Interaction method, interaction system, computing device and readable storage medium

Abstract

The invention discloses an interaction method, an interaction system, computing equipment and a computer readable storage medium. The interaction method comprises the following steps: acquiring first question information; and generating and / or outputting interaction information corresponding to the first problem information according to a response strategy corresponding to a target service scene when the target service scene corresponding to the first problem information is identified to be different from the current service scene through the AIGC large model. According to the invention, whether the obtained problem information updates or changes the current service scene can be conveniently and accurately judged through the AIGC large model, and when the current service scene is updated or changed, the user intention corresponding to the problem information can be accurately understood according to the service processing logic corresponding to the updated or changed target service scene, so that the user experience is improved. And outputting the corresponding interaction information so as to execute the business event corresponding to the target business scene, thereby meeting the actual demand of the user. The user intention understanding ability can be improved in the interaction service, and the user experience is improved.

Description

Technical Field

[0001] The present application relates to the field of interactive technology, and in particular to an interactive method, an interactive system, a computing device, and a computer-readable storage medium. Background Art

[0002] The existing interactive service business orchestration methods include prompt word orchestration and process orchestration. Among them, prompt word orchestration writes business logic into prompt words and plug-ins, and uses the capabilities of semantic understanding and intent recognition to perform business flow. Among them, the existing process orchestration has multiple business process nodes, each of which has independent business processing logic, and can understand user intent and execute corresponding business events according to the business processing logic. Usually an existing process orchestration corresponds to a business scenario.

[0003] However, the existing interactive service orchestration methods cannot promptly identify when the user's conversation switches to a different business scenario, which can easily lead to misunderstanding of the user's intent and executing the wrong business event, seriously affecting the user experience. Therefore, how to improve the accuracy of understanding the user's intent in interactive services is a technical problem that technicians in this field need to solve urgently. Summary of the invention

[0004] The purpose of this application is to provide an interaction method, an interaction system, a computing device and a computer-readable storage medium, which can improve the accuracy of understanding user intentions in interactive services, thereby improving user experience.

[0005] To achieve the above objectives: In a first aspect, an embodiment of the present application provides an interaction method, including: obtaining first question information; when identifying through the AIGC big model that the target business scenario corresponding to the first question information is different from the current business scenario, generating and / or outputting interaction information corresponding to the first question information according to the response strategy corresponding to the target business scenario.

[0006] In one implementation, the target business scenario includes one of the following: a repair scenario, a consultation scenario, a business opportunity scenario, and a manual transfer scenario.

[0007] In one implementation, the response strategy corresponding to the repair scenario includes at least one of the following: When it is identified that the first problem information corresponds to a repair intention, first interaction information is output, and the first interaction information is used to trigger a repair process; When it is identified that the first question information corresponds to the manual entry intention of the work order, second interaction information is generated and output, and the second interaction information includes to-do tasks for the business party and / or reply information for the demand party; When it is identified that the first question information corresponds to the work order machine entry intention, third interaction information is generated and output, and the third interaction information is used to trigger the question and answer process of the work order entry.

[0008] In one embodiment, the response strategy corresponding to the consultation scenario includes: generating answer information for the first question information through the AIGC big model, and outputting the answer information.

[0009] In one embodiment, when the AIGC big model identifies that the target business scenario corresponding to the first question information is different from the current business scenario, after the step of generating and / or outputting interactive information corresponding to the first question information according to the response strategy corresponding to the target business scenario, it includes: obtaining second question information; when the AIGC big model identifies that the second question information corresponds to the target business scenario, obtaining question and answer context information corresponding to the target business scenario; parsing the question and answer context information and the second question information through the AIGC big model to generate and output interactive information corresponding to the second question information.

[0010] In one implementation, the question information is natural language information.

[0011] In one embodiment, the step of generating and / or outputting interactive information corresponding to question information includes: extracting emotion determination features in the question information, the emotion determination features including a combination of one or more of the pitch of speech, the rhythm of speech, the vocabulary of the text, and the sentence structure of the text; performing emotion determination based on the emotion determination features to determine the current conversation emotion; matching reply request information based on the current conversation emotion, the reply request information including at least one of the tone of reply and the vocabulary of reply; and outputting the reply information corresponding to the question information according to the reply requirement information.

[0012] In one embodiment, the step of performing emotion determination based on emotion determination features to determine the current conversation emotion includes: obtaining question and answer context information corresponding to the target business scenario; determining the historical conversation emotion based on the question and answer context information; performing emotion determination based on the emotion determination features and the historical conversation emotions to determine the current conversation emotion.

