Sub - process Configuration Method, Device, Equipment and Medium Based on Dialogue Robot

Abstract

The present invention relates to the field of data processing, and provides a sub-process configuration method, device, equipment and medium based on a dialogue robot. The method includes: when it is detected that a target node in a preset node area moves into the dialogue editing area, obtaining the position coordinates of the target node in the dialogue editing area, and determining the distance and topological relationship between the nodes in the dialogue editing area according to the position coordinates; determining the node attribute information corresponding to each node, and extracting corresponding nodes in the dialogue editing area according to the node attribute information and topological relationship of each node to generate a sub-process; using a reference relationship table to associate and store the sub-processes according to the sub-process names; calling the corresponding sub-processes according to the sub-process names in the reference relationship table, and configuring the sub-processes according to the business process until the sub-process configuration is completed. The present invention not only reduces the processing burden of the dialogue robot, but also improves the process configuration efficiency, and is also helpful for improving the dialogue intelligence level.

Description

Technical Field

[0001] The present invention relates to the field of dialogue robots, and provides a method, apparatus, device, and medium for identifying a sub-process configuration based on a dialogue robot. Background Art

[0002] With the development of artificial intelligence technology, and also benefiting from favorable policies, capital promotion, and demand promotion, artificial intelligence technology has entered an unprecedented period of vigorous development. Among them, intelligent dialogue robots are an important direction in the development of artificial intelligence technology and are now widely used in multiple fields, such as intelligent customer service, smart home, Internet of Things, and so on.

[0003] In intelligent dialogue robot technology, the industry generally adopts the form of a question-and-answer knowledge base. It is necessary to store questions and corresponding answers in the question-and-answer knowledge base in advance. When a user inputs a question, the intelligent dialogue robot retrieves the corresponding answer to the question from the question-and-answer knowledge base and feeds it back to the user to achieve human-machine dialogue. However, this intelligent dialogue robot using the form of a question-and-answer knowledge base can only support one-round dialogue. When a user has multiple rounds of dialogue with the intelligent dialogue robot, the intelligent dialogue robot cannot feed back an answer to the user based on the context information included in the multiple rounds of dialogue, but can only regard each round of dialogue as an independent dialogue separately. In this way, when the questions input by the user in each round of dialogue are not related, the probability of the intelligent dialogue robot giving an accurate answer is relatively high. However, when the questions input by the user in each round of dialogue are related, the intelligent dialogue robot is very likely to fail to recognize the user's true intention and feed back an incorrect or completely irrelevant answer, thereby reducing the user's willingness to use and the communication efficiency.

[0004] In practical applications, especially intelligent dialogue robots are configured through process templates. However, as the complexity of the business continues to increase, the process of each dialogue robot will become more and more complex, and the process template will also become larger and larger. Since most of the process templates are loaded and configured based on a browser, the speed of loading the process template will also become slower and slower. At the same time, and because in most cases, one robot corresponds to one business scenario, and the processes and speech configurations in the business scenario often overlap. Configuring similar functions multiple times in the large robot process template is often time-consuming and laborious, which not only increases the processing burden of the dialogue robot, makes the process of the dialogue robot more complex, is not conducive to the loading and configuration of the process template, but also reduces the loading speed of the process template. Summary of the Invention

[0005] The present invention provides a sub - process configuration method, device, equipment and medium based on a dialogue robot. Its main purpose is to obtain a configuration instruction of the dialogue robot, send a dialogue editing page request to the server according to the configuration instruction of the dialogue robot; receive the dialogue editing page returned by the server and display it. The dialogue editing page includes a preset node area and a dialogue editing area; when it is detected that a target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area, and determine the distance and topological relationship between the nodes in the dialogue editing area according to the position coordinates; determine the node attribute information corresponding to each node, and extract the corresponding nodes according to the node attribute information and topological relationship of each node in the dialogue editing area to generate a sub - process. The sub - process can be repeatedly called, avoiding repeated operations when the dialogue robot references the sub - process, reducing the processing burden of the dialogue robot, making the configuration of the dialogue robot simpler, improving the process configuration efficiency, not only saving the configuration time of the sub - process, but also helping to improve the dialogue intelligence level of the dialogue robot and meeting richer application scenarios.

[0006] To achieve the above object, the present invention provides a sub - process configuration method based on a dialogue robot, and the method includes:

[0007] Obtain a configuration instruction of the dialogue robot, and send a dialogue editing page request to the server according to the configuration instruction of the dialogue robot;

[0008] Receive the dialogue editing page returned by the server and display it. The dialogue editing page includes a preset node area and a dialogue editing area;

[0009] When it is detected that a target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area, determine the distance between the nodes in the dialogue editing area according to the position coordinates, and determine the topological relationship between the nodes by combining the distance between the nodes and the relative position relationship;

[0010] Determine the node attribute information corresponding to each node, and extract the corresponding nodes according to the node attribute information and topological relationship of each node in the dialogue editing area to generate a sub - process. The sub - process at least includes a sub - process name, a sub - process category, an intent category and a sub - process description;

[0011] Use a reference relationship table to associate and store the sub - processes according to the sub - process name;

[0012] Call the corresponding sub - process according to the sub - process name in the reference relationship table, and configure the sub - process according to the business process until the sub - process configuration is completed.

[0013] Optionally, before obtaining the configuration instruction of the dialogue robot and sending a dialogue editing page request to the server according to the configuration instruction of the dialogue robot, it includes:

[0014] Receiving a robot information configuration instruction, where the robot identifier is carried in the information configuration instruction, obtaining the corresponding information configuration page according to the robot identifier and displaying it; receiving an information configuration completion instruction, obtaining basic configuration information according to the information configuration completion instruction, and sending the basic configuration information and the robot identifier to the server.

[0015] Optionally, when it is detected that the target node in the preset node area moves into the dialogue editing area, obtaining the position coordinates of the target node in the dialogue editing area, determining the distance between the nodes in the dialogue editing area according to the position coordinates, and determining the topological relationship between the nodes by combining the distance between the nodes and the relative position relationship, includes:

[0016] When it is detected that the target node in the preset node area moves into the dialogue editing area, obtaining the position coordinates of the target node in the dialogue editing area;

[0017] Performing parsing processing on the connection lines between any two nodes formed by the existing nodes and the target node in the dialogue editing area to obtain the relative position relationship between any two nodes in the current dialogue editing area;

[0018] Calculating the distance between any two nodes in the current dialogue editing area according to the position coordinates of each node, and obtaining the topological relationship between the nodes according to the distance between the nodes and the relative position relationship.

