A dialogue navigation method and electronic device

Abstract

The application discloses a dialogue navigation method and an electronic device. The method comprises the following steps: in response to the appearance of new dialogue data in a dialogue interaction interface, obtaining at least one dialogue prompt label generated based on historical dialogue data in the dialogue interaction interface; determining whether an association condition is met between the new dialogue data and the at least one dialogue prompt label based on dialogue summary information of the at least one dialogue prompt label; in response to the existence of a target dialogue prompt label meeting the association condition in the at least one dialogue prompt label, establishing an association relationship between the new dialogue data and the target dialogue prompt label, and updating dialogue summary information of the target dialogue prompt label based on at least the new dialogue data.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a dialogue navigation method and electronic device. Background Technology

[0002] In the field of artificial intelligence, users can perform a variety of different tasks by interacting with models or intelligent agents through dialogue. For example, they can ask questions, assign tasks, or execute tasks by interacting with models or intelligent agents.

[0003] When a user interacts with a model or agent in a large number of dialogue rounds, if the user wants to view certain dialogue content from past interactions, they may need to sequentially browse through the dialogue data of each round in order to locate the dialogue content they need to review, which is complex and inefficient. Summary of the Invention

[0004] On the one hand, this application provides a dialogue navigation method, including:

[0005] In response to the appearance of new dialogue data in the dialogue interaction interface, at least one dialogue prompt label generated based on historical dialogue data in the dialogue interaction interface is obtained.

[0006] Based on the dialogue summary information of at least one of the dialogue prompt tags, determine whether the newly added dialogue data and at least one of the dialogue prompt tags meet the association conditions.

[0007] In response to the existence of a target dialogue prompt tag that meets the association conditions in at least one of the dialogue prompt tags, an association relationship is established between the newly added dialogue data and the target dialogue prompt tag, and the dialogue summary information of the target dialogue prompt tag is updated based at least on the newly added dialogue data.

[0008] In one possible implementation, the dialogue navigation method also includes:

[0009] In response to the absence of a target dialogue prompt tag that meets the association conditions in at least one of the dialogue prompt tags, a new dialogue prompt tag associated with the new dialogue data is generated, and the dialogue summary information corresponding to the new dialogue prompt tag is determined;

[0010] The newly added dialogue prompt label is displayed on the dialogue interaction interface.

[0011] In yet another possible implementation, the dialogue navigation method also includes:

[0012] Based at least on the dialogue summary information of the newly added dialogue prompt tag and the dialogue summary information of at least one of the dialogue prompt tags, a first association relationship between the newly added dialogue prompt tag and at least one of the dialogue prompt tags is determined;

[0013] The newly added dialogue prompt label is displayed on the dialogue interaction interface, including:

[0014] Based on the first association relationship, determine the display arrangement relationship between the newly added dialogue prompt label and at least one of the dialogue prompt labels;

[0015] Based on the display layout, at least one of the dialogue prompt labels and the newly added dialogue prompt label are displayed in the dialogue interaction interface.

[0016] In another possible implementation, determining whether the newly added dialogue data and at least one of the dialogue prompt tags satisfy an association condition based on the dialogue summary information of at least one of the dialogue prompt tags includes:

[0017] Determine the matching result between the newly added dialogue data and the dialogue summary information of the dialogue prompt label on at least one feature dimension, wherein the matching result is used to characterize whether the newly added dialogue data and the dialogue prompt label are related on the feature dimension;

[0018] Based on the matching results corresponding to the dialogue prompt tags, it is determined whether there is a correlation between the newly added dialogue data and the dialogue prompt tags.

[0019] In another possible implementation, determining the matching result of the newly added dialogue data and the dialogue summary information of the dialogue prompt label in at least one feature dimension includes at least one of the following:

[0020] Based on the first semantic feature of the newly added dialogue data and the second semantic feature corresponding to the dialogue summary information of the dialogue prompt tag, it is determined whether there is a semantic correlation between the newly added dialogue data and the dialogue prompt tag;

[0021] Based on the first intent feature corresponding to the newly added dialogue data and the second intent feature corresponding to the dialogue summary information of the dialogue prompt label, it is determined whether there is a matter correlation between the first matter described by the newly added dialogue data and the second matter described by the dialogue summary information of the dialogue prompt label.

[0022] The step of determining whether there is a correlation between the newly added dialogue data and the dialogue prompt tag based on the matching result corresponding to the dialogue prompt tag includes:

[0023] Determine whether the newly added dialogue data and the dialogue prompt label have at least one of the semantic relevance and the event association.

[0024] In yet another possible implementation, the dialogue navigation method also includes at least one of the following:

[0025] In response to the existence of a target dialogue prompt tag that meets the association conditions in at least one of the dialogue prompt tags, the tag name of the target dialogue prompt tag is updated based on the newly added dialogue data;

[0026] In response to updating the dialogue summary information corresponding to the target dialogue prompt label or generating the new dialogue prompt label, the dialogue prompt labels in the dialogue interaction interface are merged or split.

[0027] In yet another possible implementation, the dialogue navigation method also includes:

[0028] Based on the dialogue summary information corresponding to the updated target dialogue prompt tag, a second association relationship between the updated target dialogue prompt tag and other dialogue prompt tags is determined;

[0029] Based on the second association relationship, the display arrangement relationship between the updated target dialogue prompt label and the other dialogue prompt labels in the dialogue interaction interface is adjusted.

[0030] In another possible implementation, determining the second association between the updated target dialogue prompt tag and other dialogue prompt tags based on the dialogue summary information corresponding to the updated target dialogue prompt tag includes:

[0031] Based on the user's operation information on the target dialogue prompt tag and other dialogue prompt tags, adjust the weight information between the target dialogue prompt tag and other dialogue prompt tags;

[0032] Based on the weight information and the dialogue summary information corresponding to the updated target dialogue prompt label, a second association relationship between the updated target dialogue prompt label and other dialogue prompt labels is determined.

[0033] In yet another possible implementation, the dialogue navigation method also includes at least one of the following:

[0034] In response to the user's first action on the dialog prompt label, display the dialog summary information corresponding to the dialog prompt label;

[0035] In response to a second action by the user on the dialog prompt label, at least one piece of historical dialogue data associated with the dialog prompt label is displayed.

[0036] In another aspect, this application also provides an electronic device comprising: a display module and a processor;

[0037] The display module is used to display the interactive dialogue interface;

[0038] The processor is configured to: respond to the appearance of new dialogue data in the dialogue interaction interface; obtain at least one dialogue prompt tag generated based on historical dialogue data in the dialogue interaction interface; determine whether the new dialogue data and at least one dialogue prompt tag satisfy an association condition based on the dialogue summary information of the at least one dialogue prompt tag; and, in response to the existence of a target dialogue prompt tag that satisfies the association condition among the at least one dialogue prompt tag, establish an association relationship between the new dialogue data and the target dialogue prompt tag, and update the dialogue summary information of the target dialogue prompt tag based at least on the new dialogue data. Attached Figure Description

[0039] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0040] Figure 1 A flowchart illustrating the dialogue navigation method provided in this application;

[0041] Figure 2 Another flowchart illustrating the dialogue navigation method provided in this application;

[0042] Figure 3 Another flowchart illustrating the dialogue navigation method provided in this application;

[0043] Figure 4 Another flowchart illustrating the dialogue navigation method provided in this application;

[0044] Figure 5 This is an example diagram showing multiple dialogue prompt labels displayed in a dialog interaction interface;

[0045] Figure 6 An example diagram is shown showing how the dialogue prompt labels in the dialogue interaction interface are updated in response to the presence of new dialogue data.

