Intelligent customer service system and method based on multi-agent collaboration and electronic equipment

Abstract

The invention discloses an intelligent customer service system and method based on multi-agent cooperation and electronic equipment, and belongs to the technical field of computers. The system comprises a supervisor Agent, a processing Agent and an answering Agent, the supervisor Agent is used for acquiring multi-modal information input by a user, after analyzing the multi-modal information, assigning a target task to different processing Agents, calling a corresponding answer Agent according to a target task processing result fed back by the processing Agents, receiving a processing result of the answer Agent, and generating a multi-modal reply; the processing Agent is used for processing the target task and feeding back a processing result of the target task to the supervisor Agent; and the answer Agent is used for calling the corresponding capability and tool according to the calling instruction of the main Agent to obtain a processing result, and feeding back the processing result to the main Agent. According to the system, the problems of ordering, using and complaining of multiple products can be efficiently solved, independent and cooperative work of the intelligent agents is achieved through the intelligent agent center control, and the customer satisfaction is improved.

Description

Technical Field

[0001] This application belongs to the field of computer technology, and particularly relates to an intelligent customer service system, method, and electronic device based on multi-agent collaboration. Background Art

[0002] Current intelligent customer service systems usually use a single model or independent agents to handle user problems, and it is difficult to cope with the complex requirements of multiple products and scenarios. For example, when a user orders product A, payment issues may be involved, and when using product B, technical failures may be encountered. A single model cannot handle these cross-product and cross-domain problems simultaneously, resulting in a poor user experience. In addition, traditional systems lack an agent collaboration mechanism and cannot dynamically allocate tasks according to user requests, leading to low processing efficiency.

[0003] In view of the above problems, this application proposes an intelligent customer service system, method, and electronic device based on multi-agent collaboration. Summary of the Invention

[0004] To solve the deficiencies of the existing technology, this application provides an intelligent customer service system, method, and electronic device based on multi-agent collaboration, which solves the problems in the existing technology such as the lack of an agent collaboration mechanism and the inability to dynamically allocate tasks according to user requests, resulting in low processing efficiency.

[0005] The technical effects to be achieved by this application are realized through the following solutions: In a first aspect, this application provides an intelligent customer service system based on multi-agent collaboration, which includes a supervisor Agent, a processing Agent, and an answering Agent; The supervisor Agent is used to obtain the multi-modal information input by the user, after parsing the multi-modal information, dispatch the target task to different processing Agents, call the corresponding answering Agent according to the processing result of the target task feedback by the processing Agent, receive the processing result of the answering Agent, and generate a multi-modal reply; The processing Agent is used to process the target task and feedback the processing result of the target task to the supervisor Agent; The answering Agent is used to call the corresponding capabilities and tools according to the call instruction of the supervisor Agent, obtain the processing result, and feedback the processing result to the supervisor Agent.

[0006] In some embodiments, the supervisor Agent includes a multi-modal information receiver, an information parser, a task dispatcher, a simple question answerer, and a multi-modal generator; where The multi-modal information receiver is used to receive the multi-modal information input by the user; The information parser is used to parse multi-modal information, extract key questions, requirements, and features, and obtain a parsing result; The task dispatcher is used to assign target tasks to corresponding processing Agents according to the parsing result; The simple question answerer is used to directly give answers to conventional questions; The multi-modal generator is used to generate a multi-modal response according to the processing result provided by the answering Agent.

[0007] In some embodiments, the processing Agents include: Text processing Agent, image processing Agent, video processing Agent, and file processing Agent.

[0008] In some embodiments, the answering Agents include: Product ordering Agent, product usage Agent, and product after-sales Agent.

[0009] In a second aspect, the present application provides an intelligent customer service method based on multi-agent collaboration. The method is executed by the intelligent customer service system based on multi-agent collaboration described in any one of the above. The method includes: The supervisor Agent obtains the multi-modal information input by the user, parses the multi-modal information, assigns target tasks to different processing Agents, calls the corresponding answering Agents according to the target task processing results fed back by the processing Agents, receives the processing results of the answering Agents, and generates a multi-modal response; The processing Agent processes the target task and feeds back the target task processing result to the supervisor Agent; The answering Agent calls the corresponding capabilities and tools according to the call instruction of the supervisor Agent, obtains the processing result, and feeds back the real-time processing result to the supervisor Agent.

