Method and apparatus for identifying repeated dialogue, electronic device, and readable storage medium

Abstract

The application provides a repeated dialogue identification method and device, electronic equipment and readable storage medium. In the process of artificial intelligence dialogue broadcasting first reply content, in response to a second consultation dialogue proposed by a user, whether the second consultation dialogue is an invalid consultation dialogue is judged according to the intention and / or second entity information of the second consultation dialogue. When the interval between the proposing time of the first consultation dialogue and the proposing time of the second consultation dialogue is greater than a preset time threshold, the first reply content is repeatedly broadcasted. When the interval is less than or equal to the preset interval threshold, the second consultation dialogue is ignored and the first reply content is continuously broadcasted. In this way, the user can avoid interrupting the voice interaction process by using repeated consultation dialogue in the case of unintentional interruption, and the response frequency of the artificial intelligence dialogue robot in the interaction process is reduced.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, electronic device, and readable storage medium for identifying repetitive dialogues. Background Technology

[0002] With the continuous development of artificial intelligence technology, the application of AI chatbots has become increasingly widespread, penetrating into all aspects of life. Consequently, people's expectations and demands for AI chatbots are also rising; they hope these chatbots can answer a wider range of questions and handle more complex tasks. However, due to the inherent delay in voice interaction, users often repeat the same questions or express previously stated content when they don't receive timely feedback. This causes the chatbot to be interrupted at the beginning of its broadcast by repeated questions, even though the content broadcast after the interruption is essentially the same as the previous broadcast. This delays and increases the time required for users to complete their business transactions. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide a method, apparatus, electronic device and readable storage medium for identifying repeated dialogues, which can determine whether a second consultation dialogue proposed by a user is invalid during the process of responding to a first consultation dialogue proposed by the user, so as to avoid the interference of invalid consultation dialogues on the voice interaction process, improve business processing efficiency and reduce business processing time.

[0004] This application provides a method for identifying repeated dialogues, the method comprising:

[0005] During the process of broadcasting the first response to the first consultation dialogue proposed by the user, in response to the second consultation dialogue proposed by the user, the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue are determined.

[0006] Based on the dialogue intent and / or the second entity information, determine whether the second consultation dialogue is an invalid consultation dialogue; wherein, the invalid consultation dialogue includes duplicate consultation dialogues and abnormal consultation dialogues;

[0007] When the second consultation dialogue is a repeated consultation dialogue, the time interval between the initiation of the second consultation dialogue and the first consultation dialogue is determined;

[0008] When the time interval between the requests is less than or equal to a preset time threshold, the second consultation dialogue is ignored, and the first response content is continued to be broadcast.

[0009] When the time interval between requests exceeds a preset threshold, the first response is replayed.

[0010] In one possible implementation, the identification method further includes:

[0011] When the second consultation dialogue is an abnormal consultation dialogue, ignore the second consultation dialogue and continue to broadcast the first response content.

[0012] In one possible implementation, the dialogue intent includes a second business intent, and determining whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent includes:

[0013] Based on the probability score of the second consultation dialogue having the second business intention, it is determined whether the second business intention can represent the true business intention of the second consultation dialogue; wherein, the second business intention is the candidate business intention with the highest probability score among the second consultation dialogues;

[0014] If not, the second consultation dialogue is determined to be an abnormal consultation dialogue;

[0015] If so, determine whether the second business intent is the same as the first business intent of the first consultation dialogue;

[0016] When the second business intent is the same as the first business intent, the second consultation dialogue is determined to be a repeated consultation dialogue;

[0017] When the second business intent is different from the first business intent, the second consultation dialogue is determined to be a valid consultation dialogue.

[0018] In one possible implementation, determining whether the second business intent can characterize the true business intent of the second consultation dialogue based on the probability score of the second consultation dialogue having the second business intent includes:

[0019] If the probability score of the second consultation dialogue having the second business intention is greater than or equal to the preset score threshold, then it is determined that the second business intention can represent the true business intention of the second consultation dialogue.

[0020] or,

[0021] If the probability score of the second consultation dialogue having the second business intention is less than the preset score threshold, and the difference between the second and the second high probability score is greater than or equal to the preset difference threshold, then it is determined that the second business intention can characterize the true intention of the second consultation dialogue.

[0022] In one possible implementation, the dialogue intent includes an entity intent, and determining whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent and the second entity information includes:

[0023] If the dialogue intent is an entity intent and the second entity information indicates that no new entity conditions have been extracted from the second consultation dialogue, the second consultation dialogue is determined to be an abnormal consultation dialogue.

[0024] If the dialogue intent is an entity intent and the second entity information includes newly added entity conditions extracted from the second consultation dialogue, then based on the newly added entity conditions, it is determined whether the second consultation dialogue is an invalid consultation dialogue.

[0025] In one possible implementation, determining whether the second consultation dialogue is invalid based on the newly added entity condition includes:

[0026] Determine whether the newly added entity condition is related to the first business intent of the first consultation dialogue;

[0027] If not, the second consultation dialogue is determined to be an abnormal consultation dialogue;

[0028] If so, using the newly added entity conditions, by searching the business knowledge graph, determine the content of the second response to the second consultation dialogue, and determine whether the content of the second response is consistent with the content of the first response;

[0029] If they match, the second consultation dialogue is determined to be a duplicate consultation dialogue;

[0030] If there is no discrepancy, the second consultation dialogue is determined to be a valid consultation dialogue.

[0031] In one possible implementation, the step of responding to the second consultation dialogue initiated by the user, determining the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue, includes:

[0032] In response to the second consultation dialogue proposed by the user, a natural language understanding model is used to determine the probability score between the second consultation dialogue and each candidate intent, and the natural language understanding model is used to extract the second entity information carried by the second consultation dialogue;

[0033] The candidate intent with the highest probability score is determined as the dialogue intent of the second consultation dialogue.