[0013] In a second aspect, an embodiment of the present application provides an interactive system, comprising: an AIGC gateway node, a business interaction node and a response node; the AIGC gateway node is used to implement the interactive method according to any one of claims 1 to 7; the business interaction node is used to call the AIGC gateway node to identify the target business scenario corresponding to the received business problem information, and when the target business scenario is different from the current business scenario corresponding to the business interaction node, enter the target business scenario through the AIGC gateway node, or, when the target business scenario is the same as the current business scenario corresponding to the business interaction node, conduct a business dialogue with the business interaction logic corresponding to the business interaction node; the response node is used to perform a business response according to the business dialogue result fed back by the business interaction node.

[0014] In a third aspect, an embodiment of the present application provides a computing device, comprising: a processor and a memory storing a computer program, and when the processor runs the computer program, the steps of the interaction method described in any one of the above items are implemented.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of the interaction method described in any one of the above items are implemented.

[0016] The interactive method, interactive system, computing device and computer-readable storage medium provided by the embodiments of the present application include: obtaining first question information; when the target business scenario corresponding to the first question information is identified to be different from the current business scenario through the AIGC big model, generating and / or outputting interactive information corresponding to the first question information according to the response strategy corresponding to the target business scenario. Through the technical solution of the present application, it is possible to conveniently and accurately determine whether the acquired question information has updated or changed the current business scenario through the AIGC big model, and when the current business scenario is updated or changed, it is possible to accurately understand the user intention corresponding to the question information according to the business processing logic corresponding to the updated or changed target business scenario, and output the corresponding interactive information to execute the business event corresponding to the target business scenario, thereby meeting the actual needs of the user. In this way, the technical solution of the present application can improve the ability to understand user intentions in interactive services, thereby improving user experience.

[0017] In addition, the interactive system includes: an AIGC gateway node, a business interaction node and a response node; the AIGC gateway node is used to implement the interactive method according to any one of claims 1 to 7; the business interaction node is used to call the AIGC gateway node to identify the target business scenario corresponding to the received business problem information, and when the target business scenario is different from the current business scenario corresponding to the business interaction node, the target business scenario is entered through the AIGC gateway node, or, when the target business scenario is the same as the current business scenario corresponding to the business interaction node, the business interaction logic corresponding to the business interaction node is used to conduct a business dialogue; the response node is used to conduct a business response according to the business dialogue result fed back by the business interaction node. The interactive system provided by the present application can understand the user's scenario intention at the business interaction node through the AIGC gateway node, thereby jumping to the target business scenario corresponding to the scenario intention, so that the interactive system can capture the changes in the user's scenario intention in real time and flexibly, thereby switching the business scenario, so that the interactive system provides effective interactive services in business scenarios that meet user needs, thereby improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application. In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor.

[0019] Figure 1 It is a flowchart of the interaction method provided in the first embodiment of the present application.

[0020] Figure 2 It is a schematic diagram of the framework of the interactive system provided in the second embodiment of the present application.

[0021] Figure 3 It is a workflow diagram of an interactive system applied to a customer service business environment for a device according to an example of the present application.

[0022] Figure 4 It is a structural diagram of the computing device provided by this application.

[0023] The realization of the purpose, functional features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. The above-mentioned drawings have shown clear embodiments of this application, which will be described in more detail later. These drawings and textual descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0024] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0025] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element. In addition, components, features, and elements with the same name in different embodiments of the present application may have the same meaning or different meanings, and their specific meanings need to be determined by their explanation in the specific embodiment or further combined with the context of the specific embodiment.

[0026] It should be understood that, although the terms first, second, third, etc. may be used to describe various information in this article, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this article, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "at the time of..." or "when..." or "in response to determination". Furthermore, as used in this article, the singular forms "one", "one" and "the" are intended to also include plural forms, unless there is an opposite indication in the context. It should be further understood that the terms "comprising", "including" indicate that there are described features, steps, operations, elements, components, projects, kinds, and / or groups, but do not exclude the existence, occurrence or addition of one or more other features, steps, operations, elements, components, projects, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Thus, “A, B, or C” or “A, B and / or C” means “any of the following: A; B; C; A and B; A and C; B and C; A, B, and C.” An exception to this definition will occur only when a combination of elements, functions, steps, or operations are inherently mutually exclusive in some manner.

[0027] It should be understood that, although the various steps in the flowchart in the embodiment of the present application are displayed in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and it can be performed in other orders. Moreover, at least a portion of the steps in the figure may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and their execution order is not necessarily performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0028] It should be noted that, in this article, step codes such as S11, S12, etc. are used for the purpose of expressing the corresponding content more clearly and concisely, and do not constitute a substantial limitation on the sequence. When implementing it specifically, those skilled in the art may execute S12 first and then S11, etc., but these should all be within the scope of protection of this application.

[0029] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0030] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present application, and have no specific meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.