[0019] Optionally, determining the node attribute information corresponding to each of the nodes, and extracting the corresponding node generation sub-process according to the node attribute information and the topological relationship of each node in the dialogue editing area, includes:

[0020] Extracting the node configuration information of each node in the dialogue editing area, determining the node attribute information corresponding to each of the nodes according to the node configuration information of each node, and the node attribute information includes a control node and a conversation node;

[0021] Determining the topological relationship between the nodes from the position relationship formed by the connection lines between any two nodes in the dialogue editing area, and the types of the connection lines include non-intentional lines and intentional lines;

[0022] Dividing the node attribute information and the topological relationship of each node in the dialogue editing area according to the intention of the business process to generate at least one initial sub-process;

[0023] Adjust the node attribute information and node connections in the generated initial sub-process according to the sub-process rules to generate a sub-process that conforms to the sub-process rules. Each of the sub-processes corresponds to at least one version.

[0024] Optionally, before associatively storing the sub-processes according to the sub-process names using the reference relationship table, the following steps are further included:

[0025] Verify the sub-process according to the pre-set verification rules of the sub-process. After the verification is qualified, associatively store the sub-process according to the sub-process name and the reference relationship table; where the pre-set verification rules include at least one of the following:

[0026] When it is detected that the current sub-process is called by a robot, then this sub-process cannot be taken offline; if you want to modify this sub-process, you can only copy this sub-process for modification;

[0027] Determine whether the type of a node before the end node in the current sub-process has changed. If there is a change and the current sub-process is called by a robot, then this sub-process cannot be used.

[0028] Optionally, calling the corresponding sub-process according to the sub-process name in the reference relationship table and configuring the sub-process according to the business process includes:

[0029] Extract the intention list associated with the business process of the current sub-process to be configured. The intention list contains at least one intention;

[0030] Match the intentions in the intention list according to the intention categories corresponding to the sub-process names in the reference relationship table;

[0031] When it is detected that the intention category corresponding to any sub-process name is the same as or similar to the intention, then the matching is successful, and determine the intention association mapping between the sub-process name and the business process;

[0032] Use the intention of the business process to call the associated sub-process, and add the sub-process to the corresponding node in combination with the positional relationship and node attribute information of each node in the current business process for configuration.

[0033] Optionally, matching the intentions in the intention list according to the intention categories corresponding to the sub-process names in the reference relationship table includes:

[0034] Use the text clustering algorithm to cluster the intention categories corresponding to the sub-process names in the reference relationship table; obtain the category center points of the clustering categories;

[0035] Calculate the distance value between the intention in the intention list and the category center point;

[0036] Compare the distance value between the intention in the intention list and the category center point with a preset distance threshold. When the distance value between the intention in the intention list and the category center point is less than the preset distance threshold, it indicates that the intention in the intention list is the same as or similar to the intention category corresponding to the subprocess name in the reference relationship table.

[0037] In addition, to achieve the above object, the present invention further provides a subprocess configuration device based on a dialogue robot, and the device includes:

[0038] An acquisition module, configured to acquire a configuration instruction of the dialogue robot, and send a dialogue editing page request to the server according to the configuration instruction of the dialogue robot;

[0039] A dialogue editing module, configured to receive and display the dialogue editing page returned by the server, where the dialogue editing page includes a preset node area and a dialogue editing area;

[0040] A distance and relationship determination module, configured to, when detecting that a target node in the preset node area moves to the dialogue editing area, acquire the position coordinates of the target node in the dialogue editing area, determine the distance between nodes in the dialogue editing area according to the position coordinates, and determine the topological relationship between nodes in combination with the distance and relative position relationship between the nodes;

[0041] A subprocess generation module, configured to determine the node attribute information corresponding to each of the nodes, and extract corresponding nodes according to the node attribute information and topological relationship of each node in the dialogue editing area to generate a subprocess, where the subprocess at least includes a subprocess name, a subprocess category, an intention category, and a subprocess description;

[0042] An association storage module, configured to associatively store the subprocesses according to the subprocess names by using a reference relationship table;

[0043] A subprocess configuration module, configured to call the corresponding subprocess according to the subprocess name in the reference relationship table, and configure the subprocess according to the business process until the subprocess configuration is completed.

[0044] In addition, to achieve the above object, the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the computer program, the steps of the method described in any one of the above embodiments are implemented.

[0045] In addition, to achieve the above object, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in any one of the above embodiments are implemented.

[0046] The sub - process configuration method, device, equipment and medium based on a dialogue robot proposed by the present invention obtain a configuration instruction of the dialogue robot, send a dialogue editing page request to a server according to the configuration instruction of the dialogue robot; receive the dialogue editing page returned by the server and display it. The dialogue editing page includes a preset node area and a dialogue editing area; when it is detected that a target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area, determine the distances and topological relationships between the nodes in the dialogue editing area according to the position coordinates; determine the node attribute information corresponding to each node, and extract corresponding nodes according to the node attribute information and topological relationships of each node in the dialogue editing area to generate a sub - process, which can repeatedly call the sub - process, avoid repeated operations when the dialogue robot references the sub - process, reduce the processing burden of the dialogue robot, make the configuration of the dialogue robot simpler, improve the process configuration efficiency, not only save the configuration time of the sub - process, but also help to improve the dialogue intelligence level of the dialogue robot and meet richer application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flowchart of a sub - process configuration method based on a dialogue robot provided in an embodiment of the present invention;

[0048] Figure 2 It is a flowchart of distance and relationship determination in the sub - process configuration method based on a dialogue robot provided in an embodiment of the present invention;

[0049] Figure 3 It is a flowchart of sub - process generation in the sub - process configuration method based on a dialogue robot provided in an embodiment of the present invention;

[0050] Figure 4 It is a flowchart of sub - process configuration in the sub - process configuration method based on a dialogue robot provided in an embodiment of the present invention;

[0051] Figure 5 It is a structural diagram of a sub - process configuration device based on a dialogue robot provided in an embodiment of the present invention;

[0052] Figure 6 It is a structural diagram of a computer device provided in an embodiment of the present invention.