[0034] This application embodiment also provides a device for identifying repeated dialogues, the device comprising:

[0035] The intent determination module is used to determine the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue in response to the second consultation dialogue proposed by the user during the process of broadcasting the first response content of the first consultation dialogue proposed by the user.

[0036] The judgment module is used to determine whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent and / or the second entity information; wherein, the invalid consultation dialogue includes duplicate consultation dialogue and abnormal consultation dialogue;

[0037] An interval comparison module is used to determine the time interval between the second consultation dialogue and the first consultation dialogue when the second consultation dialogue is a repeated consultation dialogue.

[0038] The first continuous broadcasting module is used to ignore the second consultation dialogue and continue broadcasting the first reply content when the time interval between the requests is less than or equal to a preset time interval threshold.

[0039] The rebroadcast module is used to rebroadcast the first response content when the time interval between the requests exceeds a preset time interval threshold.

[0040] In one possible implementation, the identification device further includes a second continuous broadcast module, the second continuous broadcast module being used for:

[0041] When the second consultation dialogue is an abnormal consultation dialogue, ignore the second consultation dialogue and continue to broadcast the first response content.

[0042] In one possible implementation, the dialogue intent includes a second business intent, and when the judgment module determines whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent, the judgment module is configured to:

[0043] Based on the probability score of the second consultation dialogue having the second business intention, it is determined whether the second business intention can represent the true business intention of the second consultation dialogue; wherein, the second business intention is the candidate business intention with the highest probability score among the second consultation dialogues;

[0044] If not, the second consultation dialogue is determined to be an abnormal consultation dialogue;

[0045] If so, determine whether the second business intent is the same as the first business intent of the first consultation dialogue;

[0046] When the second business intent is the same as the first business intent, the second consultation dialogue is determined to be a repeated consultation dialogue;

[0047] When the second business intent is different from the first business intent, the second consultation dialogue is determined to be a valid consultation dialogue.

[0048] In one possible implementation, when the judging module determines whether the second business intent can represent the true business intent of the second consultation dialogue based on the probability score of the second consultation dialogue having the second business intent, the judging module is used to:

[0049] If the probability score of the second consultation dialogue having the second business intention is greater than or equal to the preset score threshold, then it is determined that the second business intention can represent the true business intention of the second consultation dialogue.

[0050] or,

[0051] If the probability score of the second consultation dialogue having the second business intention is less than the preset score threshold, and the difference between the second and the second high probability score is greater than or equal to the preset difference threshold, then it is determined that the second business intention can characterize the true intention of the second consultation dialogue.

[0052] In one possible implementation, the dialogue intent includes an entity intent, and when the determining module determines whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent and the second entity information, the determining module is configured to:

[0053] If the dialogue intent is an entity intent and the second entity information indicates that no new entity conditions have been extracted from the second consultation dialogue, the second consultation dialogue is determined to be an abnormal consultation dialogue.

[0054] If the dialogue intent is an entity intent and the second entity information includes newly added entity conditions extracted from the second consultation dialogue, then based on the newly added entity conditions, it is determined whether the second consultation dialogue is an invalid consultation dialogue.

[0055] In one possible implementation, when the determining module is used to determine whether the second consultation dialogue is invalid based on the newly added entity condition, the determining module is used to:

[0056] Determine whether the newly added entity condition is related to the first business intent of the first consultation dialogue;

[0057] If not, the second consultation dialogue is determined to be an abnormal consultation dialogue;

[0058] If so, using the newly added entity conditions, by searching the business knowledge graph, determine the content of the second response to the second consultation dialogue, and determine whether the content of the second response is consistent with the content of the first response;

[0059] If they match, the second consultation dialogue is determined to be a duplicate consultation dialogue;

[0060] If there is no discrepancy, the second consultation dialogue is determined to be a valid consultation dialogue.

[0061] In one possible implementation, when the intent determination module determines the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue in response to the user's second consultation dialogue, the intent determination module is configured to:

[0062] In response to the second consultation dialogue proposed by the user, a natural language understanding model is used to determine the probability score between the second consultation dialogue and each candidate intent, and the natural language understanding model is used to extract the second entity information carried by the second consultation dialogue;

[0063] The candidate intent with the highest probability score is determined as the dialogue intent of the second consultation dialogue.

[0064] This application also provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the repeated dialogue identification method described above are performed.

[0065] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the repetitive dialogue identification method described above.

[0066] The method, apparatus, electronic device, and readable storage medium for identifying repeated dialogues provided in this application, during the process of broadcasting the first response content of a first consultation dialogue initiated by a user, responds to a second consultation dialogue initiated by the user, determines the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue; determines whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent and / or the second entity information; when the second consultation dialogue is a repeated consultation dialogue, determines the time interval between the initiation of the second consultation dialogue and the first consultation dialogue; when the initiation time interval is less than or equal to a preset interval threshold, ignores the second consultation dialogue and continues to broadcast the first response content; when the initiation time interval is greater than the preset interval threshold, rebroadcasts the first response content. In this way, it can prevent users from unintentionally interrupting the voice interaction process with repeated consultation dialogues, reduce the response frequency of the AI ​​dialogue robot during the interaction process, and improve the efficiency of human-computer interaction.