[0031] The following is an explanation of the terms that may be involved in this application, as follows: AIGC, or Artificial Intelligence Generated Content, is a technology that uses artificial intelligence technology, especially machine learning and deep learning, to generate content with certain creativity and quality, and it has excellent recognition capabilities for user intent. In this application, a large model based on AIGC (referred to as AIGC large model) is used as intelligent customer service, using its excellent recognition capabilities for user intent to identify business scenarios based on user conversations (or question information); AIGC large models, such as ChatGPT, are an important part of AIGC technology. Through large-scale data training, they can understand and generate natural language and provide an intelligent conversation experience.

[0032] Business orchestration can represent the planning of business processes based on the business attributes corresponding to a business scenario, as well as the formulation of business logic for each business node in the business process.

[0033] First embodiment See also Figure 1 , is an interactive method provided in the first embodiment of the present application, which can be executed by a computing device provided in the embodiment of the present application, and the computing device can be implemented in software and / or hardware. Optionally, the computing device is, for example, one of a terminal device and a server.

[0034] This embodiment provides an interaction method, including the following steps (for example, step S11 to step S12): Step S11: Obtain first question information.

[0035] In one embodiment, the first question information may be natural language information, such as user dialogue information. In addition, the first question information may be processed language information obtained by extracting and combining features of natural speech information. In this case, using the processed language information to import the corresponding model can improve the efficiency of the model output result.

[0036] In one embodiment, the processed language information is obtained after feature extraction and feature combination of natural speech information, for example, modal particles in the natural speech information are eliminated, characteristic vocabulary of the natural speech information is extracted and recombined according to the language arrangement logic, etc.

[0037] Step S12: When the AIGC big model identifies that the target business scenario corresponding to the first question information is different from the current business scenario, the interactive information corresponding to the first question information is generated and / or output according to the response strategy corresponding to the target business scenario.

[0038] In one embodiment, the interaction information may represent answer information of the question information and signal information triggering various paths.

[0039] In one embodiment, a business scenario can represent a description or definition of a scenario in a specific business environment, so as to better understand and handle various business situations in the scenario. For example, the business scenarios in the customer service business environment of small commodities can be divided into pre-sales consultation scenarios, after-sales service scenarios, and manual transfer scenarios, etc.; for another example, the business scenarios in the customer service business environment of bulk equipment can be divided into repair scenarios, consultation scenarios, business opportunity scenarios, and manual transfer scenarios, etc. For another example, the business scenarios in the livelihood service scenarios of government departments can be divided into demand entry scenarios, manual transfer scenarios, policy interpretation scenarios, etc.

[0040] In one implementation, the response strategy corresponding to the business scenario may represent a method of determining the business goal corresponding to the problem information and / or triggering the implementation process of the business goal in a specific business scenario.

[0041] In one embodiment, step S12: when the AIGC big model identifies that the target business scenario corresponding to the first question information is different from the current business scenario, before generating and / or outputting the interactive information corresponding to the first question information according to the response strategy corresponding to the target business scenario, it may include: obtaining the current business scenario.

[0042] In one implementation, the current business scenario is obtained in a manner including but not limited to one of the following: In response to the setting operation, the current business scenario is set; The AIGC big model is used to identify the historical question and answer context information to determine the current business scenario; The initialization scenario is used as the current business scenario.

[0043] In one embodiment, the AIGC big model can record multiple conversations or requests to track and record the processing status of a business (for example, completed or uncompleted), so that the current business scenario can be determined based on the processing status of the business and the business attributes corresponding to the business.

[0044] In one embodiment, the initialization business scenario may represent a default scenario corresponding to when no operation is performed or when the interactive service is reset, which may represent the initialization state of the interactive service. Thus, in the technical solution of this embodiment, when the initialization scenario is used as the current business scenario, as long as any target business scenario is identified according to the question information, it can be directly determined that the target business scenario corresponding to the question information is different from the current business scenario.

[0045] In one implementation, the technical solution of this embodiment can pre-set a scenario library, which includes multiple business scenarios and their corresponding scenario attribute features, so that the AIGC big model can access the preset scenario library and extract scenario attribute features of the problem information when the problem information is obtained, so that the business scenario corresponding to the problem information can be determined based on the extracted scenario attribute features.

[0046] In one implementation, the target business scenario includes, but is not limited to, one of the following: a repair scenario, a consultation scenario, a business opportunity scenario, and a manual transfer scenario.

[0047] In one implementation, the response strategy corresponding to the repair scenario includes but is not limited to at least one of the following: When it is identified that the first problem information corresponds to a repair intention, first interaction information is output, and the first interaction information is used to trigger a repair process; When it is identified that the first question information corresponds to the manual entry intention of the work order, second interaction information is generated and output, and the second interaction information includes to-do tasks for the business party and / or reply information for the demand party; When it is identified that the first question information corresponds to the work order machine entry intention, third interaction information is generated and output, and the third interaction information is used to trigger the question and answer process of the work order entry.