[0053] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] It should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0055] To facilitate the understanding of this application, the concepts involved in this application will be explained first.

[0056] In one embodiment, it is a schematic flowchart of a sub - process configuration method based on a dialogue robot provided in an embodiment of the present invention. Refer to Figure 1 As shown, the method includes the following steps:

[0057] Step S101, obtain the configuration instruction of the dialogue robot, and send a dialogue editing page request to the server according to the configuration instruction of the dialogue robot;

[0058] Among them, the terminal communicates with the server through the network. The terminal receives the dialogue robot configuration instruction and sends a dialogue flow editing page request to the server according to the dialogue robot configuration instruction. The terminal can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, and portable wearable devices. The server can be implemented by an independent server or a server cluster composed of multiple servers.

[0059] It should be noted that a dialogue robot is a computer program that conducts conversations via dialogue or text, can simulate human conversations, and pass the Turing test.

[0060] In the above - mentioned manner, according to the configuration instruction of the dialogue robot, a dialogue flow editing page request is generated, which can perform page editing on the dialogue robot, thereby meeting various editing requirements of users and helping to improve the process editing ability of the dialogue robot.

[0061] Step S102, receive the dialogue editing page returned by the server and display it. The dialogue editing page includes a preset node area and a dialogue editing area;

[0062] Among them, a dialogue robot is a computer program that conducts conversations via dialogue or text, can simulate human conversations, and pass the Turing test. Chatbots can be used for practical purposes, such as customer service, information acquisition, and conversation skill training, etc. A preset node is a component used for dialogue configuration, which is pre - set and can be directly used. The preset node can include components such as a narrator dialogue component, a fixed dialogue component, a fixed Q&A component, an intention dialogue component, and a scoring dialogue component, etc. The dialogue editing area refers to the area used for editing dialogues. Basic configuration information components are pre - set in the dialogue editing area. The basic configuration information components are components generated according to the basic configuration information of the dialogue robot. The basic configuration information can be the gender, age, and position of the dialogue robot, etc.

[0063] Specifically, the terminal receives a dialogue robot configuration instruction, and sends a dialogue editing page request to the server according to the dialogue robot configuration instruction. The server receives the dialogue editing page request, obtains the pre-set dialogue editing page according to the dialogue editing page request, and returns the dialogue editing page to the terminal. The terminal receives the dialogue editing page returned by the server and displays it. The dialogue editing page includes a preset node area and a dialogue editing area, and there is a basic configuration information component in the dialogue editing area.

[0064] Step S103, when it is detected that the target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area, determine the distance between the nodes in the dialogue editing area according to the position coordinates, and combine the distance between the nodes and the relative position relationship to determine the topological relationship between the nodes;

[0065] Among them, the target node refers to a component selected from each preset component and is used to configure the dialogue in the dialogue flow. The existing node refers to a component generated according to the robot's basic configuration information that already exists in the dialogue editing area, or other components that have been configured, and is used to determine the position of the component to be configured in the dialogue editing area.

[0066] For example, obtain the position coordinates of the target node in the dialogue editing area. By obtaining the position coordinates of the existing nodes and the target node in the dialogue editing area, use the position coordinates between any two nodes to determine the distance between any two nodes, and through the analysis of the connection line between any two nodes composed of the existing nodes and the target node in the dialogue editing area, obtain the relative position relationship between any two nodes in the current dialogue editing area; calculate the distance between any two nodes in the current dialogue editing area according to the position coordinates of each node, and obtain the topological relationship between the nodes according to the distance between the nodes and the relative position relationship.

[0067] Step S104, determine the node attribute information corresponding to each of the nodes, and extract the corresponding node generation sub-process according to the node attribute information and the topological relationship of each node in the dialogue editing area. The sub-process includes at least a sub-process name, a sub-process category, an intent category, and a sub-process description;

[0068] Among them, the topological relationship includes the logical relationship (topological structure relationship) between each node, that is, the business process between each node can be expressed in the form of a directed graph, and each node reversely constitutes the entire business process.

[0069] In the above manner, it is possible to quickly enter the process editing mode of the dialogue robot. By editing the process in the editing page and the topological relationship of each node constituting the process, the specific functions of each node can be clearly understood, which is convenient for subsequent actual applications.

[0070] Step S105, associatively store the sub-processes according to the sub-process names by using a reference relationship table;

[0071] Among them, the reference relationship table associatively stores sub-processes according to sub-process parameters such as sub-process names, sub-process categories, intent categories, and sub-process descriptions. For example, the reference relationship table forms a mapping relationship, and in a way similar to a directory, it is convenient for subsequent calls.

[0072] Step S106, call the corresponding sub-process according to the sub-process name in the reference relationship table, and configure the sub-process according to the business process until the sub-process configuration is completed.

[0073] Among them, the business type or intent of the business process to be configured currently and the sub-process name in the reference relationship table call the corresponding sub-process, and then integrate the sub-process into the business process to be configured to achieve integration, which improves the efficiency of the process configuration of the dialogue robot.

[0074] In this embodiment, during the process of making the dialogue process of the robot, the general and reusable processes can be saved as sub-processes, and the sub-processes can be introduced in the robot editing, so as to avoid repeated operations of process editing and save the time for configuring the process; at the same time, sub-process parameters, functions, and tasks are associated and bound in the sub-process, and they can automatically fit perfectly with the robot that references the sub-process during operation. If there are any non-fitting places, error prompts will be given during the editing process, so that the template configurator does not need to pay too much attention, simplifying many operations and potential running bugs.

[0075] Optionally, before obtaining the configuration instruction of the dialogue robot and sending a dialogue editing page request to the server according to the configuration instruction of the dialogue robot, it includes:

[0076] Receive a robot information configuration instruction, where the robot identifier is carried in the information configuration instruction, obtain the corresponding information configuration page according to the robot identifier and display it; receive an information configuration completion instruction, obtain the basic configuration information according to the information configuration completion instruction, and send the basic configuration information and the robot identifier to the server.

[0077] Among them, the robot identifier is used to uniquely identify the robot to be configured. The robot identifier can be a name, a digital ID, a string, etc. The information configuration page is a page for configuring the basic information of the robot.