[0067] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0068] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0069] Figure 1 A flowchart illustrating a method for identifying repeated dialogues provided in an embodiment of this application;

[0070] Figure 2 This is a schematic diagram of a dialogue interaction process provided in an embodiment of this application;

[0071] Figure 3 One of the structural schematic diagrams of a repeating dialogue recognition device provided in an embodiment of this application;

[0072] Figure 4 A second schematic diagram of the structure of a repeating dialogue recognition device provided in an embodiment of this application;

[0073] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0075] Research has found that due to the inherent delay in voice interaction, users often repeat the same questions or express previously stated content when they do not receive timely feedback from the AI ​​chatbot. This causes the AI ​​chatbot to be interrupted at the beginning of its broadcast by repeated questions, even though the content broadcast after the interruption is essentially the same as the previous broadcast. Consequently, this delays users' business processing time and increases the overall processing time.

[0076] Based on this, the embodiments of this application provide a method for identifying repeated dialogues, which can prevent invalid consultation dialogues from interrupting the current interaction process, improve the interaction efficiency between users and artificial intelligence chatbots, and thus achieve the goal of improving business processing efficiency.

[0077] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for identifying repetitive dialogues provided in an embodiment of this application. Figure 1 As shown in the embodiments of this application, the method for identifying repeated dialogues includes:

[0078] S101. During the process of broadcasting the first response to the first consultation dialogue proposed by the user, in response to the second consultation dialogue proposed by the user, determine the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue.

[0079] S102. Determine whether the second consultation dialogue is invalid based on the dialogue intent and / or the second entity information.

[0080] S103. When the second consultation dialogue is a repeated consultation dialogue, determine the time interval between the initiation of the second consultation dialogue and the first consultation dialogue.

[0081] S104. When the time interval is less than or equal to a preset time interval threshold, ignore the second consultation dialogue and continue to broadcast the first reply content.

[0082] S105. When the time interval between the proposed responses exceeds a preset time interval threshold, the first response content is replayed.

[0083] This application provides a method for identifying repeated dialogues. During the process of an AI-powered dialogue system broadcasting the first response to a user's first inquiry, if the user initiates a second inquiry, the method determines the dialogue intent and / or the second entity information carried by the second inquiry. To prevent invalid inquiries from interfering with the user's voice interaction with the AI ​​dialogue robot, the method judges whether the second inquiry is invalid based on the determined dialogue intent and / or second entity information. If the second inquiry is valid, the method broadcasts the second response to the user. If the second inquiry is a repeated inquiry, the method considers the time the user initiated the second inquiry to determine whether the first response needs to be broadcast again. If the time interval between inquiries exceeds a preset time threshold, the first response is broadcast again to ensure the user receives timely information. If the time interval is less than or equal to the preset time interval threshold, the second inquiry is ignored, and the first response continues to be broadcast. This avoids unintentional interruptions by users using repeated inquiries to disrupt the voice interaction, reduces the response frequency of the AI ​​dialogue robot during the interaction, and improves human-computer interaction efficiency.

[0084] Users can consult or complete business procedures by engaging in dialogue with an AI-powered chatbot. In other words, users can interact with the AI ​​chatbot to consult or complete business procedures.

[0085] Because AI chatbots require a certain response time to answer a user's initial inquiry, there is a delay in the voice interaction process. Often, users who don't receive timely feedback may feel they haven't expressed themselves clearly. Therefore, they might ask the same question in a second inquiry (which could be the same dialogue or a different expression) just as the chatbot is preparing or beginning to read its response. However, in these cases, the chatbot needs to provide the same answer; repeated responses only increase the chatbot's response frequency and slow down processing. Therefore, upon receiving a second inquiry, it's necessary to determine whether it's invalid based on the intent of the second inquiry and / or the second entity information it carries.

[0086] In step S101, during the process of broadcasting the first response to the first consultation dialogue proposed by the user, in response to the second consultation dialogue proposed by the user, the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue are determined by using a natural language understanding model.

[0087] The dialogue intent includes business intent and entity intent in a specific scenario. When the dialogue intent of the second consultation dialogue is an entity intent, it indicates that the second consultation dialogue represents a certain entity condition. In this case, it can also be said that the user wants to supplement the entity condition carried by the first consultation dialogue through the second consultation dialogue.

[0088] Specific scenarios may include one or more of the following: banking services, community services, and hospital visits.

[0089] When the specific scenario is a banking business transaction scenario, the dialogue intent can represent the services provided by the bank. Specifically, the dialogue intent may include one or more of the following: deposit_processing_processing procedure, deposit_agency_processing materials, withdrawal_processing_processing procedure, withdrawal_agency_processing materials, loan_processing_processing procedure, and loan_agency_processing materials.

[0090] When the specific scenario is a community business processing scenario, the dialogue intent can represent the services provided by the community. Specifically, the dialogue intent may include one or more of the following: unemployment registration - processing - processing procedure, unemployment registration - agency - processing materials, residence permit - processing - processing procedure, residence permit agency - processing materials, employment and entrepreneurship certificate - application - application procedure, and employment and entrepreneurship certificate - agency - processing materials.

[0091] Natural Language Processing (NLU) technology refers to the technology of communicating between natural language and computers, also known as computational linguistics. On one hand, it is a branch of language information processing; on the other hand, it is one of the core topics of artificial intelligence (AI).

[0092] A Natural Language Understanding (NLU) model is a trained model that can recognize the intent indicated by natural language. It is mainly used to identify the intent corresponding to natural language. For example, the BERT model can be used. Here, multiple sample consultation dialogues and the intent label corresponding to each sample consultation dialogue can be used to train the BERT model to obtain the NLU model.

[0093] NLU models can be built for specific business scenarios, such as banking, community services, and hospital visits. A pre-trained NLU model can identify entity information (i.e., entity words) carried in the consultation dialogue (the expressions used by users when inquiring about relevant business issues) to determine various business intentions related to the business scenario. For example, in a community service scenario, business intentions include: "unemployment registration process," "residence permit application process," and "employment and entrepreneurship certificate application materials." Furthermore, the pre-trained NLU model can extract entity information carried in the consultation dialogue during the identification process.

