The server preprocesses these data, including data cleaning, removing noise and redundant information, and defining entities, attributes and relationships, to generate preprocessed original graph data, which includes compressed graph data corresponding to the previous N rounds of dialogue, and newly collected data corresponding to the current Nth round of dialogue messages.

[0077] Next, the server simplifies the preprocessed original graph data and constructs a graph to obtain the graph data to be processed corresponding to the Nth round of dialogue. It performs compression processing on the graph data to be processed, including removing redundant information between spectra, inter-band compression to reduce data dimensions, intra-band compression to reduce data volume, and the application of sparse coding and dictionary elements, and finally generates compressed graph data.

[0078] Then, the compressed graph data on the server side is sent to the user terminal. The user terminal decompresses the compressed graph data according to the decompression algorithm and parameters corresponding to the compression on the server side, obtains the reconstructed graph, and displays it to the user in a graphical way.

[0079] When there are subsequent rounds of multiple rounds of dialogue, such as the N+1th round of dialogue, the user can continue to input new N+1th round dialogue messages. After receiving these N+1th round dialogue messages, the server will generate new graph data to be processed based on the new request and the previously stored graph data, and repeat the above compression and sending process. After the user terminal receives the new compressed graph data, it decompresses and reconstructs the graph. In particular, for the graph content that is different from the reconstructed graph corresponding to the N+1th round of dialogue messages and the reconstructed graph corresponding to the Nth round of dialogue messages, the user terminal will display it in a way that is different from the reconstructed graph corresponding to the Nth round of dialogue messages, so that the user can clearly identify new changes and information. In this way, the entire solution supports users to gradually and deeply query and analyze graph data through multiple rounds of dialogue, while ensuring efficient transmission and intuitive display of data.

[0080] Figure 4 is a schematic diagram of a framework of a graph generation device based on multi-round dialogues applied to a server provided in an embodiment of the present application, such as Figure 4As shown, based on the same inventive concept, an embodiment of the present application provides a graph generation device based on multi-round dialogues, which is applied to a server, and the device includes: A conversation message receiving module 11, configured to receive an Nth round of conversation message input by a user terminal for the Nth round of conversation; The to-be-processed graph data generating module 12 is used to query the database and perform data preprocessing according to the N-th round of dialogue messages to obtain the to-be-processed graph data corresponding to the N-th round of dialogue; A to-be-processed graph creation module 13, used to create a to-be-processed graph according to the to-be-processed graph data; A compression module 14, used for compressing and storing the to-be-processed graph to obtain compressed graph data, wherein the stored compressed graph data is used to respond to multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; The sending module 15 is used to send the compressed graph data to the user terminal, so that the user terminal can decompress and reconstruct the compressed graph data to obtain a reconstructed graph, and display the reconstructed graph.

[0081] Optionally, the device further comprises: The N+1th round of dialogue message receiving module is used to receive the N+1th round of dialogue message input by the user terminal for the N+1th round of dialogue; A compressed graph data acquisition module, used to acquire from the database a plurality of compressed graph data corresponding to the first N rounds of dialogue before the N+1th round of dialogue; a collection data acquisition module, configured to query the database according to the N+1th round of dialogue message to obtain the collection data corresponding to the N+1th round of dialogue message, wherein the collection data corresponding to the N+1th round of dialogue message is obtained by querying the content of the N+1th round of dialogue message that is different from the first N rounds of dialogue messages corresponding to the first N rounds of dialogue; The module for acquiring graph data to be processed is used to obtain the graph data to be processed corresponding to the N+1th round of dialogue according to the multiple compressed graph data corresponding to the first N rounds of dialogue and the collected data corresponding to the N+1th round of dialogue messages.

[0082] Optionally, the compression module 14 includes: A removal unit, used to remove redundant information between spectra in the spectrum to be processed; An inter-band compression unit, used to reduce the data dimension and redundancy of the to-be-processed graph through inter-band compression, so as to obtain a preliminarily compressed to-be-processed graph; An in-band compression unit, used for in-band compression of the preliminarily compressed to-be-processed graph to reduce the amount of data in the preliminarily compressed to-be-processed graph to obtain the compressed graph data; The sparse coding and dictionary representation unit is used to perform sparse coding during the inter-band compression process and the intra-band compression process, and to obtain the data representation of the to-be-processed graph using dictionary elements.

[0083] Optionally, the to-be-processed graph data generating module 12 includes: A parsing and verification unit, used for parsing and verifying the Nth round of dialogue message; an integration unit, configured to query the database to obtain original graph data according to the parsed and verified N-round dialogue message, wherein the original graph data at least includes a plurality of compressed graph data corresponding to the first N-1 rounds of dialogue and a collection message corresponding to the N-round dialogue message, wherein the collection data corresponding to the N-round dialogue message is obtained by querying the message content in the N-round dialogue message that is different from the first N-1 round dialogue message corresponding to the first N-1 round dialogue; A preprocessing unit, configured to perform data preprocessing on the original graph data, wherein the data preprocessing includes data cleaning, and removing noise, outliers, and redundant information; and generating preprocessed original graph data by defining entities, attributes, and relationships based on the preprocessed original graph data; The simplification unit is used to simplify the pre-processed original graph data to obtain the graph data to be processed corresponding to the Nth round of dialogue.