[0078] Specifically, the terminal receives a robot information configuration instruction, which carries the robot identifier to be configured. The terminal parses the robot information configuration instruction to obtain the robot identifier, and sends an information configuration page acquisition request to the server according to the robot identifier. After receiving the request, the server returns an information configuration page to the terminal, and the terminal acquires and displays the information configuration page returned by the server.

[0079] Specifically, the terminal obtains the basic configuration information of the robot through the information configuration page. At this time, the terminal receives an information configuration completion instruction through the information configuration page, obtains the basic configuration information according to the information configuration completion instruction, and then sends the basic configuration information and the robot identifier to the server. After receiving the basic configuration information and the robot identifier, the server associates and saves the basic configuration information and the robot identifier.

[0080] In the above embodiment, the basic configuration information of the robot is pre-configured through the information configuration page, and the basic configuration information and the robot identifier are associated and saved, improving the efficiency of configuring the dialogue robot for the robot.

[0081] Optionally, when it is detected that the target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area, determine the distance between the nodes in the dialogue editing area according to the position coordinates, and determine the topological relationship between the nodes in combination with the distance and relative position relationship between the nodes. For details, see Figure 2 This is the schematic diagram of the distance and relationship determination process in the sub-process configuration method based on a dialogue robot provided in an embodiment of the present invention, which is described in detail as follows:

[0082] Step S201, when it is detected that the target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area;

[0083] Specifically, for example, obtain the selected target node in the current preset node area, move the target node in the dialogue editing area, and obtain the position coordinates of the target node in the current dialogue editing area.

[0084] For another example, obtain the point of interest (POI) data of at least two data sources of the target node. The POI data has corresponding position coordinates; cluster the position coordinates of the POI data of at least two data sources to obtain at least two clustering clusters; each clustering cluster includes at least one POI data; score at least two clustering clusters respectively, and determine the position coordinates of the target node according to the position coordinates of the POI data in each clustering cluster corresponding to the scoring result.

[0085] Step S202: Parse the connection lines between any two nodes composed of the existing nodes and the target node in the dialogue editing area to obtain the relative position relationship between any two nodes in the current dialogue editing area.

[0086] Specifically, by parsing (such as using regular expression rules) the connection lines between any two nodes, the relative position relationship between any two nodes in the current dialogue editing area can be obtained. For example, the connection relationship, directed relationship, etc. between the nodes.

[0087] Step S203: Calculate the distance between any two nodes in the current dialogue editing area according to the position coordinates of each node, and obtain the topological relationship between the nodes based on the distance and relative position relationship between the nodes.

[0088] Specifically, after obtaining the position coordinates of any two nodes, the distance between the nodes can be determined using the coordinate distance (such as the Pythagorean theorem), and the topological relationship between each node can be obtained based on the distance and relative position relationship between the nodes.

[0089] In this embodiment, the distance and relative position relationship between the nodes in the current robot dialogue process can be obtained to get the topological relationship between each node, which is beneficial to subsequent segmentation or extraction of one or more sub-processes.

[0090] Optionally, determine the node attribute information corresponding to each node, and extract the corresponding nodes according to the node attribute information and topological relationship of each node in the dialogue editing area to generate sub-processes. For details, see Figure 3 , which is the schematic diagram of the sub-process generation process in the sub-process configuration method based on a dialogue robot provided in an embodiment of the present invention, and is described in detail as follows:

[0091] Step S301: Extract the node configuration information of each node in the dialogue editing area, and determine the node attribute information corresponding to each node according to the node configuration information of each node. The node attribute information includes control nodes and dialogue nodes.

[0092] Among them, since each node has initial configuration information during initial configuration, when extracting the node configuration information of each node in the dialogue editing area, if the configuration information of the node has not been changed, the initial configuration information is used as the configuration information of the unchanged node. At the same time, if the configuration information has been changed, the changed configuration information of the node can be called.

[0093] Step S302: Determine the topological relationship between the nodes according to the position relationship formed by the connection lines between any two nodes in the dialogue editing area. The types of the connection lines include non-intentional lines and intentional lines.

[0094] Among them, within the dialogue editing area, the connections between each node form the positional relationship between the nodes, and the topological relationship between the nodes is determined by this positional relationship. The type of this connection selects the corresponding non-intent line or intent line according to whether it is a control node or a conversation node.

[0095] Step S303: Divide the node attribute information and topological relationship of each node in the dialogue editing area according to the intent of the business process to generate at least one initial sub-process.

[0096] Among them, since there may be multiple intents in the dialogue editing area, multiple sub-processes can be divided according to the types of intents.

[0097] Step S304: Adjust the node attribute information and node connections in the generated initial sub-process with reference to the sub-process rules to generate a sub-process that conforms to the sub-process rules. One such sub-process corresponds to at least one version.

[0098] Among them, the sub-process rules can be set artificially. For example, the type of the node before the end node of a certain sub-process, the type of the connection, etc. can be set, which will not be elaborated here.

[0099] For example, by adding a new sub-process, it also includes:

[0100] Use the newly added sub-process to draw nodes to obtain a list of selectable sub-processes; determine a certain sub-process in the sub-process table, use the front end to store the business ID of the sub-process into the sub-process, and save the type of the sub-process node.

[0101] Among them, the sub-process list includes the dialogue robot ID, the dialogue robot name, the robot release status, and the operation process; the sub-process also includes a voice sub-process and a text sub-process. Therefore, a voice robot can only reference the voice sub-process, and a text robot can only reference the text sub-process.

[0102] Through the above method, during the process of making the dialogue process of the dialogue robot, the general and reusable processes are saved as sub-processes, and the sub-processes are introduced in the robot editing, thus avoiding repeated operations in process editing and saving the time for configuring the process.

[0103] Optionally, before using the reference relationship table to associate and store the sub-processes according to the sub-process name, it further includes:

[0104] Verify the sub-process according to the preset verification rules of the sub-process. After the verification is qualified, associate and store the sub-process with the reference relationship table according to the sub-process name; among them, the preset verification rules include at least one of the following:

[0105] When it is detected that the current sub - process is called by a robot, the sub - process cannot be taken offline; if you want to modify the sub - process, you can only copy the sub - process and modify it in the form of a new template;

[0106] Determine whether the type of a node before the end node in the current sub - process has changed. If it has changed and the current sub - process is called by a robot, then the sub - process cannot be used (i.e., publish the sub - process).