[0094] Here, the first response includes the result of the first consultation dialogue or the guiding dialogue generated based on the first consultation dialogue.

[0095] In one implementation, step S101 includes: in response to the second consultation dialogue proposed by the user, using a natural language understanding model to determine the probability score between the second consultation dialogue and each candidate intent, and using the natural language understanding model to extract the second entity information carried by the second consultation dialogue; and determining the candidate intent with the highest probability score as the dialogue intent of the second consultation dialogue.

[0096] In this step, after obtaining the second consultation dialogue proposed by the user, the second consultation dialogue is input into a pre-trained natural language understanding model. The natural language understanding model is used to determine the probability score of the second consultation dialogue having any candidate intent. Here, the probability score represents the likelihood that the second consultation dialogue is any candidate intent; the higher the probability score, the higher the probability that the second consultation dialogue is that candidate intent. Therefore, the candidate intent with the highest probability score among multiple candidate intents can be determined as the dialogue intent possessed by the second consultation dialogue. The range of the probability score is (0, 1).

[0097] Simultaneously, a natural language understanding model is used to extract second entity information from the second consultation dialogue; here, the second entity information includes condition labels that can characterize whether the second consultation dialogue carries entity conditions and / or entity conditions carried in the second consultation dialogue.

[0098] In step S102, after determining the dialogue intent and / or second entity information of the second consultation dialogue, it is possible to determine whether the second consultation dialogue is an invalid consultation dialogue by combining the dialogue intent and / or second entity information of the second consultation dialogue.

[0099] Here, invalid consultation dialogue can refer to a repeated consultation dialogue submitted by a user when no response is received; it can also refer to an abnormal consultation dialogue submitted by a user where no response can be found; that is, invalid consultation dialogue includes both repeated consultation dialogue and abnormal consultation dialogue.

[0100] In one implementation, the dialogue intent includes a second business intent, and determining whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent includes:

[0101] Step 1: Based on the probability score of the second consultation dialogue having the second business intention, determine whether the second business intention can represent the true business intention of the second consultation dialogue.

[0102] Here, the second business intent is the candidate business intent with the highest probability score in relation to the second consultation dialogue; however, even the candidate business intent with the highest probability score may not necessarily accurately represent the true business intent indicated by the second consultation dialogue. That is, there may be cases where the determined second business intent cannot accurately represent the true business intent of the second consultation dialogue. Therefore, in order to further increase the accuracy of business intent identification, it is also necessary to judge whether the second business intent can represent the true business intent of the second consultation dialogue.

[0103] In this step, the likelihood score of the second consultation dialogue having a second business intention can be used to determine whether the second business intention can represent the true business intention of the second consultation dialogue.

[0104] Specifically, step 1 includes: if the probability score of the second consultation dialogue having the second business intention is greater than or equal to a preset score threshold, then it is determined that the second business intention can represent the true business intention of the second consultation dialogue.

[0105] In this step, the probability score of the second consultation dialogue having a second business intention is compared with a preset score threshold. If the probability score of having a second business intention is greater than or equal to the preset score threshold, it is determined that the second business intention can represent the true business intention of the second consultation dialogue; otherwise, it is determined that the second business intention cannot represent the true business intention of the second consultation dialogue.

[0106] Alternatively, if the probability score of the second consultation dialogue having the second business intention is less than the preset score threshold, and the difference between the second and the second high probability score is greater than or equal to the preset difference threshold, then it is determined that the second business intention can characterize the true intention of the second consultation dialogue.

[0107] In this step, if the probability score of the second consultation dialogue having a second business intention is less than a preset score threshold, the difference between the probability score of the second consultation dialogue having a second business intention and the second highest probability score is compared with a preset difference threshold. If the difference between the two is greater than or equal to the preset difference threshold, it is determined that the second business intention can represent the true business intention of the second consultation dialogue; otherwise, it is determined that the second business intention cannot represent the true business intention of the second consultation dialogue.

[0108] The preset score threshold can be a fixed constant, such as 0.7; similarly, the preset difference threshold can also be a fixed constant, such as 0.25. The sources of the preset score threshold and preset difference threshold are as follows: After training the NLU intent model, it is tested using a large number of test sets (the test sets contain test consultation dialogues and test intent labels corresponding to each test consultation dialogue). The test results mainly include the scores corresponding to the highest-scoring business intent and the second-highest-scoring business intent matched by the test consultation dialogue in the NLU model, as well as the difference between the two. The highest-scoring business intent being the same as the intent indicated by the test intent label is considered correctly identified; otherwise, it is considered incorrectly identified.

[0109] Statistical results show that when the highest business intent score is greater than or equal to 0.7, the credibility of the intent result is relatively high and can be considered a credible result.

[0110] When the highest business intent score is less than 0.7, further calculation is performed on the difference between the highest and second-highest business intent scores in each training sample group. The calculated differences for each group are statistically analyzed and categorized to determine an effective difference threshold. There are two categorization methods: one is based on total values, such as [0,1), [0.1,1), [0.2,1); the other is based on intervals, such as [0,0.1), [0.1,0.2), [0.2,0.3).

[0111] Statistical analysis revealed that when the difference between the highest-scoring business intent and the second-highest-scoring intent score is between 0.2 and 0.3, the number of misidentifications of unreliable results as reliable results in a test sample can be controlled to within 3, while the number of correctly identified results is greater than or equal to 42. This demonstrates that the positive benefits far outweigh the negative impacts for a test sample, and the negative impacts can be kept within an acceptable range. Therefore, when the highest-scoring intent score is less than 0.7, or when the difference between the highest-scoring and second-highest-scoring business intent scores is greater than or equal to 0.25, the highest-scoring business intent is considered a reliable result.