[0046] Figure 7 This is a schematic diagram illustrating one implementation process of merging dialog prompt labels in this application;

[0047] Figure 8 This is a schematic diagram illustrating one implementation process of splitting dialog prompt labels in this application;

[0048] Figure 9 This is an example diagram of a navigation area used to display dialog prompt labels in this application being in an editable state;

[0049] Figure 10 A schematic diagram of the composition structure of an electronic device provided in this application. Detailed Implementation

[0050] The solution proposed in this application can be applied to intelligent dialogue scenarios based on models or intelligent agents, such as dialogue interaction based on intelligent agents or models to achieve purposes such as problem consultation, task allocation, task instruction and processing, without any specific limitations.

[0051] The solution proposed in this application enables a more efficient and convenient way to locate the dialogue content that needs to be reviewed from historical dialogue data.

[0052] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0053] As will be known to those skilled in the art, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0054] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0055] like Figure 1 This is a flowchart illustrating a dialogue navigation method provided in this application. The method in this embodiment can be applied to electronic devices, such as mobile terminals like mobile phones or laptops; it can also be a server or a device node in a distributed system, etc., without limitation.

[0056] The method in this embodiment may include:

[0057] S101, in response to the appearance of new dialogue data in the dialogue interaction interface, obtain at least one dialogue prompt label generated based on historical dialogue data in the dialogue interaction interface.

[0058] The dialogue interaction interface is the interface output by the electronic device, used for users to interact with models or intelligent agents through dialogue.

[0059] New dialogue data refers to newly generated dialogue content in the dialogue interaction interface.

[0060] The newly added dialogue data can be a dialogue content, such as the question entered by the user or the response output by the electronic device based on the model or intelligent agent.

[0061] In one alternative approach, in order to reduce the amount of data processing and to accurately determine the dialogue prompt tags associated with the new dialogue data, the new dialogue data can be a set of dialogue data, also known as a round of dialogue data, which may include: the most recently entered question content in the dialogue interaction interface and the response content to that question content.

[0062] In this application, a dialogue prompt label is associated with at least one historical dialogue data in the dialogue interaction interface, and the dialogue prompt label is used to locate the at least one historical dialogue data. For example, the dialogue prompt label is associated with the start and end positions corresponding to at least one historical dialogue data (i.e., the context boundary of the historical dialogue data associated with the dialogue prompt label). The historical dialogue data can be dialogue data generated in the dialogue interaction interface in the past; the historical dialogue data can be a single dialogue content or a set of dialogue data (i.e., a round of dialogue data), without any specific limitations.

[0063] S102, based on the dialogue summary information of at least one dialogue prompt label, determine whether the newly added dialogue data and at least one dialogue prompt label meet the association conditions.

[0064] The dialogue summary information of a dialogue prompt tag is a semantic summary of at least one historical dialogue data associated with that dialogue prompt tag. For example, the dialogue summary information of that dialogue prompt tag can be generated by performing semantic analysis and summary extraction on at least one historical dialogue data associated with that dialogue prompt tag.

[0065] The dialogue summary information of the dialogue prompt label can reflect the key semantics of each historical dialogue data associated with the dialogue prompt label, the task status of the task to which it belongs, the user intent or the problem to be dealt with and its processing stage, and can also characterize the basis for the formation of the dialogue prompt label.

[0066] The association condition between newly added dialogue data and dialogue prompt tags indicates a correlation between the newly added dialogue data and the historical dialogue data associated with that dialogue prompt tag. For example, the association condition between newly added dialogue data and dialogue prompt tags can be: the historical dialogue data associated with the newly added dialogue data and the dialogue prompt tag belong to the same topic, the same plan (such as different chapters of the same plan), the same discussion goal (such as the discussion content under the same discussion or different directions of the same goal), the same task (such as different execution steps or approval stages of the same task), or other interaction data, such as the newly added dialogue data and the historical dialogue data associated with the dialogue prompt tag having other contextual or causal relationships, etc., without any specific restrictions.

[0067] For example, the historical dialogue data associated with the dialogue prompt label includes historical dialogue data related to the submission of an application pending approval. New dialogue data consists of interaction data regarding the approval status of that application; therefore, the new dialogue data and the dialogue prompt label satisfy the association condition.

[0068] S103, in response to the existence of a target dialogue prompt label that meets the association conditions in at least one dialogue prompt label, establish an association relationship between the newly added dialogue data and the target dialogue prompt label, and update the dialogue summary information of the target dialogue prompt label based at least on the newly added dialogue data.

[0069] In this application, for ease of distinction, the dialogue prompt label that meets the association condition with the newly added dialogue data is referred to as the target dialogue prompt label. There may be one or more target dialogue prompt labels, without limitation.

[0070] The process of establishing the association between the newly added dialogue data and the target dialogue prompt label can be either a correspondence between the index or other identification information of the newly added dialogue data and the label name or other label identifier of the target dialogue prompt label, or storing the identification information of the newly added dialogue data in the dialogue data set corresponding to the target dialogue prompt label, without any specific restrictions.

[0071] As introduced above, if the newly added dialogue data meets the association conditions with an existing dialogue prompt label, it means that the newly added dialogue data and the historical dialogue data associated with the dialogue prompt label belong to the same topic, the same task stage, or have other relationships. In this case, establishing the association relationship between the newly added dialogue data and the target dialogue prompt label can link related dialogue data to the same dialogue prompt label, so that users can view related dialogue data based on the dialogue prompt label in the future.

[0072] There are several ways to update the dialogue summary information of the target dialogue prompt label based on the newly added dialogue data, and there are no restrictions on this.

[0073] For example, in one possible implementation, the summary information of the newly added dialogue data can be added to the dialogue summary information, or the dialogue summary information of the target dialogue prompt tag can be regenerated based on the dialogue summary information of the target dialogue prompt tag and the newly added dialogue data, or the dialogue summary information of the newly added dialogue data, so as to update the dialogue summary information of the target dialogue prompt tag.

[0074] In a second possible implementation, updating the dialogue summary information of the target dialogue prompt label can also be based on the newly added dialogue data and at least one historical dialogue data associated with the target dialogue prompt label, and regenerating the dialogue summary information of the target dialogue prompt label.

[0075] Of course, since updating the dialogue summary information using the second possible implementation method requires processing a relatively large amount of data, this application may also prioritize the first possible implementation method and use the second possible implementation method once every set time interval to update the dialogue summary information of the currently determined target dialogue prompt tag.

[0076] In this embodiment, when new dialogue data appears in the dialogue interaction interface, it is first determined whether the new dialogue data meets the association conditions with existing dialogue prompt tags. If there is a target dialogue prompt tag among the existing dialogue prompt tags that meets the association conditions with the new dialogue data, the new dialogue data will be associated with the target dialogue prompt tag. This allows for efficient location of historical dialogue data based on dialogue prompt tags, reducing the complexity of data querying.