[0107] For example, when using the relationship between each dialogue robot and the sub - process in the reference relationship table, obtain the reference relationship table of the dialogue robot and the sub - process; when receiving the save and publish instructions, determine whether there are parameters with the same name but different types in the parameter set of the save and publish interface and the parameter sets of the sub - processes referenced by all dialogue robots. If not, the verification is successful and the publish operation and save operation can be performed. If so, the verification fails and no operation is performed.

[0108] Specifically, the intent, function functions of the sub - process (each sub - process node contains a function function, each function function has a function name, and the parameters of each function function include a processing class and an entity variable) and the tasks do not need to be verified and are configured in the corresponding template of the sub - process; by calling the parameters, functions, tasks, etc. in the sub - process, a perfect fit of the robot process configuration is achieved. Incompatible places will report errors during the editing process. At the same time, it enables template configurators to pay less attention, simplifying many operations and potential running bugs.

[0109] When it is detected that the front - end graph is converted into the back - end graph, change the conversion logic of the newly added sub - process. Among them, a new sub - process node object is added to the back - end graph, and the sub - process node stores the sub - process business ID;

[0110] When the save and publish interface of the dialogue robot completes the new verification, determine that all referenced sub - processes are in the published state, and save and publish the dialogue robot.

[0111] In some other embodiments, if a sub - process is referenced by a published robot, the sub - process cannot be directly taken offline. If you want to modify it, you can copy the sub - process, modify the new version of the sub - process and publish it to replace the original published sub - process.

[0112] When publishing a sub - process, it is necessary to clean the cache of the robot that references the sub - process.

[0113] By the above method, adjust the save and publish interface of the dialogue robot and verify it. Only the sub - processes that meet the verification can be saved and published, which can ensure the accuracy of the sub - processes and avoid the phenomenon of randomly referencing sub - processes in the future.

[0114] Optionally, call the corresponding subprocess according to the subprocess name in the reference relationship table, and configure the subprocess according to the business process. For details, see Figure 4 , which is a schematic diagram of the subprocess configuration process in the subprocess configuration method based on a dialogue robot provided in an embodiment of the present invention, and is described in detail as follows:

[0115] Step S401: Extract the list of intents associated with the business process of the currently to-be-configured subprocess. The list of intents includes at least one intent;

[0116] Among them, intents can be generated according to keywords in the current dialogue process to form a list of intents. For example, a pre-trained Bert model is used to perform intent recognition on the keywords in the current dialogue process to obtain the intent recognition result.

[0117] Step S402: Match the intents in the list of intents according to the intent categories corresponding to the subprocess names in the reference relationship table;

[0118] Among them, the matching is performed by the similarity or identity between the intent and the intent category corresponding to the subprocess name in the reference relationship table.

[0119] Step S403: When it is detected that the intent category corresponding to any subprocess name is the same as or similar to the intent, the matching is successful, and the mapping between the subprocess name and the intent of the business process is determined;

[0120] Among them, the similarity is used to determine whether the intent category corresponding to any subprocess name is the same as or similar to the intent, which will not be elaborated here.

[0121] Step S404: Use the intent of the business process to call the associated subprocess, and add the subprocess to the corresponding node for configuration in combination with the positional relationship and node attribute information of each node in the current business process.

[0122] Among them, the subprocess is stored in the form of a template, and each subprocess template is associated and stored in the form of a directory through the reference relationship table. The corresponding subprocess template can be called according to the currently to-be-configured business process. The number of called subprocesses is not limited and is called according to the requirements of the business process.

[0123] Optionally, the matching the intents in the list of intents according to the intent categories corresponding to the subprocess names in the reference relationship table includes:

[0124] Using a text clustering algorithm to cluster the intent categories corresponding to the subprocess names in the reference relationship table; obtaining the category center points of the clustering categories;

[0125] For example, the text clustering algorithm is used to divide the intents corresponding to the sub - process names in the reference relationship table into several text categories; after constructing the tag library, the text clustering algorithm (e.g., tf - idf + kmeans) can be used to cluster the tags in the tag library into several text categories to complete the clustering of intents.

[0126] Calculate the distance value between the intents in the intent list and the category center point;

[0127] For example, through one or several methods such as Euclidean distance, cosine similarity, Manhattan distance, Minkowski distance, Jaccard similarity coefficient, Pearson correlation coefficient, Hamming distance or BM25 correlation score, determine the distance value between the graph and the category center point.

[0128] Compare the distance value between the intents in the intent list and the category center point with a preset distance threshold. When the distance value between the intents in the intent list and the category center point is less than the preset distance threshold, it means that the intents in the intent list are the same or similar to the intent category corresponding to the sub - process name in the reference relationship table.

[0129] For example, by determining whether the intents in the intent list of the currently to - be - configured sub - process are the same or similar to the intent category corresponding to the sub - process name in the reference relationship table, the intents that meet the requirements of the distance value and the preset distance threshold are regarded as successfully matched.

[0130] In this embodiment, clustering is performed according to the intent category of this node to generate a clustering result, and the unrecognized intent text to be adopted is configured into the intent - similar discourse to achieve an efficient and targeted optimization of the robot's intent recognition ability.

[0131] Optionally, in some other embodiments, each sub - process is called using the sub - process names in the reference relationship table, and each sub - process is merged according to the business process to obtain a fused dialogue robot process template. The method of merging each sub - process according to the business flow is described in detail as follows:

[0132] When detecting the type of the corresponding node in the parsed process template, according to the sub - process business ID corresponding to the current node, call the sub - process stored in the backend server by the business ID and parse it, and merge each node into the process template of the dialogue robot according to a preset rule.

[0133] Change all the parent nodes corresponding to the end nodes in the sub - process of the dialogue robot to point to the lower node pointed to by the sub - process node in the dialogue robot;

[0134] Change the parent node corresponding to the sub - process node of the dialogue robot to the first node of the sub - process;

[0135] If the start node of the dialogue robot happens to be a sub - process node, then set the start node of the dialogue robot to the start node within the sub - process.

[0136] Specifically, the attribute information of the node includes at least the type of the node. According to the sub - process business ID corresponding to the current node, call the sub - process stored in the back - end server by the business ID for parsing, and combine each sub - process in the main process according to the preset rules. The combination methods include but are not less than superposition and fusion. For example, split the sub - process into start nodes, intermediate nodes, and end nodes, and realize the fusion of process templates between different sub - processes according to the business logic between each node.