[0048] In one embodiment, the repair process may include a repair service interaction node and / or a repair service response node, etc. Among them, the repair service interaction node may represent a node that needs to interact with the user to determine the repair service intention or demand. Among them, the repair service response node may represent a node that does not need to interact with the user and processes information on its own.

[0049] In one embodiment, the following example illustrates a repair process in a repair scenario, wherein the repair response node may include a direct reply node, a gateway routing node, a content extraction node, and a manual task node; wherein the repair service interaction node may include an intent recognition node and a large model node; the process logic is as follows: (1) Direct reply node: outputs the prompt content for entering the basic information of the repair report, such as "Hello, please provide your device model"; (2) Intent recognition node: This node interacts with the user to receive the user's question information and performs semantic recognition on the question information to determine the device type and other information corresponding to the user's device repair intention. (3) Gateway routing node: uses the determined device type and other information as conditions to transfer to different repair sub-processes. For example, device type A corresponds to repair sub-process 1, and device type B corresponds to repair sub-process 2. Repair Sub-Process 1: Large model node: Through prompt word arrangement, it starts a dialogue with the user and collects the detailed content of the repair report; Content extraction node: extracts key information from the context of the repair report through prompt words, and enters the extracted key information into the work order template through the configured plug-in (API) to obtain the repair work order, so as to realize the work order machine entry function; Direct reply node: Output the prompt content corresponding to the status information of the repair work order, such as "Thank you for your cooperation, your repair report has been accepted."

[0050] Repair Sub-Process 2: Manual task node: Generates a to-do task based on the determined device type and other information, and sends the to-do task to the business stakeholders; Direct reply node: Outputs the prompt content corresponding to the status information of the to-do task, such as "Thank you for your cooperation, your repair report has been accepted, and an engineer will contact you directly later."

[0051] In this way, the technical solution of this embodiment can trigger the repair process through interactive information in the warranty scenario, so that users can realize convenient and accurate repair reporting.

[0052] In one embodiment, when it is identified that the first problem information corresponds to the intention of manual entry of the work order, the second interactive information is generated and output, and the second interactive information includes a to-do task for the business party and / or a reply information for the demand party. For example, when it is identified that the first problem information corresponds to the intention of manual entry of the work order, it directly jumps to the manual task node in the repair sub-process 2 included in the repair process to generate a to-do task based on the determined equipment type and other information, and sends the to-do task to the business-related party, and outputs the prompt content corresponding to the status information of the to-do task to the demand party (ie, the user) through the direct reply node, such as "Thank you for your cooperation, your repair report has been accepted, and an engineer will contact you directly later."

[0053] In one embodiment, when it is identified that the first question information corresponds to the intention of machine entry of a work order, third interaction information is generated and output, and the third interaction information is used to trigger the question-answering process of work order entry. For example, when it is identified that the first question information corresponds to the intention of machine entry of a work order, it directly jumps to the large model node in the repair sub-process 1 included in the repair process (through prompt word arrangement, dialogue with the user, and collection of detailed content of the repair report), and then flows to the content extraction node to extract key information in the context of the detailed content of the repair report through the prompt word, and enters the extracted key information into the work order template through the configured plug-in (API) to obtain the repair work order, so as to realize the work order machine entry function.

[0054] In one embodiment, the response strategy corresponding to the consultation scenario may include: when the first question information corresponds to the consultation intention, generating answer information (ie, interaction information) of the first question information through the AIGC big model, and outputting the answer information.

[0055] In one embodiment, the AIGC big model can access a preset knowledge base and other literature query platforms to generate answer information for the first question information and output the answer information.

[0056] In one implementation, the response strategy corresponding to the business opportunity scenario may include: when the first question information corresponds to the cooperation intention, outputting interactive information for triggering the question-answer process of cooperation content collection.

[0057] In one embodiment, the response strategy corresponding to the manual transfer scenario may include: when the first question information corresponds to the manual transfer intention, generating a manual docking task (ie, interaction information) and outputting it to the manual agent platform.

[0058] In one embodiment, step S12: when the target business scenario corresponding to the first question information is identified by the AIGC big model as being different from the current business scenario, generating and / or outputting an interactive message corresponding to the first question information according to the response strategy corresponding to the target business scenario, it may include: obtaining the second question information; when the second question information is identified by the AIGC big model as corresponding to the target business scenario, obtaining the question and answer context information corresponding to the target business scenario; parsing the question and answer context information and the second question information by the AIGC big model to generate and output interactive information corresponding to the second question information. In this way, the technical solution of this embodiment can record multiple conversations or requests in a business scenario through the AIGC big model, so that the user's intention can be accurately identified through the question and answer context information corresponding to the previous multiple conversations or requests, so that the answer information of the question information can be determined based on the identified user's intention to meet the actual needs of the user and improve the user experience.