[0077] Furthermore, unlike generating a new dialogue prompt label every time new dialogue data is generated, in this embodiment, if the new dialogue data and an existing dialogue prompt label meet the association conditions, this application will establish an association between the new dialogue data and the existing dialogue prompt label. This associates historical dialogue data with a relationship with the same dialogue prompt label, rather than creating a new dialogue prompt label. This allows for continuous updating of the dialogue prompt labels in the dialogue interaction interface as the dialogue data in the interface changes continuously. This not only reduces the situation of blindly adding dialogue prompt labels as the dialogue data in the interface increases, but also reduces data redundancy in dialogue prompt labels. As a result, it can reduce the situation where the inability to efficiently query dialogue prompt labels is caused by an excessive number of dialogue prompt labels.

[0078] Moreover, the dialogue prompt labels in the dialogue interaction interface can continuously evolve as the dialogue data progresses. A dialogue prompt label can be associated with multiple dialogue data that are not continuous in time but are related. This allows users to locate multiple related dialogue data based on the dialogue prompt labels, enabling users to better review the content and more efficiently retrieve the information they need based on the relevant dialogue data.

[0079] In this application, in order to more conveniently and intuitively distinguish different dialog prompt labels, the dialog prompt labels have label names.

[0080] The name of a dialog hint label can also be called its label content; it refers to the content displayed when the dialog hint label is shown. Therefore, the label name can be used to indicate the topic of at least one historical dialogue data point. The label name provides a clear understanding of the topic type and direction described by the at least one historical dialogue data point associated with the dialog hint label.

[0081] To quickly identify the topic of the historical dialogue data associated with a dialog prompt tag based on its tag name, the tag name should be shorter and more concise than the dialogue summary information. For example, the tag name could be "Prospects for Artificial Intelligence," "English Grammar," or "Introduction to Computer Science."

[0082] The dialogue prompt label is generated based on the dialogue data associated with it. For example, if at least one historical dialogue data associated with the dialogue prompt label mainly involves questions, definitions, specific compression processes, and implementation methods related to knowledge compression, then the label name for the dialogue prompt label could be "Knowledge Compression Principles".

[0083] In one alternative approach, if a target dialogue prompt tag that meets the target conditions is determined to exist, this application can also update the tag name of the target dialogue prompt tag based on the newly added dialogue data. For example, by combining the newly added dialogue data and the dialogue summary information of the target dialogue prompt tag, the tag name of the target dialogue prompt tag is regenerated to update the tag name of the target dialogue prompt tag.

[0084] As the dialogue data in the interactive interface changes, this application can dynamically update the dialogue data, dialogue summary information, and tag name associated with the dialogue prompt tag, so that the dialogue prompt tag can more accurately and flexibly establish a relationship with the dialogue data in the interactive interface, and can intuitively reflect the topic information of the dialogue data associated with the dialogue prompt tag through the tag name.

[0085] In this application, at least one dialogue prompt label may be displayed in the dialogue interaction interface; alternatively, at least one dialogue prompt label already generated on the dialogue interaction interface may be displayed when a target operation is detected, so that the user can flexibly operate the dialogue prompt label. The target operation may be an operation that triggers the display of the label menu, and there are no specific limitations.

[0086] Among these features, dialog prompt labels can be displayed in the navigation area of ​​the dialog interaction interface.

[0087] For example, the navigation area can be located at the top, side or bottom of the dialogue interface, and the navigation area can be presented in the form of a floating window or other forms, without any specific restrictions.

[0088] In this application, users can display corresponding historical dialogue data or dialogue summary information corresponding to historical dialogue data by operating the dialogue prompt labels.

[0089] For example, in one instance, in response to the user's first action on the dialog hint label, the corresponding dialog summary information is displayed. This first action can be controlling the cursor to hover over the dialog hint label using a mouse or other control, or setting a set duration for the cursor to remain on the dialog hint label, etc., without any specific limitations.

[0090] The dialog summary information corresponding to the dialog prompt label can be displayed near or adjacent to the dialog prompt label's display window or information display area, and displayed within that window or information display area. Of course, there are other possibilities for displaying the dialog summary information of the dialog prompt label, and there are no restrictions on this.

[0091] By displaying the conversation summary information of the conversation prompt label, users can understand the information summary described by the historical conversation data associated with the conversation prompt label, so as to determine whether the historical conversation data associated with the conversation prompt label is the conversation data that the user needs to view.

[0092] In yet another example, in response to a second action by the user on the dialog hint label, at least one piece of historical dialogue data associated with the dialog hint label is displayed. This second action may differ from the first action. For example, the second action could be clicking the dialog hint label, or long-pressing the dialog hint label, etc., without limitation.

[0093] The historical dialogue data can be displayed in the dialogue interaction area of ​​the dialogue interface. For example, the historical dialogue data can be displayed sequentially according to the order in which it was generated.

[0094] In this application, if no target dialogue prompt label exists that matches the association condition with the newly added dialogue data, a new dialogue prompt label can be generated for the newly added dialogue data. The following section combines... Figure 2 Please provide an explanation. For example... Figure 2 This is another flowchart illustrating the dialogue navigation method provided in this application. This embodiment can be applied to the aforementioned electronic devices.

[0095] The method in this embodiment may include:

[0096] S201, in response to the appearance of new dialogue data in the dialogue interaction interface, obtain at least one dialogue prompt label generated based on historical dialogue data in the dialogue interaction interface.

[0097] S202, based on the dialogue summary information of at least one dialogue prompt label, determine whether the newly added dialogue data and at least one dialogue prompt label meet the association conditions.

[0098] S203, in response to the existence of a target dialogue prompt label that meets the association conditions in at least one dialogue prompt label, establish an association relationship between the newly added dialogue data and the target dialogue prompt label, and update the dialogue summary information of the target dialogue prompt label based at least on the newly added dialogue data.

[0099] The above steps can be found in the relevant descriptions of the previous embodiments, and will not be repeated here.

[0100] S204, in response to the absence of a target dialogue prompt label that meets the association conditions in at least one dialogue prompt label, generate a new dialogue prompt label associated with the new dialogue data, and determine the dialogue summary information corresponding to the new dialogue prompt label.

[0101] For ease of distinction, the dialogue prompt label generated for this new dialogue data will be called the new dialogue prompt label.

[0102] Generating the new dialogue prompt label may include: adding a new dialogue prompt label to the dialogue interaction interface and establishing a relationship between the new dialogue prompt label and the new dialogue data.

[0103] Furthermore, this application can also determine the tag name of the new dialogue prompt label based on the newly added dialogue data. For example, based on the newly added dialogue data or the summary information of the newly added dialogue data, keywords that can represent the topic of the newly added dialogue data can be determined, and these keywords can be used as the tag name of the new dialogue prompt label. Of course, there are other ways to generate the tag name, and there are no restrictions on this.

[0104] The dialogue summary information for the newly added dialogue prompt label can be a semantic summary of the newly added dialogue data. For example, a semantic summary that can characterize the semantic content of the newly added dialogue data can be determined and used as the dialogue summary information for the newly added dialogue prompt label.