[0137] For example, when editing the exit part of the sub - process and below the robot that references the sub - process, many restrictions and prompts are made for the connection lines and other corresponding rules in the editing interface, so that the template configurator does not need to pay too much attention to these complex logics, and the program will help him adjust or prompt.

[0138] In some embodiments, adjustments are made through DM (process management) to regulate the processing flow of newly added sub - process nodes, and it also includes:

[0139] 1) For the sub - process, the line connecting to the end node is a non - intended line;

[0140] 2) A node before the end node cannot be used as a logic node;

[0141] 3) A node before the end node can be used as a speech node or a control node;

[0142] 4) When the sub - process is published, verify the sub - process before the change, and judge whether the type of the node before the end node has changed. If there is a change, since the current sub - process has been referenced by the inventor, the sub - process cannot be published.

[0143] Specifically, by setting restrictive conditions for the connection lines in the sub - process editing, the original rules and complex logics within the sub - process can be weakened.

[0144] Through the above - mentioned method, on the basis of the original main process, after triggering a specific speech node, call other sub - processes (speech branches) until the dialogue ends and no longer return to the original main process. By configuring the speech sub - process, it helps to improve the intelligence level of the robot's dialogue and meet more diverse scenario requirements. Because after configuring the sub - process, there will be more speech branches and more speech nodes need to be configured, which has a certain complexity. The speech logic of the sub - process can be clarified before configuration.

[0145] It should be noted that the above - mentioned sub - process management table structure is as follows:

[0146] Sub - process table:

[0147] CREATE TABLE public.aicsp_robot_subflow(

[0148] id serial NOT NULL,-- Primary key id

[0149] name varchar(60)NULL,-- Sub - process name

[0150] business_id int4 NOT NULL,-- Sub - process business Id, conceptually the same as the robot's business id (robot_id)

[0151] status int2 NULL DEFAULT 0,-- Completion status 0: Not published 1: Published

[0152] description varchar(300)NULL,-- Sub - process description

[0153] version_id varchar(30)NULL,-- Sub - process version number

[0154] version_desc varchar(1000)NULL,-- Sub - process version description

[0155] robot_param jsonb NULL,-- Input and output parameters called by the sub - process

[0156] robot_info jsonb NULL,-- Front - end graph

[0157] front_graph jsonb NULL,-- Back - end graph

[0158] nlp_lab jsonb NULL,-- The value corresponding to intentLibrary inside is the intent library

[0159] category varchar(30)NULL,-- Sub - process category, divided into CALL and IM

[0160] appid varchar(100)NULL,-- Tenant id

[0161] created_user varchar(60)NULL,-- Creating user

[0162] created_time timestamp NULL,-- Creation time

[0163] updated_user varchar(60) NULL, -- Updated user

[0164] updated_time timestamp NULL -- Update time );

[0166] Many-to-many association table for subprocesses and robots:

[0167] CREATE TABLE public.aicsp_robot_subflow_rel(

[0168] id serial NOT NULL, -- Primary key id

[0169] robot_id int4 NOT NULL, -- Robot primary key id

[0170] subflow_id int4 NOT NULL, -- Business id (business_id) of the subprocess

[0171] created_user varchar(60) NULL, -- Creating user

[0172] created_time timestamp NULL, -- Creation time

[0173] updated_user varchar(60) NULL, -- Updated user

[0174] updated_time timestamp NULL -- Update time

[0175] Among them, the type of the outgoing line connected from the subprocess node is determined by the type of the previous node of the internal end node of the subprocess;

[0176] If it is a speech node, the outgoing line connected from the subprocess node is an intent line;

[0177] If it is a control node, the outgoing line connected from the subprocess node is a non-intent line;

[0178] In the subprocess node, when replacing the subprocess, if the subprocess node has an outgoing line and the types of the previous nodes of the internal end nodes of the old and new subprocesses are different, a prompt will be popped up.

[0179] If control node => speech node, the type of the replaced outgoing line is an intent line;

[0180] If the conversation node => control node has only one outgoing connection line, replace the type of the outgoing connection line with a non-intent line; if there are multiple outgoing connection lines, delete all intent outgoing connection lines.

[0181] This embodiment provides a sub-process configuration method based on a dialogue robot. The method obtains a configuration instruction of the dialogue robot and sends a dialogue editing page request to the server according to the configuration instruction of the dialogue robot; receives the dialogue editing page returned by the server and displays it. The dialogue editing page includes a preset node area and a dialogue editing area; when it is detected that a target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area, and determine the distance and topological relationship between the nodes in the dialogue editing area according to the position coordinates; determine the node attribute information corresponding to each node, and extract corresponding nodes according to the node attribute information and topological relationship of each node in the dialogue editing area to generate a sub-process, which can repeatedly call the sub-process, avoid repeated operations when the dialogue robot references the sub-process, reduce the processing burden of the dialogue robot, make the configuration of the dialogue robot simpler, improve the process configuration efficiency, not only save the configuration time of the sub-process, but also help to improve the dialogue intelligence level of the dialogue robot and meet richer application scenarios.

[0182] In one embodiment, the present invention also provides a sub-process configuration device 500 based on a dialogue robot. Refer to Figure 5 , the device includes:

[0183] An acquisition module 501, configured to obtain a configuration instruction of a dialogue robot, and send a dialogue editing page request to a server according to the configuration instruction of the dialogue robot;

[0184] A dialogue editing module 502, configured to receive the dialogue editing page returned by the server and display it. The dialogue editing page includes a preset node area and a dialogue editing area;

[0185] A distance and relationship determination module 503, configured to obtain the position coordinates of the target node in the dialogue editing area when it is detected that the target node in the preset node area moves into the dialogue editing area, determine the distance between the nodes in the dialogue editing area according to the position coordinates, and determine the topological relationship between the nodes in combination with the distance and relative position relationship between the nodes;

[0186] A sub-process generation module 504, configured to determine the node attribute information corresponding to each node, and extract corresponding nodes according to the node attribute information and topological relationship of each node in the dialogue editing area to generate a sub-process. The sub-process at least includes a sub-process name, a sub-process category, an intent category, and a sub-process description;

[0187] The associated storage module 505 is used to store the sub - processes associated by sub - process names by using a reference relationship table;

[0188] The sub - process configuration module 506 is used to call the corresponding sub - process according to the sub - process name in the reference relationship table, configure the sub - processes according to the business process until the sub - process configuration is completed.