[0059] In one implementation, both the first question information and the second question information may be summarized as question information.

[0060] In one implementation, the question information may be natural language information.

[0061] In one embodiment, the step of generating and / or outputting interactive information corresponding to the question information may include: extracting emotion determination features in the question information; performing emotion determination according to the emotion determination features to determine the current conversation emotion; matching the reply requirement information according to the current conversation emotion; and outputting the reply information corresponding to the question information according to the reply requirement information. In this way, the technical solution of this embodiment can realize the emotion recognition integrated into the user conversation (i.e., question information), identify the user's conversation emotion, and integrate the reply requirement information adapted to the user's conversation emotion when answering questions or prompting at each reply node, so as to provide a more humanized interactive service. For example, when the user's conversation emotion represents anger, a gentle tone is used for auditory soothing when answering questions or prompting, and soft and polite words are integrated for soothing at the language expression level; for another example, when the user's conversation emotion represents boredom, humorous words are integrated and a naughty tone is used when answering questions or prompting to enhance user interest.

[0062] In one embodiment, the emotion determination feature may represent biological features that can reflect the emotions of the characters during the conversation and text features of the expression content.

[0063] In one embodiment, the emotion determination features include, but are not limited to, a combination of one or more of the following: voice pitch, voice rhythm, text vocabulary, and text sentence structure. For example, they may also include facial expressions.

[0064] In one embodiment, the pitch of the voice: analyzes the pitch of the voice, where a high pitch may indicate excitement or anxiety, and a low pitch may indicate sadness or seriousness. The rhythm of the voice: analyzes the speaking speed, frequency and length of pauses, where a fast pace may indicate tension or excitement, and a slow pace may indicate contemplation or sadness.

[0065] In one implementation, the current conversation emotion may represent the emotion of the user in the current conversation scenario.

[0066] In one embodiment, the current conversation emotion may be, but is not limited to, one of the following: happiness, sadness, fear, surprise, disgust, trust, shame, jealousy, curiosity, excitement, boredom, anxiety, frustration, expectation, satisfaction, gratitude, contempt, etc.

[0067] In one embodiment, the reply request information includes but is not limited to at least one of the tone of reply and the vocabulary of reply. Optionally, when the reply information is output by a robot with expression changing function, the reply request information may also include the expression of reply.

[0068] In one embodiment, the step of performing emotion determination based on emotion determination features to determine the current conversation emotion may include: obtaining question and answer context information corresponding to the target business scenario; determining the historical conversation emotion based on the question and answer context information; performing emotion determination based on the emotion determination features and the historical conversation emotions to determine the current conversation emotion.

[0069] In one embodiment, since a person's emotional changes during an ongoing conversation should be gradual and will not jump between two completely opposite emotions, the technical solution of this embodiment can incorporate the historical conversation emotions of the question and answer context into the process of identifying the current conversation emotions, thereby improving the accuracy of identifying the current conversation emotions.

[0070] In one embodiment, emotion determination can be implemented by using a pre-trained machine model (or emotion recognition model) to perform emotion determination. Alternatively, a machine learning or deep learning model, such as a support vector machine (SVM), a deep neural network (DNN), or a Transformer model, can be used to train the emotion recognition model.

[0071] The interaction method provided in this embodiment includes: step S11: obtaining the first question information; step S12: when the target business scenario corresponding to the first question information is identified by the AIGC big model as being different from the current business scenario, generating and / or outputting the interaction information corresponding to the first question information according to the response strategy corresponding to the target business scenario. Through the technical solution of this embodiment, it is possible to conveniently and accurately determine whether the acquired question information has updated or changed the current business scenario through the AIGC big model, and when the current business scenario is updated or changed, it is possible to accurately understand the user intention corresponding to the question information according to the business processing logic corresponding to the updated or changed target business scenario, and output the corresponding interaction information to execute the business event corresponding to the target business scenario, thereby meeting the actual needs of the user. In this way, the technical solution of this embodiment can improve the ability to understand user intentions in interactive services, thereby improving the user experience. Second embodiment See also Figure 2 , is a structural diagram of an interactive system provided in the first embodiment of the present application.

[0072] This embodiment provides an interactive system, including: an AIGC gateway node Q1, a service interaction node Q2, and a response node Q3.

[0073] The AIGC gateway node Q1 is used to implement the steps of the interaction method described in the first embodiment.

[0074] In one embodiment, the AIGC gateway node Q1 is used to identify the target business scenario corresponding to the business problem information through the AIGC big model, and determine whether the target business scenario is different from the current business scenario.