[0105] S205, a new dialogue prompt label is displayed on this dialogue interaction interface.

[0106] In this embodiment, when the newly added dialogue data does not meet the association conditions with any of the existing dialogue prompt labels, a new dialogue prompt label can be generated for the newly added dialogue data. This allows dialogue prompt labels to be added along with the newly added dialogue data in the dialogue interaction interface, enabling users to efficiently and conveniently locate the newly added dialogue data based on the newly added dialogue prompt labels.

[0107] In this embodiment, displaying the newly added dialogue prompt label in the dialogue interaction interface can be done by displaying the dialogue prompt label in the navigation area of ​​the dialogue interaction interface. For details, please refer to the previous related introduction, which will not be repeated here.

[0108] It is understandable that while displaying the newly added dialogue prompt label in the dialogue interface, at least one existing dialogue prompt label can also be displayed in the same dialogue interface. In this application, the newly added dialogue prompt label and the existing at least one dialogue prompt label can be displayed sequentially in the dialogue interface. For example, based on the generation time of the newly added dialogue prompt label and the existing dialogue prompt label, the existing dialogue prompt label and the newly added dialogue prompt label can be displayed sequentially in the navigation area of ​​the dialogue interface.

[0109] In one alternative approach, after constructing the new dialogue prompt label, the application may further determine a first association between the new dialogue prompt label and at least one existing dialogue prompt label, based at least on the dialogue summary information of the new dialogue prompt label and the dialogue summary information of at least one existing dialogue prompt label.

[0110] The first association relationship may include at least one of the following: the order of generation time, the weight relationship, the topic association relationship, and the dependency relationship between the newly added dialogue prompt tag and at least one dialogue prompt tag.

[0111] The order of generation time indicates the sequence between the generation time of the new dialogue prompt tag (or the dialogue summary information of the new dialogue prompt tag) and the generation time of the dialogue prompt tag (or the dialogue summary information of the dialogue prompt tag).

[0112] This weighting order indicates the relative weights of newly added dialog hint tags and other dialog hint tags. The weight of a newly added dialog hint tag can be a default value, or it can be determined based on the corresponding dialog summary information or newly added dialog data.

[0113] The weight of existing dialog prompt tags can be determined based on the user's actions on those tags prior to the current moment. These actions may include, but are not limited to, accessing, clicking, editing, favorited, or modifying the tag name or associated historical dialogue data of the dialog prompt tag, as well as adding questions to the historical dialogue data associated with the dialog prompt tag, and actions such as lingering on the dialog prompt tag or its associated historical dialogue data.

[0114] The user's operation information for the dialog prompt tag can represent the number of times or frequency of operations performed by the user on the dialog prompt tag or the historical dialogue data associated with the dialog prompt tag. The higher the number of times or frequency of operations represented by the operation information, the higher the weight of the dialog prompt tag.

[0115] Accordingly, the weight of the newly added dialogue prompt tag can be updated based on the user's operation information regarding the newly added dialogue prompt tag, which will not be elaborated further.

[0116] The topic association relationship is used to characterize whether there is a relationship, and the specific relationship, between the first topic content corresponding to the dialogue summary information of a newly added dialogue prompt tag and the second topic content corresponding to the dialogue summary information of an existing dialogue prompt tag. For example, if the first topic content corresponding to a newly added dialogue prompt tag and the second topic content corresponding to the dialogue summary information of a certain dialogue prompt tag are two sub-topics under the same general topic, then the newly added dialogue prompt tag and that dialogue prompt tag belong to the same topic.

[0117] The dependency relationship is the association between the new dialogue data corresponding to the new dialogue prompt label and the historical dialogue data of the existing dialogue prompt labels, determined based on the dialogue summary information of the new dialogue prompt label and the dialogue summary information of the existing dialogue prompt labels, in terms of the tasks, plans, plan chapters or items they belong to.

[0118] For example, the dependency relationship can be: the new dialogue data is a task branch or task step under the task indicated by the historical dialogue data of the dialogue prompt label, or the new dialogue data and the historical dialogue data of the dialogue prompt label belong to the discussion content of different task steps under the same task, etc.

[0119] For example, the dependency relationship is: the new dialogue data is a contextual relationship in which the solution is further optimized, supplemented or otherwise evolved based on the solution discussed in the historical dialogue data of the dialogue prompt label.

[0120] Accordingly, displaying the newly added dialogue prompt label can be done by: determining the display arrangement relationship between the newly added dialogue prompt label and at least one dialogue prompt label based on the first association relationship; and displaying at least one dialogue prompt label and the newly added dialogue prompt label in the dialogue interaction interface based on the display arrangement relationship.

[0121] In this application, the display arrangement relationship between the newly added dialogue prompt labels and at least one dialogue prompt label may include, but is not limited to, the arrangement, graph structure, tree structure relationship or other topological structure relationship presented on the dialogue interaction interface.

[0122] For example, by combining the dependencies, weight relationships, generation order, and topic associations between the newly added dialog hint label and at least one existing dialog hint label, a tree structure diagram containing the relationships between the new dialog hint label and existing dialog hint labels can be constructed and displayed on the dialog interaction interface. This tree structure diagram can reflect the various relationships and hierarchical relationships between the new dialog interface and each dialog hint label, allowing users to intuitively understand the relationships between the dialog hint labels based on the tree structure diagram.

[0123] In the above embodiments of this application, there are multiple possible implementations for determining whether the association condition between newly added dialogue data and dialogue prompt tags is met. The following, in conjunction with... Figure 3 Let's take one implementation method as an example to illustrate.

[0124] like Figure 3 This is another flowchart illustrating the dialogue navigation method provided in this application. The method in this embodiment may include:

[0125] S301, in response to the appearance of new dialogue data in the dialogue interaction interface, obtain at least one dialogue prompt label generated based on historical dialogue data in the dialogue interaction interface.

[0126] S302, for each dialogue prompt label, determine the matching result between the newly added dialogue data and the dialogue summary information of the dialogue prompt label in at least one feature dimension.

[0127] The matching results are used to characterize whether the newly added dialogue data and the dialogue prompt labels are correlated across various feature dimensions.

[0128] The feature dimension can have various possibilities. For example, the feature dimension can include one or more of the following: semantic features, intent features, and entity object features (such as people, project names, or technical terms). It can be set according to actual needs.

[0129] The specific details of the matching results will vary depending on the feature dimensions. For example, the matching result can be determined using at least one of the following possible methods:

[0130] In one possible scenario, it is possible to determine whether there is a semantic correlation between the newly added dialogue data and the dialogue prompt labels based on the first semantic features of the newly added dialogue data and the second semantic features corresponding to the dialogue summary information of the dialogue prompt labels.

[0131] Semantic features are used to characterize the semantic content expressed by dialogue data or dialogue summary information. For example, semantic features can reflect the topics, conclusions, tasks, solution versions, key content, and user-indicated dialogue data relationships expressed or involved in the dialogue data or dialogue summary information.

[0132] The user-indicated dialogue data relationship can be information entered by the user to confirm or deny the relationship between the dialogue data and other dialogue data. For example, if the dialogue data includes "ignore the above content, let's discuss a new topic below", it means that the semantics of the dialogue data indicate that there is no relationship between the dialogue data and other dialogue data.