[0189] Optionally, in some embodiments, before the acquisition module 501, there is further included: a configuration information module for

[0190] receiving a robot information configuration instruction, where the robot identifier is carried in the information configuration instruction, obtaining and displaying the corresponding information configuration page according to the robot identifier; receiving an information configuration completion instruction, obtaining basic configuration information according to the information configuration completion instruction, and sending the basic configuration information and the robot identifier to the server.

[0191] Based on the above - mentioned embodiment, the distance and relationship determination module 503 further includes: when detecting that a target node in a preset node area moves into the dialogue editing area, obtaining the position coordinates of the target node in the dialogue editing area; performing parsing processing on the connection lines between any two nodes formed by the existing nodes and the target node in the dialogue editing area to obtain the relative position relationship between any two nodes in the current dialogue editing area; calculating the distance between any two nodes in the current dialogue editing area according to the position coordinates of each node, and obtaining the topological relationship between the nodes according to the distance and relative position relationship between the nodes.

[0192] Based on the above - mentioned embodiment, the sub - process generation module 504 further includes: extracting the node configuration information of each node in the dialogue editing area, determining the node attribute information corresponding to each node according to the node configuration information of each node, where the node attribute information includes control nodes and dialogue nodes; determining the topological relationship between the nodes according to the positional relationship formed by the connection lines between any two nodes in the dialogue editing area, and the types of the connection lines include non - intention lines and intention lines; dividing the node attribute information and topological relationship of each node in the dialogue editing area according to the intention of the business process to generate at least one initial sub - process; adjusting the node attribute information and node connection lines in the generated initial sub - process with reference to the sub - process rules to generate sub - processes that conform to the sub - process rules, and at least one version corresponds to one sub - process.

[0193] Optionally, in some embodiments, before the associated storage module 505, it further includes: a verification module, which is used to verify the sub-process according to the verification rules preset for the sub-process. After the verification is qualified, the sub-process is associated and stored according to the sub-process name and the reference relationship table; wherein, the preset verification rules include at least one of the following:

[0194] When it is detected that the current sub-process is called by the robot, the sub-process cannot be taken offline; if the sub-process needs to be modified, only copy the sub-process and modify it in the form of a new template;

[0195] Judge whether the type of a node before the end node in the current sub-process has changed. If there is a change and the current sub-process is called by the robot, the sub-process cannot be used.

[0196] Optionally, on the basis of the above embodiments, the sub-process configuration module 506 further includes: extracting the intention list associated with the business process of the current sub-process to be configured, where the intention list contains at least one intention; matching the intentions in the intention list according to the intention category corresponding to the sub-process name in the reference relationship table; when it is detected that the intention category corresponding to any sub-process name is the same as or similar to the intention, the matching is successful, and the intention association mapping between the sub-process name and the business process is determined; using the intention of the business process to call the associated sub-process, and adding the sub-process to the corresponding node for configuration in combination with the positional relationship and node attribute information of each node in the current business process.

[0197] Optionally, in the above embodiments, the matching of the intentions in the intention list according to the intention category corresponding to the sub-process name in the reference relationship table includes:

[0198] Using a text clustering algorithm to cluster the intention categories corresponding to the sub-process names in the reference relationship table; obtaining the category center points of the clustering categories;

[0199] Calculating the distance value between the intention in the intention list and the category center point;

[0200] Comparing the distance value between the intention in the intention list and the category center point with a preset distance threshold. When the distance value between the intention in the intention list and the category center point is less than the preset distance threshold, it means that the intention in the intention list is the same as or similar to the intention category corresponding to the sub-process name in the reference relationship table.

[0201] This embodiment provides a sub - process configuration device based on a dialogue robot. The device obtains a configuration instruction of the dialogue robot, and sends a dialogue editing page request to the server according to the configuration instruction of the dialogue robot; receives the dialogue editing page returned by the server and displays it. The dialogue editing page includes a preset node area and a dialogue editing area; when it is detected that a target node in the preset node area moves into the dialogue editing area, it obtains the position coordinates of the target node in the dialogue editing area, and determines the distance and topological relationship between the nodes in the dialogue editing area according to the position coordinates; determines the node attribute information corresponding to each node, and extracts corresponding nodes from the node attribute information and topological relationship of each node in the dialogue editing area to generate a sub - process, which can repeatedly call the sub - process, avoid repeated operations when the dialogue robot references the sub - process, reduce the processing burden of the dialogue robot, make the configuration of the dialogue robot simpler, improve the process configuration efficiency, not only save the configuration time of the sub - process, but also help to improve the dialogue intelligence level of the dialogue robot and meet richer application scenarios.

[0202] It should be known that the above - mentioned sub - process configuration device based on a dialogue robot essentially sets up multiple modules to execute the sub - process configuration method based on a dialogue robot in any of the above - mentioned embodiments. The specific functions and technical effects can be referred to the above - mentioned embodiments and will not be elaborated here.

[0203] In one embodiment, referring to Figure 6 , this embodiment also provides a computer device 600, including a memory 601, a processor 602, and a computer program stored on the memory and executable on the processor. When the processor 602 executes the computer program, it implements the steps of the method described in any of the above - mentioned embodiments.

[0204] In one embodiment, a computer - readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the method described in any of the above - mentioned embodiments.

[0205] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is a theory, method, technology, and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0206] It should be noted that in this text, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, apparatus, article or method comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article or method. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, apparatus, article or method comprising such element.