[0084] Figure 5 is a schematic diagram of a framework of a graph generation method based on multi-round dialogues applied to a user terminal provided in an embodiment of the present application, such as Figure 5 As shown, based on the same inventive concept, an embodiment of the present application further provides a graph generation device based on multi-round dialogues, which is applied to a user terminal, and the device includes: The compressed graph data receiving module 21 is used to receive the compressed graph data for the Nth round of dialogue message sent by the server, wherein the compressed graph data is the Nth round of dialogue message input by the user terminal for the Nth round of dialogue received by the server; according to the Nth round of dialogue message, query the database and perform data preprocessing to obtain the graph data to be processed corresponding to the Nth round of dialogue; according to the graph data to be processed, create a graph to be processed; compress and store the graph to be processed, and the stored compressed graph data is used to respond to the multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; A reconstruction module 22, used to process the compressed graph data according to the decompression algorithm and parameters corresponding to the compression by the server, to obtain a reconstructed graph corresponding to the Nth round of dialogue messages; The display module 23 is used to display the reconstructed atlas to the user in a graphical manner.

[0085] Optionally, the device further comprises: A compressed graph data receiving module is used to receive the compressed graph data corresponding to the N+1th round of dialogue message sent by the server; A reconstructed graph acquisition module, used to obtain a reconstructed graph corresponding to the N+1th round of dialogue message based on the compressed graph data corresponding to the N+1th round of dialogue message; The distinguishing display module is used to display the reconstructed graph corresponding to the N+1th round of dialogue message to the user in a graphical manner, wherein the graph content that is distinguished between the reconstructed graph corresponding to the N+1th round of dialogue message and the reconstructed graph corresponding to the Nth round of dialogue message is displayed in a manner that is distinguished from the graph content of the reconstructed graph corresponding to the Nth round of dialogue message.

[0086] Based on the same inventive concept, another embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the graph generation method based on multi-round dialogue as described in any of the above embodiments.

[0087] Among them, electronic equipment refers to Figure 6 , Figure 6 is a schematic diagram of an electronic device provided by an embodiment of the present application. Figure 6 As shown, the electronic device 600 includes: a memory 610 and a processor 620. The memory 610 and the processor 620 are connected via a bus communication. A computer program is stored in the memory 610. The computer program can be executed on the processor 620 to implement the steps in the graph generation method based on multi-round dialogue disclosed in the above embodiment of the present application.

[0088] Based on the same inventive concept, another embodiment of the present application further provides a computer program product, including a computer program, which is executed by a processor to perform the graph generation method based on multi-round dialogues as described in any of the above embodiments.

[0089] Based on the same inventive concept, another embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored, wherein when the program is executed by a processor, the graph generation method based on multi-round dialogue as described in any of the above embodiments is implemented.

[0090] As for the device, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0091] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0092] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0093] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0094] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0095] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0096] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0097] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0098] The above is a detailed introduction to the graph generation method, device, equipment and medium based on multi-round dialogue provided by the present application. This article uses specific examples to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A graph generation method based on multi-round dialogue, characterized in that: Applied to the server, the method includes: Receiving an N-th round of dialogue message input by a user terminal for the N-th round of dialogue; According to the Nth round of dialogue message, query the database and perform data preprocessing to obtain the to-be-processed graph data corresponding to the Nth round of dialogue; Creating a graph to be processed according to the graph data to be processed; Compressing and storing the graph to be processed to obtain compressed graph data, wherein the stored compressed graph data is used to respond to multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; The compressed graph data is sent to the user terminal, so that the user terminal decompresses and reconstructs the compressed graph data to obtain a reconstructed graph, and displays the reconstructed graph.

2. The graph generation method based on multi-round dialogue according to claim 1 is characterized in that: The method further comprises: receiving an N+1th round of dialogue message input by the user terminal for the N+1th round of dialogue; Acquire from the database a plurality of compressed graph data corresponding to the first N rounds of dialogue before the N+1th round of dialogue; According to the N+1th round of dialogue message, query the database to obtain the collected data corresponding to the N+1th round of dialogue message, wherein the collected data corresponding to the N+1th round of dialogue message is obtained by querying the message content in the N+1th round of dialogue message that is different from the first N rounds of dialogue messages corresponding to the first N rounds of dialogue; The to-be-processed graph data corresponding to the N+1th round of dialogues are obtained according to the plurality of compressed graph data corresponding to the first N rounds of dialogues and the collected data corresponding to the messages of the N+1th round of dialogues.