[0133] In this application, determining the semantic features of dialogue data or dialogue summary information can be done by extracting semantic vectors from the dialogue data or dialogue summary information, or by using a model to analyze and determine the semantic features of the dialogue data or dialogue summary information, without any specific limitation.

[0134] For ease of distinction, this application refers to the semantic features of newly added dialogue data as the first semantic features, and the semantic features corresponding to the dialogue summary information of existing dialogue prompt tags as the second semantic features.

[0135] The semantic relevance between newly added dialogue data and dialogue prompt labels indicates a semantic connection between the first semantic feature of the newly added dialogue data and the second semantic feature corresponding to the dialogue prompt label. For example, the first and second semantic features describe the same topic, conclusion, task, or solution.

[0136] In one example, if the semantic similarity between the first semantic feature of the newly added dialogue data and the second semantic feature corresponding to the dialogue prompt label exceeds a set threshold, it is determined that the newly added dialogue data and the dialogue prompt label have semantic relevance.

[0137] In another possible scenario, the matching result can be determined by: based on the first intent feature corresponding to the newly added dialogue data and the second intent feature corresponding to the dialogue summary information of the dialogue prompt label, determining whether there is a matter-related relationship between the first matter described by the newly added dialogue data and the second matter described by the dialogue summary information of the dialogue prompt label.

[0138] The first intent feature of the newly added dialogue data is a feature determined through intent analysis of the newly added dialogue data. Unlike semantic features, which can only express the shallow semantics of the newly added dialogue data, the first intent feature is determined through deeper analysis of the newly added dialogue data to obtain the specific matters described by the newly added dialogue data.

[0139] Accordingly, the first intent feature of newly added dialogue data can characterize the first matter described by the newly added dialogue data and the matter features of the first matter. The first matter can be a task that the newly added dialogue data needs to perform, the specific steps of the task (e.g., the stage of the task and the purpose of the stage), the items discussed, the objects, or the expected goals. The matter features of a matter (e.g., the first matter or the subsequent second matter) can be one or more of the following: matter type, matter goal, and matter status.

[0140] Similarly, the second intent feature corresponding to the dialogue summary information of the dialogue prompt label is a feature obtained by performing intent analysis on the dialogue summary information. It can be used to characterize the second matter described by the dialogue summary information and the matter features of the second matter. The types of matters that the second matter may involve are similar to those of the first matter, and will not be elaborated further.

[0141] Among them, the relevance between the first item and the second item indicates that the first item and the second item are related items, such as the first item and the second item being the same item, different sub-items of the same main item, items that exchange information with each other, or items that are related, etc., without any specific restrictions.

[0142] In one example, if the similarity between the first intent feature and the second intent feature exceeds a set threshold, it can be determined that the first item and the second item are related.

[0143] S303, based on the matching results corresponding to the dialogue prompt labels, determine whether there is a correlation between the newly added dialogue data and the dialogue prompt labels.

[0144] Specifically, if the dialogue prompt label and the newly added dialogue data are correlated in at least one feature dimension, it is determined that there is a correlation between the newly added dialogue data and the dialogue prompt label.

[0145] In one possible implementation, it can be determined whether the newly added dialogue data and the dialogue prompt label have at least one of semantic relevance and event relevance. For example, if the newly added dialogue data and the dialogue prompt label have at least one of semantic relevance and event relevance, it is determined that the newly added dialogue data and the dialogue prompt label are related.

[0146] S304, in response to the existence of a target dialogue prompt label that is related to the newly added dialogue data in at least one dialogue prompt label, establish an association between the newly added dialogue data and the target dialogue prompt label, and update the dialogue summary information of the target dialogue prompt label based at least on the newly added dialogue data.

[0147] This step can be referred to in the relevant description of the previous embodiments, and will not be repeated here.

[0148] Furthermore, in response to the absence of a target dialogue prompt label associated with the newly added dialogue data in at least one dialogue prompt label, this application can also generate a new dialogue prompt label associated with the newly added dialogue data. For details, please refer to the relevant descriptions in other embodiments, which will not be repeated here.

[0149] In this embodiment, the correlation between new dialogue data and existing dialogue prompt tags is determined through feature matching in at least one dimension. This can identify both explicit topic continuation (semantic relevance) and implicit task logic continuation and other related matters. Based on this, even if a user uses different expressions to describe the same topic or matter in different dialogue data, the system can still accurately identify which existing dialogue prompt tag the new dialogue data should be assigned to through feature matching in at least one dimension. This avoids the system erroneously generating duplicate dialogue prompt tags or incorrectly classifying dialogue data into inappropriate dialogue prompt tags due to differences in expression, thereby improving the accuracy of classification and navigation labeling of multi-turn dialogue data.

[0150] In any of the above embodiments of this application, after updating the target dialog prompt label, this application may also update the display arrangement relationship between the target dialog prompt label and other dialog prompt labels, so as to adjust the display arrangement relationship between the target dialog prompt label and other dialog prompt labels in the dialog interaction interface.

[0151] For example, based on the dialogue summary information corresponding to the updated target dialogue prompt label, a second association relationship between the updated target dialogue prompt label and other dialogue prompt labels is determined; based on the second association relationship, the display arrangement relationship between the updated target dialogue prompt label and the other dialogue prompt labels in the dialogue interaction interface is adjusted.

[0152] The second association relationship may include, but is not limited to, at least one of the following: chronological order of generation, weighting, topic association, and dependency between the target dialog hint tag and other dialog hint tags. For a detailed explanation of the specific meanings of the relationships mentioned here, please refer to the preceding text. Figure 2 The relevant descriptions in the embodiments will not be repeated here.

[0153] As mentioned above, the display arrangement relationship determined here may include, but is not limited to, the arrangement, graph structure, tree structure or other topological structure relationships presented on the dialogue interaction interface. For details, please refer to the previous introduction, which will not be repeated here.

[0154] To facilitate understanding of the specific implementation of adjusting the display arrangement of dialog prompt labels in this application, combined with... Figure 4 Let's take a specific implementation as an example for illustration. For example... Figure 4 This is another flowchart illustrating the dialogue navigation method provided in this application. The method in this embodiment may include:

[0155] S401, in response to the appearance of new dialogue data in the dialogue interaction interface, obtain at least one dialogue prompt label generated based on historical dialogue data in the dialogue interaction interface.

[0156] S402, for each dialogue prompt label, based on the first semantic feature of the newly added dialogue data and the second semantic feature corresponding to the dialogue summary information of the dialogue prompt label, determine whether there is a semantic correlation between the newly added dialogue data and the dialogue prompt label.

[0157] S403, based on the first intent feature corresponding to the newly added dialogue data and the second intent feature corresponding to the dialogue summary information of the dialogue prompt label, determine whether there is a matter-related relationship between the first matter described by the newly added dialogue data and the second matter described by the dialogue summary information of the dialogue prompt label.

[0158] The above steps can be found in the relevant descriptions of the previous embodiments, and will not be repeated here.

[0159] It should be noted that this embodiment is an example of one way to determine whether newly added dialogue data and dialogue prompt labels meet the association conditions. The other implementation methods mentioned above are also applicable to this embodiment.