[0112] Step 2: If not, determine that the second consultation dialogue is an abnormal consultation dialogue.

[0113] In this step, if it is determined that the second business intent cannot represent the true business intent of the second consultation dialogue, then the second consultation dialogue is determined to be an abnormal consultation dialogue.

[0114] Step 3: If yes, determine whether the second business intent is the same as the first business intent of the first consultation dialogue; when the second business intent is the same as the first business intent, determine that the second consultation dialogue is a duplicate consultation dialogue; when the second business intent is not the same as the first business intent, determine that the second consultation dialogue is a valid consultation dialogue.

[0115] In this step, if the second business intent can represent the true business intent of the second consultation dialogue, it is also necessary to determine whether the second business intent is the same as the first business intent of the first consultation dialogue. If the second business intent is the same as the first business intent, it means that the second response content of the second consultation dialogue is the same as the first response content. At this time, the second consultation dialogue can be considered as a repeat consultation dialogue of the first consultation dialogue.

[0116] If the second business intent is different from the first business intent, it means that the second response in the second consultation dialogue is different from the first response, that is, the user has raised a new consultation question. In this case, the second consultation dialogue can be considered a valid consultation dialogue, and a response should be given to the second consultation dialogue.

[0117] In another implementation, the dialogue intent includes an entity intent, and determining whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent and the second entity information includes:

[0118] Step 4: If the dialogue intent is an entity intent and the second entity information indicates that no new entity conditions have been extracted from the second consultation dialogue, then the second consultation dialogue is determined to be an abnormal consultation dialogue.

[0119] In this step, when the dialogue intent of the second consultation dialogue is determined to be an entity intent through the natural language understanding model, that is, the second consultation dialogue is used to supplement the entity conditions carried by the first consultation dialogue, if the second entity information indicates that no new entity conditions are extracted from the second consultation dialogue, it means that the second consultation dialogue proposed by the user cannot supplement the entity conditions of the first consultation dialogue. At this time, the second consultation dialogue can be determined to be an abnormal consultation dialogue.

[0120] Step 5: If the dialogue intent is an entity intent and the second entity information includes newly added entity conditions extracted from the second consultation dialogue, determine whether the second consultation dialogue is an invalid consultation dialogue based on the newly added entity conditions.

[0121] In this step, when the dialogue intent of the second consultation dialogue is determined to be an entity intent through the natural language understanding model, that is, the second consultation dialogue is used to supplement the entity conditions carried by the first consultation dialogue, and the second entity information includes the newly added entity conditions extracted from the second consultation dialogue; then, based on the extracted newly added entity conditions, it is further determined whether the second consultation dialogue is an invalid consultation dialogue.

[0122] Specifically, determining whether the second consultation dialogue is invalid based on the newly added entity conditions includes:

[0123] Step a: Determine whether the newly added entity condition is related to the first business intent of the first consultation dialogue.

[0124] In this step, it is determined whether the newly added entity condition is related to the first business intent of the first consultation dialogue, that is, whether the newly added entity condition can serve as an entity condition that triggers the first business intent.

[0125] Step b: If not, determine that the second consultation dialogue is an abnormal consultation dialogue.

[0126] In this step, if the newly added entity condition is not related to the first business intent, that is, the newly added entity condition carried by the second consultation dialogue cannot further supplement the entity condition carried by the first consultation dialogue, then the second consultation dialogue is identified as an abnormal consultation dialogue.

[0127] Step c: If yes, using the newly added entity conditions, determine the second response content of the second consultation dialogue by searching the business knowledge graph, and determine whether the second response content is consistent with the first response content; if consistent, determine that the second consultation dialogue is a duplicate consultation dialogue; if inconsistent, determine that the second consultation dialogue is a valid consultation dialogue.

[0128] In this step, if the newly added entity condition is related to the first business intent, that is, the newly added entity condition carried in the second consultation dialogue can further supplement the entity condition carried in the first consultation dialogue to obtain a more accurate response result; then, the newly added entity condition can be used to determine the second response content for answering the second consultation dialogue by searching the pre-generated business knowledge graph, and then it can be judged again whether the second response content is consistent with the first response content; if they are consistent, the second consultation dialogue can be considered a duplicate consultation dialogue of the first consultation dialogue; if they are inconsistent, it means that the user has raised a new consultation question. At this time, the second consultation dialogue can be considered a valid consultation dialogue, and a response needs to be given to the second consultation dialogue.

[0129] Here, the business knowledge graph is pre-organized based on the business environment in which the AI ​​chatbot operates. The business knowledge graph is a network structure that can show the relationship between business intent and entity conditions. The business knowledge graph includes each entity condition, the corresponding rhetorical question prompt for each entity condition (i.e., the rhetorical question prompt that the AI ​​chatbot should give when the user asks a question but lacks some entity conditions), and the response result that should be given after the user completes all entity conditions.

[0130] For example, a user's first inquiry to the AI ​​chatbot is "I want to apply for a residence permit." After recognizing the user's intention to "apply for a residence permit," the AI ​​chatbot, in order to clarify the user's intended business, guides the user to supplement the substantive conditions required for applying for a residence permit through feedback prompts. At this time, the AI ​​chatbot provides the user with the prompts, "Which of the following services do you want to apply for: 1. Applying for a new residence permit; 2. Replacing a residence permit?" Based on the user's response to the prompts, the AI ​​chatbot clarifies that the user's intention is to "replace a residence permit" and provides the user with the initial response, the procedure for replacing a residence permit.

[0131] If, at this point, the user raises a second inquiry about "replacement of residence permit," the AI ​​chatbot recognizes "replacement of residence permit" as the entity intent and extracts the new entity condition as "replacement of residence permit." Clearly, the AI ​​chatbot's second response to "replacement of residence permit" is consistent with the first response currently being provided, thus determining the second inquiry to be a duplicate. The AI ​​chatbot does not need to respond to the second inquiry; that is, the second inquiry will not interrupt the AI ​​chatbot's broadcast.