3. The graph generation method based on multi-round dialogue according to claim 1 is characterized in that: The step of compressing and storing the to-be-processed graph to obtain compressed graph data includes: Removing redundant information between spectra in the spectrum to be processed; By means of inter-band compression, the data dimension and redundancy of the to-be-processed graph are reduced to obtain a preliminarily compressed to-be-processed graph; The preliminarily compressed to-be-processed graph is in-band compressed to reduce the amount of data in the preliminarily compressed to-be-processed graph to obtain the compressed graph data; During the inter-band compression process and the intra-band compression process, sparse coding is performed, and dictionary elements are used to obtain data representation of the graph to be processed.

4. The graph generation method based on multi-round dialogue according to claim 1 is characterized in that: The querying of a database and performing data preprocessing according to the N-th round of dialogue message to obtain the to-be-processed graph data corresponding to the N-th round of dialogue includes: Parsing and verifying the Nth round of dialogue message; According to the N-round dialogue message after the analysis and verification, the original graph data is queried in the database, wherein the original graph data at least includes a plurality of compressed graph data corresponding to the first N-1 rounds of dialogue and a collection message corresponding to the N-round dialogue message, and the collection data corresponding to the N-round dialogue message is obtained by querying the message content in the N-round dialogue message that is different from the first N-1 round dialogue message corresponding to the first N-1 round dialogue; Performing data preprocessing on the original graph data, wherein the data preprocessing includes data cleaning and removing noise, outliers and redundant information; generating preprocessed original graph data by defining entities, attributes and relationships according to the preprocessed original graph data; The preprocessed original graph data is simplified to obtain the graph data to be processed corresponding to the Nth round of dialogue.

5. A graph generation method based on multi-round dialogue, characterized in that: Applied to a user terminal, the method comprises: Receive compressed graph data for the Nth round of dialogue message sent by the server, wherein the compressed graph data is the Nth round of dialogue message input by the user terminal for the Nth round of dialogue received by the server; query the database and perform data preprocessing according to the Nth round of dialogue message to obtain the graph data to be processed corresponding to the Nth round of dialogue; create a graph to be processed according to the graph data to be processed; compress and store the graph to be processed, wherein the stored compressed graph data is used to respond to multiple rounds of dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; Processing the compressed graph data according to the decompression algorithm and parameters corresponding to the compression by the server to obtain a reconstructed graph corresponding to the Nth round of dialogue messages; The reconstructed reconstruction map is displayed to the user in a graphical manner.

6. The graph generation method based on multi-round dialogue according to claim 5 is characterized in that: The method further comprises: Receive the compressed graph data corresponding to the N+1th round of dialogue message sent by the server; Based on the compressed graph data corresponding to the N+1th round of dialogue message, a reconstructed graph corresponding to the N+1th round of dialogue message is obtained; The reconstructed graph corresponding to the N+1-th round of dialogue message is displayed to the user in a graphical manner, wherein the graph contents that are different from the reconstructed graph corresponding to the N+1-th round of dialogue message and the reconstructed graph corresponding to the N-th round of dialogue message are displayed in a manner that is different from the graph contents of the reconstructed graph corresponding to the N-th round of dialogue message.

7. A graph generation device based on multi-round dialogue, characterized in that: Applied to the server, the device comprises: A dialogue message receiving module, used to receive the Nth round of dialogue message input by the user terminal for the Nth round of dialogue; A to-be-processed graph data generating module is used to query a database and perform data preprocessing according to the N-th round of dialogue messages to obtain the to-be-processed graph data corresponding to the N-th round of dialogue; A to-be-processed graph creation module, used to create a to-be-processed graph based on the to-be-processed graph data; A compression module, used for compressing and storing the to-be-processed graph to obtain compressed graph data, wherein the stored compressed graph data is used to respond to multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; A sending module is used to send the compressed graph data to the user terminal, so that the user terminal can decompress and reconstruct the compressed graph data to obtain a reconstructed graph, and display the reconstructed graph.

8. A graph generation device based on multi-round dialogue, characterized in that: Applied to a user terminal, the device comprises: A compressed graph data receiving module is used to receive compressed graph data for the Nth round of dialogue message sent by the server, wherein the compressed graph data is the Nth round of dialogue message input by the user terminal for the Nth round of dialogue received by the server; according to the Nth round of dialogue message, query the database and perform data preprocessing to obtain the graph data to be processed corresponding to the Nth round of dialogue; according to the graph data to be processed, create a graph to be processed; compress and store the graph to be processed, and the stored compressed graph data is used to respond to the multi-round dialogue messages input by the user terminal in multiple dialogues after the Nth round of dialogue; A reconstruction module, used to process the compressed graph data according to the decompression algorithm and parameters corresponding to the compression by the server, to obtain a reconstructed graph corresponding to the Nth round of dialogue messages; The display module is used to display the reconstructed reconstruction map to the user in a graphical manner.

9. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the graph generation method based on multi-round dialogue as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, wherein when the computer program is executed by a processor, the method for generating a graph based on multi-round dialogue as described in any one of claims 1-6 is implemented.

Images