[0075] Among them, the business interaction node Q2 is used to call the AIGC gateway node Q1 to identify the target business scenario corresponding to the received business problem information, and when the target business scenario is different from the current business scenario corresponding to the business interaction node Q2, enter the target business scenario through the AIGC gateway node Q1, or, when the target business scenario is the same as the current business scenario corresponding to the business interaction node Q2, conduct business dialogue with the business interaction logic corresponding to the business interaction node Q2.

[0076] The response node Q3 is used to perform a business response according to the business dialogue result fed back by the business interaction node Q2.

[0077] The interactive system provided in this embodiment can understand the user's scenario intention at the business interaction node Q2 through the AIGC gateway node Q1, and thus jump to the target business scenario corresponding to the scenario intention. In this way, the interactive system can capture the changes in the user's scenario intention in real time and flexibly, thereby switching the business scenario, so that the interactive system can provide effective interactive services in business scenarios that meet user needs, thereby improving the user experience.

[0078] Based on the same inventive concept as the above embodiments, see Figure 3 The following example illustrates an interactive system applied in a customer service environment for equipment, and uses a repair scenario to illustrate the technical solution of the aforementioned embodiment.

[0079] 1. Device support for interactive systems, including but not limited to: (1) Computer: used to run the Java virtual machine and configure PC programs; (2) The Java virtual machine is used to compile and run Java programs.

[0080] The interactive system includes AIGC gateway nodes, business interaction nodes and response nodes, and each of the aforementioned nodes can be configured through a Java virtual machine.

[0081] In one embodiment, see Figure 3 ,The business interaction nodes in the interactive system under the repair scenario include intent recognition nodes and large model nodes. The response nodes in the interactive system under the repair scenario include direct reply nodes, gateway routing nodes, content extraction nodes, and manual task nodes.

[0082] 2. Introduction to nodes of interactive system: 1. Start node: the node where the process starts; 2. AIGC gateway node: Through the AIGC gateway node, the data is forwarded to the corresponding business scenario according to the user's scenario intention, such as forwarding to the warranty scenario.

[0083] In one embodiment, the AIGC gateway node is one of the core nodes of the interactive system. It can implement generative artificial intelligence. In all subsequent nodes that interact with users (such as intent recognition nodes and large model nodes), it is necessary to pass through the AIGC gateway node by default to make business scenario judgment. If it is found that the user's business intention switches the current business scenario, it is necessary to dynamically and non-invasively switch to the new business scenario.

[0084] 3. Direct reply node: the node that directly replies to the customer; 4. Intent recognition node: In business scenarios, it identifies the user's intent based on the user's conversation; 5. Gateway routing node: performs gateway routing based on contextual conditions or recognition results of intent recognition nodes; 6. Large model node: a node that interacts with the large model; 7. Content extraction node: extract key information from the context; 8. Manual task node: a node that requires human participation; 9. End node: the node where the process ends.

[0085] Workflow of interactive system: 1. After the customer's intention to report a repair is identified through the AIGC gateway node, the repair scenario is entered; the process includes business scenarios such as repair reporting, consultation, business opportunities, and manual transfer.

[0086] 2. Reply directly to the node and send to the customer according to the configuration: Hello, please provide your device model.

[0087] 3. Intent recognition node, configure prompt words: Identify the customer's device type based on the customer's semantics.

[0088] When having a conversation with a customer, the default entry is from the AIGC gateway node. If the customer's scenario intention is found to have changed, the current business scenario is jumped out and the new business scenario is entered from the AIGC gateway node.

[0089] 4. Gateway routing node, configured with routing conditions: use the device type as a condition to transfer to different repair processing flows (device type A transfers to repair 1, device type B transfers to repair 2, and repair 2 has a manual task node); Repair 1: 5. Large model node, prompt words: Start a conversation with the customer by arranging prompt words, for example, Hello, what is the number of your device? When starting a conversation with a customer, the default entry is from the AIGC gateway node. If the customer's scenario intention is found to have changed, the current business scenario is jumped out and the new business scenario is entered from the AIGC gateway node. 6. Content extraction node, configured with prompt words: extract key information from the context through prompt words, such as extracting device numbers, fault imagination, etc.

[0090] In addition, it is also equipped with a plug-in (API) to store the extracted key information into the business system to generate documents; 7. Directly reply to the node and send a message to the customer based on the configuration: Thank you for your cooperation, your repair request has been accepted.

[0091] Repair 1: 8. Manual task node, used to generate a to-do item for business stakeholders; 9. Directly reply to the node and send it to the customer according to the configuration: Thank you for your cooperation. Your repair request has been accepted and the engineer will contact you directly later. This example can provide a document entry solution based on patterns, rules and AI big models through an interactive system.