[0132] If, at this point, the user raises a second inquiry regarding "new residence permit application," the AI ​​chatbot identifies "new residence permit application" as the entity intent and extracts the new entity condition as "new residence permit application." Clearly, the AI ​​chatbot's second response to "new residence permit application" is inconsistent with the first response currently being provided. Therefore, the second inquiry is deemed a valid inquiry, and the AI ​​chatbot must respond to it; that is, at this point, the AI ​​chatbot must provide the user with the new application process for "new residence permit application."

[0133] During a user's inquiry, the surrounding environment may affect the user's ability to hear the AI ​​chatbot's message clearly at first. In such cases, the user may wish to repeat the inquiry so that the AI ​​chatbot can repeat the message. Therefore, in practice, although it may be determined that the second inquiry is a repeat of the first inquiry, it is still necessary to further determine whether the first response needs to be re-recited based on the time the inquiry was initiated.

[0134] In step S103, after determining that the second consultation dialogue is an invalid consultation dialogue, if the second consultation dialogue is a duplicate consultation dialogue of the first consultation dialogue, then it is necessary to further determine the time interval between the time of the first consultation dialogue and the time of the second consultation dialogue.

[0135] In step S104, if the time interval is less than or equal to the preset interval threshold, it can be considered that the user is only making a repeated inquiry without receiving a timely reply. In this case, the second inquiry dialogue made by the user can be ignored, and the first reply content can be broadcast again.

[0136] In step S105, if the time interval is greater than the preset interval threshold, it can be assumed that the user has not heard the first reply being broadcast clearly. In this case, the first reply needs to be broadcast to the user again.

[0137] In one embodiment, the identification method further includes: step S106 (e.g. Figure 1 (as shown) If not, using the dialogue intent and / or the second entity information, by searching the business knowledge graph, determine the second response content for answering the second consultation dialogue, and broadcast the second response content to the user.

[0138] In step S106, if it is determined that the second consultation dialogue is not an invalid consultation dialogue, that is, the second consultation dialogue is a valid consultation dialogue, then it means that a response needs to be given to the second consultation dialogue raised by the user. At this time, the second response content can be determined by searching the business knowledge graph based on the dialogue intent of the second consultation dialogue and / or the second entity information extracted from the second consultation dialogue, and the second response content is broadcast to the user.

[0139] Here, if the second response content of the second consultation dialogue has already been obtained through the business knowledge graph during the process of determining whether the second consultation dialogue is invalid, then there is no need to obtain it again; the already obtained second response content can be broadcast directly.

[0140] In one embodiment, the identification method further includes: when the second consultation dialogue is an abnormal consultation dialogue, ignoring the second consultation dialogue and continuing to broadcast the first reply content.

[0141] In this step, when the second consultation dialogue is determined to be an abnormal consultation dialogue, it means that the second consultation dialogue neither raises new consultation questions nor can it supplement the entity information under the first business intent indicated by the first consultation dialogue. In this case, the second consultation dialogue raised by the user can be ignored, and the first response content can continue to be broadcast.

[0142] Please see Figure 2 , Figure 2 This is a schematic diagram of a dialogue interaction process provided in an embodiment of this application. Figure 2 As shown, step 201: broadcast the first response content of the first consultation dialogue to the user; step 202: determine the dialogue intent and / or second entity information of the second consultation dialogue received by the user through a Natural Language Understanding (NLU) model; step 203: determine whether the dialogue intent of the second consultation dialogue has a second business intent or an entity intent; if it is a second business intent, proceed to step 204; if it is an entity intent, proceed to step 209; step 204: determine whether the second business intent can represent the true business intent of the second consultation dialogue; if yes, proceed to step 205; if no, proceed to step 208; step 205: determine whether the second business intent is the same as the first business intent of the first consultation dialogue; if yes, proceed to step 206; if no, proceed to step 207; step 206: determine that the second consultation dialogue is a repeat of the first consultation dialogue, and do not interrupt the broadcast of the first response content; Step 207: Determine if the second consultation dialogue is a valid consultation dialogue, interrupt the broadcast of the first response content, and broadcast the second response content corresponding to the second consultation dialogue to the user; Step 208: Determine if the second consultation dialogue is an abnormal consultation dialogue, and do not interrupt the broadcast of the first response content; Step 209: Determine if the second entity information carries any new entity conditions. If not, proceed to step 208; if yes, proceed to step 210; Step 210: Determine if the new entity conditions are related to the first business intent of the first consultation dialogue. If not, proceed to step 208; if yes, proceed to step 211; Step 211: Using the new entity conditions carried by the second entity information, determine the second response content of the second consultation dialogue by searching the business knowledge graph; Step 212: Determine if the second response content is consistent with the first response content. If consistent, proceed to step 206; if inconsistent, proceed to step 207.

[0143] The method for identifying repeated dialogues provided in this application embodiment, during the process of broadcasting the first response content of a first consultation dialogue initiated by a user, responds to a second consultation dialogue initiated by the user by determining the dialogue intent and / or the second entity information carried by the second consultation dialogue; based on the dialogue intent and / or the second entity information, determining whether the second consultation dialogue is an invalid consultation dialogue; when the second consultation dialogue is a repeated consultation dialogue, determining the time interval between the initiation of the second consultation dialogue and the first consultation dialogue; when the initiation time interval is less than or equal to a preset interval threshold, ignoring the second consultation dialogue and continuing to broadcast the first response content; when the initiation time interval is greater than the preset interval threshold, rebroadcasting the first response content. In this way, it can prevent users from unintentionally interrupting the voice interaction process with repeated consultation dialogues, reducing the response frequency of the AI ​​dialogue robot during the interaction process, and improving the efficiency of human-computer interaction.