[0160] S404, in response to the absence of a target dialogue prompt label that is semantically related or related to the new dialogue data among the at least one dialogue prompt label, a new dialogue prompt label associated with the new dialogue data is generated, and the label name and dialogue summary information corresponding to the new dialogue prompt label are determined.

[0161] S405, based at least on the dialogue summary information of the newly added dialogue prompt label and the dialogue summary information of at least one dialogue prompt label, determine a first association relationship between the newly added dialogue prompt label and at least one dialogue prompt label.

[0162] S406, based on the first association relationship, determine the display arrangement relationship between the newly added dialog prompt label and at least one dialog prompt label.

[0163] For example, the first association relationship may include the weight relationship, dependency relationship, and generation time order relationship between the newly added dialog prompt label and each dialog prompt label.

[0164] The display arrangement can be the order between newly added dialog hint labels and at least one dialog hint label, or it can be a tree structure or other topological relationship, etc., without any restrictions.

[0165] S407, based on the display layout, displays at least one dialogue prompt label and a newly added dialogue prompt label in the dialogue interaction interface.

[0166] For example, based on this display layout, the newly added dialogue prompt label and other dialogue prompt labels can be displayed in the navigation area at the top, side, or bottom of the dialogue interaction interface. This navigation area can be a floating window or a portion of the dialogue interaction interface, without any specific restrictions.

[0167] S408, in response to the existence of a target dialogue prompt label in the at least one dialogue prompt label that is semantically related or related to the new dialogue data, establish an association between the new dialogue data and the target dialogue prompt label, and update the label name and dialogue summary information of the target dialogue prompt label based at least on the new dialogue data.

[0168] S409, based on the dialogue summary information corresponding to the updated target dialogue prompt label, determine the second association relationship between the updated target dialogue prompt label and other dialogue prompt labels.

[0169] For example, the second association relationship may include at least one of the following: the weight relationship between the target dialog prompt tag and other dialog prompt tags, the dependency relationship, and the order of generation time.

[0170] In one example, if the second association includes a weighted relationship, determining the second association may include: adjusting the weight information between the target dialog tag and other dialog tags based on the user's operation information on the target dialog tag and other dialog tags.

[0171] As mentioned earlier, this operation information can characterize user actions such as clicking, editing, modifying, and accessing the target dialog hint tag, the dialog data associated with the target dialog hint tag, other dialog hint tags, and the dialog data associated with other dialog hint tags. Specifically, the higher the number or frequency of user operations on the dialog hint tag (e.g., the target dialog hint tag or other dialog hint tags) and its associated dialog data, the higher the importance of the dialog hint tag, and the higher its weight can be.

[0172] Accordingly, based on the weight information and the dialogue summary information corresponding to the updated target dialogue prompt label, a second association relationship between the updated target dialogue prompt label and other dialogue prompt labels is determined.

[0173] For example, based on the updated dialogue summary information of the target dialogue cue tag and the dialogue summary information of other dialogue cue tags, the dependency relationship between the target dialogue cue tag and other dialogue cue tags can be determined. Combining weight information and dependency relationship, the topological relationship between the target dialogue cue tag and other dialogue cue tags can be determined.

[0174] S410, based on the second association, adjust the display arrangement of the updated target dialogue prompt label and other dialogue prompt labels in the dialogue interaction interface.

[0175] For example, the second association can characterize the order of each dialogue prompt label, and can adjust the order of the target dialogue prompt label and other dialogue prompt labels in the dialogue interaction interface.

[0176] For example, if the second relationship represents the topological relationship between the various dialogue prompt labels, then the topological relationship presented in the dialogue interaction interface can be adjusted.

[0177] To facilitate understanding of the scheme in this application, the following is combined with... Figure 5 and Figure 6 Let's illustrate this with a simple example:

[0178] Among them, such as Figure 5 This is an example diagram showing multiple dialogue prompt labels displayed in a dialog interaction interface. Figure 6 This diagram illustrates an example of updating the dialogue prompt labels in the dialogue interface in response to the presence of new dialogue data.

[0179] For ease of understanding, Figure 5 and Figure 6 The example illustrates how China and Israel display dialogue prompts sequentially in the dialogue interaction interface.

[0180] Depend on Figure 5 It can be seen that the top of the dialogue interaction interface displays 5 dialogue prompt labels in the form of a floating layer. The label names of these 5 dialogue prompt labels 501 are: "Transformer Breakthrough", "Early NLP and Word Vectors", "Pre-training and Exploration", "Human Alignment Instructions" and "Multimodal Data Analysis".

[0181] contrast Figure 5 and Figure 6 It can be seen that a new round of dialogue data has been added to the dialogue interaction interface: "What are the future applications of natural language processing?" and "In the future, natural language processing (NLP) is likely to be expanded to other application scenarios." Assuming that this new round of dialogue data is related to the existing dialogue prompt label "early NLP and word vectors," a correlation can be established between the new dialogue data and this dialogue prompt label.

[0182] Because the newly added dialogue data is incorporated into the dialogue prompt label "Early NLP and Word Vectors," this application will also update the label name of this "Early NLP and Word Vectors" dialog prompt label and its display arrangement with other dialog prompt labels. For example... Figure 6 It can be seen that the label name of the dialog prompt "Early NLP and Word Vectors" has been updated to "NLP and Word Vectors". At the same time, the display order of the updated dialog prompt label relative to other dialog prompt labels has changed, with the updated dialog prompt label being placed at the top.

[0183] Of course, while updating the label name of the dialog prompt label "Early NLP and Word Vectors", the dialog summary information associated with the dialog prompt label can also be updated, which will not be elaborated here.

[0184] In this embodiment, when the dialogue summary information of the target dialogue prompt tag is updated, the semantic content carried by the target dialogue prompt tag may change, and its association with other dialogue prompt tags may also change. Based on this, by redetermining the second association between the updated target dialogue prompt tag and other dialogue prompt tags, and adjusting the display arrangement of each dialogue prompt tag in the dialogue interaction interface based on the second association, users can understand the association between each dialogue prompt tag in a timely manner, thereby enabling them to select dialogue prompt tags and locate the dialogue data they need to view more reasonably and efficiently.

[0185] As described in the above embodiments of this application, in these embodiments, the dialogue prompt label can serve as a dynamic interaction anchor for locating dialogue data in the dialogue interaction interface. However, unlike static labels that only mark the topic corresponding to the dialogue data, this dynamic interaction anchor, in addition to associating with the corresponding dialogue data, also records the context boundary (such as the start and end positions of at least one associated historical dialogue) and semantic feature information such as task status information, user intent information, and formation basis corresponding to the associated dialogue data. A dynamic interaction anchor can associate multi-turn dialogue data that is not continuous in time but has a semantic relationship, and organize the multi-turn dialogue data scattered in long dialogues into navigable information units through the association relationship, thereby supporting users to quickly locate historical dialogue content and re-enter the corresponding dialogue context.