[0207] The serial numbers of the above-described embodiments of the present invention are for description purposes only and do not represent the superiority or inferiority of the embodiments. Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium as described above (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0208] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the description and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A sub - process configuration method based on a dialogue robot, characterized in that, the method includes: Obtain the configuration instruction of the dialogue robot, and send a dialogue editing page request to the server according to the configuration instruction of the dialogue robot; Receive the dialogue editing page returned by the server and display it. The dialogue editing page includes a preset node area and a dialogue editing area; When it is detected that the target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area, determine the distance between the nodes in the dialogue editing area according to the position coordinates, and determine the topological relationship between the nodes by combining the distance between the nodes and the relative position relationship; Determine the node attribute information corresponding to each node. Extract the corresponding nodes to generate a sub - process according to the node attribute information and topological relationship of each node in the dialogue editing area. The sub - process includes at least a sub - process name, a sub - process category, an intent category, and a sub - process description; among them, extract the node configuration information of each node in the dialogue editing area, determine the node attribute information corresponding to each node according to the node configuration information of each node. The node attribute information includes a control node and a speech node; determine the topological relationship between the nodes according to the connection line between any two nodes in the dialogue editing area. The type of the connection line includes a non - intent line and an intent line; divide the node attribute information and topological relationship of each node in the dialogue editing area according to the intent of the business process to generate at least one initial sub - process; adjust the node attribute information and node connection lines in the generated initial sub - process with reference to the sub - process rules to generate a sub - process that conforms to the sub - process rules. One sub - process corresponds to at least one version; Use a reference relationship table to associate and store the sub - processes according to the sub - process name; Call the corresponding sub - process according to the sub - process name in the reference relationship table, and configure the sub - process according to the business process until the sub - process configuration is completed.

2. The sub - process configuration method based on a dialogue robot according to claim 1, characterized in that, before obtaining the configuration instruction of the dialogue robot and sending a dialogue editing page request to the server according to the configuration instruction of the dialogue robot, it includes: Receive a robot information configuration instruction. The information configuration instruction carries a robot identifier, obtain the corresponding information configuration page according to the robot identifier and display it; receive an information configuration completion instruction, obtain basic configuration information according to the information configuration completion instruction, and send the basic configuration information and the robot identifier to the server.

3. The sub - process configuration method based on a dialogue robot according to claim 1, characterized in that, when it is detected that the target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area, determine the distance between the nodes in the dialogue editing area according to the position coordinates, and determine the topological relationship between the nodes by combining the distance between the nodes and the relative position relationship, including: When it is detected that the target node in the preset node area moves into the dialogue editing area, obtain the position coordinates of the target node in the dialogue editing area; Perform parsing processing on the connection lines between any two nodes composed of the existing nodes and the target node in the dialogue editing area to obtain the relative position relationship between any two nodes in the current dialogue editing area; Calculate the distance between any two nodes in the current dialogue editing area according to the position coordinates of each node, and obtain the topological relationship between the nodes according to the distance between the nodes and the relative position relationship.

4. The sub-process configuration method based on a dialogue robot according to claim 3, characterized in that, before using the reference relationship table to associate and store the sub-processes according to the sub-process names, it further includes: Verify the sub-process according to the preset verification rules of the sub-process. After the verification is qualified, associate and store the sub-process with the reference relationship table according to the sub-process name; wherein, the preset verification rules include at least one of the following: When it is detected that the current sub-process is called by the robot, the sub-process cannot be taken offline; if you want to modify the sub-process, you can only copy the sub-process for modification; Judge whether the type of a node before the end node in the current sub-process has changed. If it has changed and the current sub-process is called by the robot, the sub-process cannot be used.

5. The sub-process configuration method based on a dialogue robot according to any one of claims 1-3, characterized in that, calling the corresponding sub-process according to the sub-process name in the reference relationship table, and configuring the sub-process according to the business process, including: extracting the intent list associated with the business process of the current sub-process to be configured, where the intent list includes at least one intent; matching the intents in the intent list according to the intent categories corresponding to the sub-process names in the reference relationship table; when it is detected that the intent category corresponding to any sub-process name is the same as or similar to the intent, the matching is successful, and the mapping between the sub-process name and the intent of the business process is determined; using the intent of the business process to call the associated sub-process, and adding the sub-process to the corresponding node in combination with the position relationship and node attribute information of each node in the current business process for configuration.

6. The sub-process configuration method based on a dialogue robot according to claim 5, characterized in that, the matching the intents in the intent list according to the intent categories corresponding to the sub-process names in the reference relationship table includes: using a text clustering algorithm to cluster the intent categories corresponding to the sub-process names in the reference relationship table; obtaining the category center points of the clustering categories; calculating the distance value between the intent in the intent list and the category center point; comparing the distance value between the intent in the intent list and the category center point with a preset distance threshold. When the distance value between the intent in the intent list and the category center point is less than the preset distance threshold, it means that the intent in the intent list is the same as or similar to the intent category corresponding to the sub-process name in the reference relationship table.

7. A sub - process configuration device based on a dialogue robot, characterized in that, the device comprises: an acquisition module, configured to acquire a configuration instruction of the dialogue robot and send a dialogue editing page request to a server according to the configuration instruction of the dialogue robot; a dialogue editing module, configured to receive and display the dialogue editing page returned by the server, where the dialogue editing page includes a preset node area and a dialogue editing area; a distance and relationship determination module, configured to, when detecting that a target node in the preset node area moves into the dialogue editing area, acquire the position coordinates of the target node in the dialogue editing area, determine the distance between nodes in the dialogue editing area according to the position coordinates, and determine the topological relationship between nodes by combining the distance between nodes and the relative position relationship; a sub - process generation module, configured to determine the node attribute information corresponding to each node, extract corresponding nodes according to the node attribute information and topological relationship of each node in the dialogue editing area to generate a sub - process, where the sub - process at least includes a sub - process name, a sub - process category, an intent category, and a sub - process description; wherein, extract the node configuration information of each node in the dialogue editing area, determine the node attribute information corresponding to each node according to the node configuration information of each node, and the node attribute information includes a control node and a speech node; determine the topological relationship between nodes from the positional relationship formed by the connection line between any two nodes in the dialogue editing area, and the types of the connection lines include non - intent lines and intent lines; divide the node attribute information and topological relationship of each node in the dialogue editing area according to the intent of the business process to generate at least one initial sub - process; adjust the node attribute information and node connection lines in the generated initial sub - process with reference to the sub - process rules to generate a sub - process that conforms to the sub - process rules, and one sub - process corresponds to at least one version; an associated storage module, configured to associate and store the sub - processes according to the sub - process name by using a reference relationship table; a sub - process configuration module, configured to call the corresponding sub - process according to the sub - process name in the reference relationship table, and configure the sub - process according to the business process until the sub - process configuration is completed.

8. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, when the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer - readable storage medium, on which a computer program is stored, characterized in that, when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images