[0144] Please see Figure 3 , Figure 4 , Figure 3 This is one of the structural schematic diagrams of a repeat dialogue recognition device provided in an embodiment of this application. Figure 4 This is a second schematic diagram of a repeating dialogue recognition device provided in an embodiment of this application. Figure 3 As shown, the identification device 300 includes:

[0145] The intent determination module 310 is used to determine the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue in response to the second consultation dialogue proposed by the user during the process of broadcasting the first response content of the first consultation dialogue proposed by the user.

[0146] The judgment module 320 is used to determine whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent and / or the second entity information; wherein, the invalid consultation dialogue includes duplicate consultation dialogue and abnormal consultation dialogue;

[0147] The interval comparison module 330 is used to determine the time interval between the second consultation dialogue and the first consultation dialogue when the second consultation dialogue is a repeated consultation dialogue.

[0148] The first continuous broadcasting module 340 is used to ignore the second consultation dialogue and continue broadcasting the first reply content when the time interval between the requests is less than or equal to a preset time interval threshold.

[0149] The rebroadcast module 350 is used to rebroadcast the first response content when the time interval between the requests exceeds a preset time interval threshold.

[0150] Furthermore, such as Figure 4 As shown, the identification device 300 further includes a second continuous broadcast module 360, which is used for:

[0151] When the second consultation dialogue is an abnormal consultation dialogue, ignore the second consultation dialogue and continue to broadcast the first response content.

[0152] Furthermore, the dialogue intent includes a second business intent. When the judgment module 320 determines whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent, the judgment module 320 is used to:

[0153] Based on the probability score of the second consultation dialogue having the second business intention, it is determined whether the second business intention can represent the true business intention of the second consultation dialogue; wherein, the second business intention is the candidate business intention with the highest probability score among the second consultation dialogues;

[0154] If not, the second consultation dialogue is determined to be an abnormal consultation dialogue;

[0155] If so, determine whether the second business intent is the same as the first business intent of the first consultation dialogue;

[0156] When the second business intent is the same as the first business intent, the second consultation dialogue is determined to be a repeated consultation dialogue;

[0157] When the second business intent is different from the first business intent, the second consultation dialogue is determined to be a valid consultation dialogue.

[0158] Furthermore, when determining whether the second business intent can represent the true business intent of the second consultation dialogue based on the probability score of the second consultation dialogue having the second business intent, the judgment module 320 is used to:

[0159] If the probability score of the second consultation dialogue having the second business intention is greater than or equal to the preset score threshold, then it is determined that the second business intention can represent the true business intention of the second consultation dialogue.

[0160] or,

[0161] If the probability score of the second consultation dialogue having the second business intention is less than the preset score threshold, and the difference between the second and the second high probability score is greater than or equal to the preset difference threshold, then it is determined that the second business intention can characterize the true intention of the second consultation dialogue.

[0162] Furthermore, the dialogue intent includes entity intent. When the judgment module 320 determines whether the second consultation dialogue is an invalid consultation dialogue based on the dialogue intent and the second entity information, the judgment module 320 is used to:

[0163] If the dialogue intent is an entity intent and the second entity information indicates that no new entity conditions have been extracted from the second consultation dialogue, the second consultation dialogue is determined to be an abnormal consultation dialogue.

[0164] If the dialogue intent is an entity intent and the second entity information includes newly added entity conditions extracted from the second consultation dialogue, then based on the newly added entity conditions, it is determined whether the second consultation dialogue is an invalid consultation dialogue.

[0165] Furthermore, when determining whether the second consultation dialogue is invalid based on the newly added entity conditions, the determination module 320 is used to:

[0166] Determine whether the newly added entity condition is related to the first business intent of the first consultation dialogue;

[0167] If not, the second consultation dialogue is determined to be an abnormal consultation dialogue;

[0168] If so, using the newly added entity conditions, by searching the business knowledge graph, determine the content of the second response to the second consultation dialogue, and determine whether the content of the second response is consistent with the content of the first response;

[0169] If they match, the second consultation dialogue is determined to be a duplicate consultation dialogue;

[0170] If there is no discrepancy, the second consultation dialogue is determined to be a valid consultation dialogue.

[0171] Furthermore, when the intent determination module 310 determines the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue in response to the user's second consultation dialogue, the intent determination module 310 is used to:

[0172] In response to the second consultation dialogue proposed by the user, a natural language understanding model is used to determine the probability score between the second consultation dialogue and each candidate intent, and the natural language understanding model is used to extract the second entity information carried by the second consultation dialogue;

[0173] The candidate intent with the highest probability score is determined as the dialogue intent of the second consultation dialogue.

[0174] The duplicate dialogue identification device provided in this application embodiment, during the process of broadcasting the first response content of a first consultation dialogue proposed by a user, responds to a second consultation dialogue proposed by the user, determines the dialogue intent of the second consultation dialogue and / or the second entity information carried by the second consultation dialogue; based on the dialogue intent and / or the second entity information, determines whether the second consultation dialogue is an invalid consultation dialogue; when the second consultation dialogue is a duplicate consultation dialogue, determines the time interval between the second consultation dialogue and the first consultation dialogue; when the time interval is less than or equal to a preset interval threshold, ignores the second consultation dialogue and continues to broadcast the first response content; when the time interval is greater than the preset interval threshold, rebroadcasts the first response content. In this way, it can prevent users from unintentionally interrupting the voice interaction process with duplicate consultation dialogues, reduce the response frequency of the AI ​​dialogue robot during the interaction process, and improve the efficiency of human-computer interaction.

[0175] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.