[0186] Furthermore, different dynamic interaction anchors can form a relationship network based on at least one of the following: time association, dependency association, reference association, jump association, and topic extension association. The electronic device can dynamically adjust the weight of the anchors and the relationships between them based on user behavior information such as accessing, returning to, staying on, editing, saving, or continuing to ask follow-up questions, thereby adjusting the relationship network. The device then reorganizes the display arrangement of the dynamic interaction anchors in the dialogue interface according to the adjusted relationship network. After a user selects any dynamic interaction anchor, the electronic device can not only display the historical dialogue data associated with that anchor, but also restore the corresponding task status, pending items, and related contextual content, allowing the user to quickly return to the historical discussion state and continue the subsequent interaction process.

[0187] In any of the above embodiments of this application, as the dialogue data in the dialogue interaction interface changes, the appropriate dialogue prompt labels or the number of dialogue prompt labels to be displayed in the dialogue interaction interface will also change. Based on this, in order to display the dialogue prompt labels more reasonably to facilitate user selection and operation, the dialogue summary information corresponding to the target dialogue prompt label or the generation of new dialogue prompt labels can be updated, and the dialogue prompt labels in the dialogue interaction interface can be merged or split.

[0188] The process of merging dialogue prompt labels refers to merging at least two dialogue prompt labels into one merged dialogue prompt label, and establishing an association between the historical dialogue data associated with the at least two dialogue prompt labels and the merged dialogue prompt label.

[0189] The following example illustrates one way to merge dialog prompt labels. For instance... Figure 7 This is a schematic diagram illustrating one implementation process of merging dialog prompt labels in this application. This embodiment may include:

[0190] S701, in response to generating new dialogue prompt tags or updating the dialogue summary information of the target dialogue prompt tag, if the total number of new dialogue prompt tags and at least one dialogue prompt tag exceeds a set number, based on the dialogue summary information corresponding to each dialogue prompt tag, determine at least one dialogue prompt tag group that meets the merging condition.

[0191] This dialog prompt label group includes at least two dialog prompt labels.

[0192] In one example, the dialogue summary information corresponding to each dialogue prompt label in a dialogue prompt label group that meets the merging conditions is related. For example, the dialogue summary information corresponding to each dialogue prompt label in a dialogue prompt label group is semantically related, or the matters described by the dialogue summary information of each dialogue prompt label in a dialogue prompt label are related.

[0193] In this application, the number of dialog prompt tag groups that need to be merged can be determined based on the set quantity. For example, if the difference between the total number of all current dialog prompt tags and the set quantity is relatively small, a small number of dialog prompt tag groups that need to be merged can be determined; if the difference between the total number and the set quantity is large, a relatively large number of dialog prompt tag groups can be determined.

[0194] S702, based on the dialogue summary information of each dialogue prompt label in the dialogue prompt label group, merge each dialogue prompt label in the dialogue prompt label group into one dialogue prompt label to obtain the merged dialogue prompt label.

[0195] Merging each dialogue prompt tag in a dialogue prompt tag group into a single dialogue prompt tag may include: constructing a merged dialogue prompt tag based on at least one of the dialogue summary information and tag name of each dialogue prompt tag in the dialogue prompt tag group, and determining the tag name of the merged dialogue prompt tag.

[0196] Of course, while or after merging the merged dialog hint tags, the dialog summary information of the merged dialog hint tags can be determined by combining the dialog summary information or historical dialogue data corresponding to each dialog hint tag in the dialog hint tag group.

[0197] In this application, splitting at least one dialog prompt label can be performed by splitting one or more separable dialog prompt labels from the at least one dialog prompt label. For ease of distinction, the dialog prompt labels split from the dialog prompt label are called sub-dialog prompt labels.

[0198] The essence of splitting dialogue prompt labels is to divide the historical dialogue data associated with the dialog prompt label to be split into at least two groups of historical dialogue data, and then reconstruct sub-dialogue prompt labels for each group of historical dialogue data. The following explanation uses one implementation method as an example. Figure 8 This is a schematic diagram illustrating one implementation process of splitting dialog prompt labels in this application. This embodiment may include:

[0199] S801, in response to generating dialogue summary information for adding a new dialogue prompt label or updating the target dialogue prompt label, a splittable candidate dialogue prompt label is determined from the at least one dialogue prompt label based on at least one historical dialogue data associated with the dialogue prompt label.

[0200] Among them, the splittable candidate dialogue prompt labels can be dialogue prompt labels that meet the splitting conditions, such as the splitting condition being that the number of historical dialogue data associated with the dialogue prompt label exceeds a first quantity threshold.

[0201] The splittable candidate dialog prompt label can also be a dialog prompt label that needs to be split, as indicated by the user through a command.

[0202] As an optional approach, when generating new dialogue hint tags or updating the dialogue summary information of the target dialogue hint tag, if a tag splitting instruction is obtained or the tag splitting condition is met, the operation of determining the candidate dialogue hint tags that can be split is performed.

[0203] The tag splitting instruction can be a user-input command used to split dialog prompt tags. This instruction can either indicate splittable candidate dialog prompt tags, or it can choose not to indicate the candidate dialog prompt tags to be split, allowing the electronic device to determine the candidate dialog prompt tags that meet the splitting criteria.

[0204] The tag splitting condition can be that the number of at least one dialogue prompt tag in the dialogue interaction interface is less than a second quantity threshold, or other conditions can be set as needed, without any specific restrictions.

[0205] S802, for each candidate dialogue prompt label, based on at least one candidate historical dialogue data associated with the candidate dialogue prompt label, the candidate dialogue prompt label is split into at least two sub-prompt labels.

[0206] Each sub-hint label is associated with at least one portion of candidate historical dialogue data. For example, the correlation between candidate historical dialogue data associated with the same sub-hint label is stronger than that between candidate historical dialogue data associated with other sub-hint labels.

[0207] Specifically, while extracting sub-hint tags, the tag name and corresponding dialogue summary information of each sub-hint tag can be determined based on the candidate historical dialogue data associated with that sub-hint tag. The tag name of the sub-hint tag is used to indicate the topic of each candidate historical dialogue data associated with that sub-hint tag.

[0208] S803, display the at least two sub-prompt labels in the dialogue interaction interface.

[0209] In this application, when the candidate dialogue prompt label is split into at least two sub-prompt labels, only the split at least two sub-prompt labels can be displayed in the dialogue interaction interface, and the candidate dialogue prompt label can be omitted.

[0210] Furthermore, to enable users to more intuitively understand the relationships between the various sub-hint labels, the candidate dialog hint label and its sub-hint labels can be displayed simultaneously. For example, based on the association between the candidate dialog hint label and its sub-hint labels, the topological relationship between the candidate dialog hint label and its sub-hint labels can be shown.

[0211] Of course, the dialog interaction interface can also display the arrangement order or topological relationship of the sub-hint label with other dialog hint labels, etc. For details, please refer to the previous introduction, which will not be repeated here.

[0212] In the above embodiments of this application, users can actively edit the dialogue prompt labels in the dialogue interaction interface, such as renaming the dialogue prompt labels, deleting the dialogue prompt labels, triggering the merging of dialogue prompt labels, and modifying the order or other display arrangement of the dialogue prompt labels.

[0213] For example, when a user clicks on or otherwise triggers an action in the navigation area used to display dialog prompt labels, the navigation area enters a label editing state. Figure 9 As shown, after the navigation area enters edit mode, it can add a "Refactor" option to trigger the reconstruction of dialog hint labels, as well as other editing options; it can also select dialog hint labels, such as selecting... Figure 9 The section on "NPU and word vectors" is used to rename the dialog prompt labels, etc., but details will not be elaborated further.