[0092] The existing business orchestration methods of interactive services include prompt word orchestration and process orchestration. The disadvantage of prompt word orchestration is that when the business logic is more complex (for example, the business scenario changes), the accuracy of large model recognition will drop significantly, and there may be semantic similarity (i.e., similar meaning) or confusion between the various parts of the business process of different business scenarios, resulting in unsatisfactory results (frequent recognition errors). However, when the interactive service is implemented through the interactive system of this example, the AIGC gateway node can be used to capture the changes in the user's scene intentions in real time and flexibly, thereby switching the business scenario, thereby ensuring the accuracy of large model recognition, and different business scenarios can be used to distinguish semantically similar (i.e., similar meaning) or confused user dialogues, which can effectively understand the user's dialogue intentions and improve the interaction effect. Therefore, the technical solution of this example can overcome the defects of the above-mentioned prior art. In addition, the disadvantages of the existing process orchestration (stateless): it is only suitable for one-time transactional operations, and cannot meet the orchestration of multiple rounds of dialogues and more complex business processes (i.e., multiple dialogues or requests to complete a business together). At the same time, it cannot meet the changes in customer intentions in real time and dynamically during the business orchestration process. However, when the interactive service is implemented through the interactive system of this example, it is possible to complete a business (such as a repair service) in combination with the context of the conversation, and the AIGC gateway node can capture the changes in the user's scene intentions in real time and flexibly, thereby switching the business scene to effectively understand the user's conversation intentions and improve the interaction effect. Therefore, the technical solution of this example can overcome the above-mentioned defects of the prior art.

[0093] The technical solution of this example has the following technical effects: 1. Combined with the big model, design a stateful process orchestration engine (combined with the conversation context) to adapt to the orchestration of multi-round conversations and multi-step complex business scenarios.

[0094] 2. In stateful process orchestration, AIGC gateway nodes are designed to be able to capture the user's changing scenario intentions in real time and flexibly at any time.

[0095] 3. Optimize the stateless process orchestration based on AIGC technology currently available on the market. For example, adjust the stateless process orchestration on the market from the original start node, conditional branch, direct reply, large model node, and question classifier to: start node, AIGC gateway node, manual task node, gateway routing node, direct reply node, intent recognition node, content intelligent extraction node, large model node, and end node. To better adapt to the complex business scenarios of enterprises and expand the scope and breadth of AIGC technology implementation. In this way, this example provides a business orchestration solution based on the AIGC large model. By combining the AIGC large model, transforming the traditional process engine, and based on the characteristics of both for AIGC technology, the AIGC gateway node is added in a pioneering way to meet the various complex needs of AIGC technology implementation, break the limitations of AIGC technology implementation, and create higher value.

[0096] Furthermore, the technical solution of this example combines the AIGC big model to design a stateful process orchestration engine to adapt to multi-round dialogues and multi-step complex business scenario orchestration. The node information of a business request in the process can be stored, combined with the context of the dialogue with the AIGC big model, to realize the step-by-step flow of complex business processes.

[0097] Furthermore, the technical solution of this example designs an AIGC gateway node in the stateful process orchestration to be able to capture the user's changing intentions in real time and flexibly. In a complex business scenario, in the process of solving problems for users, users may switch their scene intentions at any time. Through the AIGC gateway node, when the user switches the scene intention, the system can know it, avoiding users from wasting time in the wrong interaction process, thereby effectively improving the user experience.

[0098] Based on the same inventive concept as the above embodiments, the present application embodiment provides a computing device, such as Figure 4 As shown, the device includes: a processor 310 and a memory 311 storing a computer program; wherein, Figure 4 The processor 310 shown in the figure is not used to indicate that the number of the processor 310 is one, but is only used to indicate the positional relationship of the processor 310 relative to other devices. In actual applications, the number of the processor 310 may be one or more; similarly, Figure 4 The memory 311 shown in the figure has the same meaning, that is, it is only used to refer to the position relationship of the memory 311 relative to other devices. In practical applications, the number of memories 311 can be one or more. When the processor 310 runs the computer program, the interactive method applied to the above device is implemented, for example, see Figure 1Step S11: Obtain first question information; Step S12: When the AIGC big model identifies that the target business scenario corresponding to the first question information is different from the current business scenario, generate and / or output interactive information corresponding to the first question information according to the response strategy corresponding to the target business scenario.

[0099] The device may also include: at least one network interface 312. The various components in the device are coupled together via a bus system 313. It is understood that the bus system 313 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 313 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, Figure 4 Various buses are labeled as bus system 313.

[0100] The memory 311 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory may be a disk memory or a tape memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAMbus random access memory (DRRAM, Direct Rambus Random Access Memory).The memory 311 described in the embodiments of the present application is intended to include but is not limited to these and any other suitable types of memories.