[0176] The memory 520 stores machine-readable instructions executable by the processor 510. When the electronic device 500 is running, the processor 510 and the memory 520 communicate via the bus 530. When the machine-readable instructions are executed by the processor 510, they can perform the operations described above. Figure 1 The steps of the repeated dialogue identification method in the method embodiment shown are described in detail in the method embodiment, and will not be repeated here.

[0177] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 1 The steps of the repeated dialogue identification method in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0178] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0179] The units described as separate components may or may not be physically separate. 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0180] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0181] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0182] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method of identifying a repetitive dialogue, characterized by, The identification method comprises: In the process of broadcasting the first reply content of the first consultation dialogue proposed by a user, in response to a second consultation dialogue proposed by the user, determining a dialogue intention possessed by the second consultation dialogue and second entity information carried by the second consultation dialogue; According to the dialogue intention and the second entity information, determining whether the second consultation dialogue is an invalid consultation dialogue; wherein the invalid consultation dialogue comprises a repeated consultation dialogue and an abnormal consultation dialogue; the dialogue intention comprises an entity intention, and the determination of whether the second consultation dialogue is an invalid consultation dialogue according to the dialogue intention and the second entity information comprises: if the dialogue intention is an entity intention and the second entity information indicates that no new entity condition is extracted from the second consultation dialogue, determining that the second consultation dialogue is an abnormal consultation dialogue; if the dialogue intention is an entity intention and the second entity information comprises a new entity condition extracted from the second consultation dialogue, determining whether the second consultation dialogue is an invalid consultation dialogue according to the new entity condition; the determination of whether the second consultation dialogue is an invalid consultation dialogue according to the new entity condition comprises: determining whether the new entity condition has an association relationship with a first business intention of the first consultation dialogue; if not, determining that the second consultation dialogue is an abnormal consultation dialogue; if yes, using the new entity condition to determine second reply content of the second consultation dialogue by searching a business knowledge graph, and determining whether the second reply content is consistent with the first reply content; if consistent, determining that the second consultation dialogue is a repeated consultation dialogue; if not consistent, determining that the second consultation dialogue is a valid consultation dialogue; When the second consultation dialogue is a repeated consultation dialogue, determining an interval of the proposing time of the second consultation dialogue and the first consultation dialogue; When the interval of the proposing time is less than or equal to a preset interval threshold, ignoring the second consultation dialogue and continuing to broadcast the first reply content; When the interval of the proposing time is greater than the preset interval threshold, re-broadcasting the first reply content.

2. The identification method according to claim 1, characterized in that, The identification method further comprises: When the second consultation dialogue is an abnormal consultation dialogue, ignoring the second consultation dialogue and continuing to broadcast the first reply content.

3. The identification method according to claim 1, characterized in that, The determination of the dialogue intention possessed by the second consultation dialogue and the second entity information carried by the second consultation dialogue in response to the second consultation dialogue proposed by the user comprises: In response to the second consultation dialogue proposed by the user, using a natural language understanding model to determine a possibility score between the second consultation dialogue and each candidate intention, and using the natural language understanding model to extract second entity information carried by the second consultation dialogue; Determining the candidate intention with the highest possibility score as the dialogue intention possessed by the second consultation dialogue.

4. An apparatus for identifying a repeat conversation, the apparatus comprising: The identification device comprises: The intention determination module is configured to determine, in a process of broadcasting a first reply content of a first consultation dialogue initiated by a user, a dialogue intention possessed by a second consultation dialogue initiated by the user and second entity information carried by the second consultation dialogue; The judgment module is configured to determine, according to the dialogue intention and the second entity information, whether the second consultation dialogue is an invalid consultation dialogue; the invalid consultation dialogue includes a repeated consultation dialogue and an abnormal consultation dialogue; the dialogue intention includes an entity intention; when determining, according to the dialogue intention and the second entity information, whether the second consultation dialogue is an invalid consultation dialogue, the judgment module is configured to: if the dialogue intention is an entity intention and the second entity information indicates that no new entity condition is extracted from the second consultation dialogue, determine that the second consultation dialogue is an abnormal consultation dialogue; if the dialogue intention is an entity intention and the second entity information includes a new entity condition extracted from the second consultation dialogue, determine, according to the new entity condition, whether the second consultation dialogue is an invalid consultation dialogue; when determining, according to the new entity condition, whether the second consultation dialogue is an invalid consultation dialogue, the judgment module is configured to: determine whether the new entity condition has an association relationship with a first business intention of the first consultation dialogue; if not, determine that the second consultation dialogue is an abnormal consultation dialogue; if yes, determine second reply content of the second consultation dialogue by searching a business knowledge graph by using the new entity condition, and determine whether the second reply content is consistent with the first reply content; if yes, determine that the second consultation dialogue is a repeated consultation dialogue; if not, determine that the second consultation dialogue is a valid consultation dialogue; The interval comparison module is configured to determine, when the second consultation dialogue is a repeated consultation dialogue, a time interval between initiation of the second consultation dialogue and the first consultation dialogue; The first continuous broadcasting module is configured to ignore the second consultation dialogue and continue broadcasting the first reply content when the time interval is less than or equal to a preset interval threshold. The re-broadcasting module is configured to re-broadcast the first reply content when the time interval is greater than the preset interval threshold.

5. An electronic device, comprising: The processor, the memory and the bus, the memory stores machine readable instructions executable by the processor, when the electronic device runs, the processor and the memory communicate through the bus, the machine readable instructions are executed by the processor to execute the steps of the repeated dialogue identification method in any one of claims 1 to 3. The computer readable storage medium stores a computer program, the computer program is executed by the processor to execute the steps of the repeated dialogue identification method in any one of claims 1 to 3.

6. A computer readable storage medium characterized by, ​

Images