[0214] Furthermore, this application also provides an electronic device. For example... Figure 10 As shown, it illustrates a schematic diagram of the composition structure of the electronic device, which includes at least a display module 1001 and a processor 1002;

[0215] The display module 1001 is used to display the interactive dialogue interface;

[0216] The processor 1002 is configured to, in response to the appearance of new dialogue data in the dialogue interaction interface, obtain at least one dialogue prompt label generated based on historical dialogue data in the dialogue interaction interface; determine, based on the dialogue summary information of at least one of the dialogue prompt labels, whether the new dialogue data and at least one of the dialogue prompt labels satisfy an association condition; and, in response to the existence of a target dialogue prompt label that satisfies the association condition among at least one of the dialogue prompt labels, establish an association relationship between the new dialogue data and the target dialogue prompt label, and update the dialogue summary information of the target dialogue prompt label based at least on the new dialogue data.

[0217] The specific operations performed by the processor can be found in the relevant operations of the previous embodiments, and will not be repeated here.

[0218] Understandably, the electronic device may also include a memory 1003 for storing programs required by the processor to perform operations.

[0219] Furthermore, the electronic device may also include an input unit 1004 for inputting dialogue data.

[0220] Of course, the electronic device can also have more than Figure 10 There are no restrictions on the number of more or fewer components.

[0221] This application also provides a computer program product, including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement any of the dialogue navigation methods provided in this application.

[0222] This application also provides a computer-readable storage medium that carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any of the dialogue navigation methods provided in this application.

[0223] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0224] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence or the part that has been contributed, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0225] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0226] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

Claims

1. A dialogue navigation method, comprising: In response to the appearance of new dialogue data in the dialogue interaction interface, at least one dialogue prompt label generated based on historical dialogue data in the dialogue interaction interface is obtained. Based on the dialogue summary information of at least one of the dialogue prompt tags, determine whether the newly added dialogue data and at least one of the dialogue prompt tags meet the association conditions. In response to the existence of a target dialogue prompt tag that meets the association conditions in at least one of the dialogue prompt tags, an association relationship is established between the newly added dialogue data and the target dialogue prompt tag, and the dialogue summary information of the target dialogue prompt tag is updated based at least on the newly added dialogue data.

2. The dialogue navigation method according to claim 1 further includes: In response to the absence of a target dialogue prompt tag that meets the association conditions in at least one of the dialogue prompt tags, a new dialogue prompt tag associated with the new dialogue data is generated, and the dialogue summary information corresponding to the new dialogue prompt tag is determined; The newly added dialogue prompt label is displayed on the dialogue interaction interface.

3. The dialogue navigation method according to claim 2 further includes: Based at least on the dialogue summary information of the newly added dialogue prompt tag and the dialogue summary information of at least one of the dialogue prompt tags, a first association relationship between the newly added dialogue prompt tag and at least one of the dialogue prompt tags is determined; The newly added dialogue prompt label is displayed on the dialogue interaction interface, including: Based on the first association relationship, determine the display arrangement relationship between the newly added dialogue prompt label and at least one of the dialogue prompt labels; Based on the display layout, at least one of the dialogue prompt labels and the newly added dialogue prompt label are displayed in the dialogue interaction interface.

4. The dialogue navigation method according to claim 1, wherein determining whether the newly added dialogue data and at least one of the dialogue prompt tags satisfy an association condition based on the dialogue summary information of at least one of the dialogue prompt tags includes: Determine the matching result between the newly added dialogue data and the dialogue summary information of the dialogue prompt label on at least one feature dimension, wherein the matching result is used to characterize whether the newly added dialogue data and the dialogue prompt label are related on the feature dimension; Based on the matching results corresponding to the dialogue prompt tags, it is determined whether there is a correlation between the newly added dialogue data and the dialogue prompt tags.

5. The dialogue navigation method according to claim 4, wherein determining the matching result of the newly added dialogue data and the dialogue summary information of the dialogue prompt label in at least one feature dimension includes at least one of the following: Based on the first semantic feature of the newly added dialogue data and the second semantic feature corresponding to the dialogue summary information of the dialogue prompt tag, it is determined whether there is a semantic correlation between the newly added dialogue data and the dialogue prompt tag; Based on the first intent feature corresponding to the newly added dialogue data and the second intent feature corresponding to the dialogue summary information of the dialogue prompt label, it is determined whether there is a matter correlation between the first matter described by the newly added dialogue data and the second matter described by the dialogue summary information of the dialogue prompt label. The step of determining whether there is a correlation between the newly added dialogue data and the dialogue prompt tag based on the matching result corresponding to the dialogue prompt tag includes: Determine whether the newly added dialogue data and the dialogue prompt label have at least one of the semantic relevance and the event association.

6. The dialogue navigation method according to claim 1 or 2, further comprising at least one of the following: In response to the existence of a target dialogue prompt tag that meets the association conditions in at least one of the dialogue prompt tags, the tag name of the target dialogue prompt tag is updated based on the newly added dialogue data; In response to updating the dialogue summary information corresponding to the target dialogue prompt label or generating the new dialogue prompt label, the dialogue prompt labels in the dialogue interaction interface are merged or split.

7. The dialogue navigation method according to claim 1, further comprising: Based on the dialogue summary information corresponding to the updated target dialogue prompt tag, a second association relationship between the updated target dialogue prompt tag and other dialogue prompt tags is determined; Based on the second association relationship, the display arrangement relationship between the updated target dialogue prompt label and the other dialogue prompt labels in the dialogue interaction interface is adjusted.

8. The dialogue navigation method according to claim 7, wherein determining the second association relationship between the updated target dialogue prompt tag and other dialogue prompt tags based on the dialogue summary information corresponding to the updated target dialogue prompt tag includes: Based on the user's operation information on the target dialogue prompt tag and other dialogue prompt tags, adjust the weight information between the target dialogue prompt tag and other dialogue prompt tags; Based on the weight information and the dialogue summary information corresponding to the updated target dialogue prompt label, a second association relationship between the updated target dialogue prompt label and other dialogue prompt labels is determined.

9. The dialogue navigation method according to claim 1, further comprising at least one of the following: In response to the user's first action on the dialog prompt label, display the dialog summary information corresponding to the dialog prompt label; In response to a second action by the user on the dialog prompt label, at least one piece of historical dialogue data associated with the dialog prompt label is displayed.

10. An electronic device, comprising: Display module and processor; The display module is used to display the interactive dialogue interface; The processor is configured to: respond to the appearance of new dialogue data in the dialogue interaction interface; obtain at least one dialogue prompt tag generated based on historical dialogue data in the dialogue interaction interface; determine whether the new dialogue data and at least one dialogue prompt tag satisfy an association condition based on the dialogue summary information of the at least one dialogue prompt tag; and, in response to the existence of a target dialogue prompt tag that satisfies the association condition among the at least one dialogue prompt tag, establish an association relationship between the new dialogue data and the target dialogue prompt tag, and update the dialogue summary information of the target dialogue prompt tag based at least on the new dialogue data.

Images