[0101] The memory 311 in the embodiment of the present application is used to store various types of data to support the operation of the device. Examples of these data include: any computer program used to operate on the device, such as an operating system and an application program. Among them, the operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and handle hardware-based tasks. The application program may include various applications, such as a media player (MediaPlayer), a browser (Browser), etc., which are used to implement various application services. Here, the program that implements the interactive method provided in the embodiment of the present application can be included in the application program.

[0102] Based on the same inventive concept as the above-mentioned embodiment, this embodiment further provides a computer-readable storage medium, in which a computer program is stored. The computer-readable storage medium may be a ferromagnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); it may also be various devices including one or any combination of the above-mentioned memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, etc. When the computer program stored in the computer-readable storage medium is executed by the processor, the interactive method provided in the above-mentioned embodiment is implemented. For the specific steps implemented when the computer program is executed by the processor, please refer to Figure 1 Step S11 shown: obtain first question information; step S12: when the AIGC big model identifies that the target business scenario corresponding to the first question information is different from the current business scenario, generate and / or output interactive information corresponding to the first question information according to the response strategy corresponding to the target business scenario.

[0103] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] In this document, the terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion of elements other than those listed and may also include additional elements not expressly listed.

[0105] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An interactive method, characterized in that: include: Get the first question information; When the AIGC big model identifies that the target business scenario corresponding to the first question information is different from the current business scenario, the interactive information corresponding to the first question information is generated and / or output according to the response strategy corresponding to the target business scenario.

2. The method according to claim 1, characterized in that The target business scenario includes one of the following: a repair scenario, a consultation scenario, a business opportunity scenario, and a manual transfer scenario.

3. The method according to claim 2, characterized in that The response strategy corresponding to the repair scenario includes at least one of the following: When it is identified that the first problem information corresponds to a repair intention, first interaction information is output, where the first interaction information is used to trigger a repair process; When it is identified that the first problem information corresponds to a manual entry intention of a work order, second interaction information is generated and output, where the second interaction information includes to-do tasks for the business party and / or reply information for the demand party; When it is identified that the first question information corresponds to the work order machine entry intention, third interaction information is generated and output, and the third interaction information is used to trigger the question and answer process of the work order entry.

4. The method according to claim 2, characterized in that: The response strategies corresponding to the consultation scenarios include: The answer information of the first question information is generated by the AIGC big model, and the answer information is output.

5. The method according to claim 1, characterized in that: After the step of generating and / or outputting interaction information corresponding to the first question information according to a response strategy corresponding to the target business scenario when the target business scenario corresponding to the first question information is identified by the AIGC big model as being different from the current business scenario, the method includes: Obtaining the second question information; When the AIGC big model identifies that the second question information corresponds to the target business scenario, obtaining question and answer context information corresponding to the target business scenario; The question-answer context information and the second question information are parsed by the AIGC big model to generate and output interaction information corresponding to the second question information.

6. The method according to any one of claims 1 to 5, characterized in that: The question information is natural language information; The step of generating and / or outputting interactive information corresponding to the question information comprises: Extracting emotion determination features from the question information, wherein the emotion determination features include a combination of one or more of voice pitch, voice rhythm, text vocabulary, and text sentence structure; Performing emotion determination according to the emotion determination feature to determine the current conversation emotion; Matching reply request information according to the current conversation emotion, wherein the reply request information includes at least one of a reply tone and a reply vocabulary; The answer information corresponding to the question information is output and controlled according to the answer requirement information.

7. The method according to claim 6, characterized in that The step of performing emotion determination according to the emotion determination feature to determine the current dialogue emotion includes: Obtaining question and answer context information corresponding to the target business scenario; Determining historical conversation emotions based on the question-answer context information; Emotion determination is performed according to the emotion determination feature and the historical dialogue emotion to determine the current dialogue emotion.

8. An interactive system, characterized in that: include: AIGC gateway node, business interaction node and response node; The AIGC gateway node is used to implement the interaction method according to any one of claims 1 to 7; The service interaction node is used to call the AIGC gateway node to identify the target service scenario corresponding to the received service problem information, and when the target service scenario is different from the current service scenario corresponding to the service interaction node, enter the target service scenario through the AIGC gateway node, or, when the target service scenario is the same as the current service scenario corresponding to the service interaction node, conduct a service dialogue with the service interaction logic corresponding to the service interaction node; The response node is used to perform a service response according to the service dialogue result fed back by the service interaction node.

9. A computing device, characterized in that include: A processor and a memory storing a computer program, wherein when the processor runs the computer program, the steps of the interaction method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the steps of the interactive method according to any one of claims 1 to 7 are implemented.

Images