[0010] In some embodiments, the supervisor Agent obtains the multi-modal information input by the user, parses the multi-modal information, assigns target tasks to different processing Agents, calls the corresponding answering Agents according to the target task processing results fed back by the processing Agents, receives the processing results of the answering Agents, and generates a multi-modal response, including: The multi-modal information receiver of the supervisor Agent receives the multi-modal information input by the user; The information parser of the supervisor Agent parses the multi-modal information, extracts key questions, requirements, and features, and obtains a parsing result; The task dispatcher of the supervisor Agent assigns the target task to the corresponding processing Agent according to the parsing result, receives the processing result of the target task fed back by the processing Agent, calls the corresponding answering Agent according to the processing result of the target task, and receives the processing result of the answering Agent; The multi-modal generator of the supervisor Agent is used to generate a multi-modal reply according to the processing result provided by the answering Agent.

[0011] In some embodiments, after the information parser of the supervisor Agent obtains the parsing result, it determines whether it belongs to a regular question based on the parsing result. If it belongs to a regular question, the simple question solver of the supervisor Agent directly gives an answer to the regular question; if it does not belong to a regular question, the parsing result is sent to the task dispatcher of the supervisor Agent for the assignment of the target task.

[0012] In some embodiments, the processing Agent includes: a text processing Agent, a picture processing Agent, a video processing Agent, and a file processing Agent; the answering Agent includes: a product ordering Agent, a product using Agent, and a product after-sales Agent.

[0013] In some embodiments, the multi-modal information includes: text, pictures, videos, and files.

[0014] In a third aspect, the present application provides an electronic device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method described in any one of the foregoing is implemented.

[0015] Through the intelligent customer service system, method, and electronic device based on multi-agent collaboration provided by the present application, the system can efficiently process multi-product ordering, use, and complaint problems, and realize the independent and collaborative work of the agents through the agent central control, improving customer satisfaction. Description of the Drawings

[0016] In order to more clearly illustrate the embodiments of the present application or the existing technical solutions, the following will briefly introduce the drawings required for the description of the embodiments or the existing technical solutions. Obviously, the drawings in the following description are only some embodiments recorded in the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 It is a schematic structural diagram of an intelligent customer service system based on multi-agent collaboration in an embodiment of the present application; Figure 2 A schematic block diagram of an electronic device in an embodiment of the present application. Detailed implementation manners

[0018] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part rather than all of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the scope of protection of the present application.

[0019] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present application should have the ordinary meanings understood by those of ordinary skill in the art to which the present application pertains. The terms "first", "second" and similar terms used in one or more embodiments of the present application do not denote any order, quantity or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this term cover the elements or objects listed after this term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left" and "right" are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0020] Currently, there are many related technologies involving intelligent agent customer service, such as: A multi-agent integrated customer service method and system with the patent publication number of CN119359316A, which relates to a multi-agent integrated customer service method and system. The method includes: processing the consultation questions raised by the actively incoming users through the dialogue management agent to obtain multi-dimensional data information and generating a dialogue response to the consultation questions based on this; analyzing the consultation intention of the actively incoming users through the intention recognition agent based on the feedback of the actively incoming users after the dialogue response; determining the outgoing operations to be executed through the outgoing decision agent based on the consultation intention of the actively incoming users and their recent activity records; and then executing the outgoing operations through the outgoing execution agent. Through the present application, the integrated customer service of multiple agents is realized, and the collaborative division of labor of multiple different agents is achieved, each focusing on its specific business process, so as to provide better services for customers. Moreover, it can also actively call and contact customers by analyzing customer data, track customer problems or conduct active marketing, effectively improving customer satisfaction.

[0021] An intelligent customer service Q&A method based on large language model technology with the patent publication number CN118364084A, which relates to the technical field of customer service. Specifically, it discloses an intelligent customer service Q&A method based on large language model technology, including: preprocessing the current question text information of the user; training the constructed intelligent customer service model to obtain the trained intelligent customer service model; judging the type of the preprocessed current question text information; if the current question is a modal particle or a related question, then combine the corresponding prompt words and chat records and input them into the trained intelligent customer service model to obtain the reply content; if the current question is a new topic, then combine the corresponding prompt words and input them into the trained intelligent customer service model to obtain the reply content; perform security verification on the reply content to output the verified reply content. The intelligent customer service Q&A method based on large language model technology provided by the present invention can improve the intelligence and accuracy of the answers and improve the application scenarios and user experience of intelligent customer service Q&A.

[0022] An online Q&A prompt word optimization generation method based on large language model with the patent publication number CN118093836A, which is applicable to the technical field of large language models, and provides a large language model prompt word generation method, system, terminal device and medium. It includes constructing a personalized corpus based on user data; obtaining the prompt words input by the user and splitting the prompt words; determining the root node in the core words and determining the relationship between this core word and other unit texts according to the syntactic relationship; constructing a prompt word parsing tree according to the root node and the relationship; calculating the evolution probability and node depth probability, evolving the prompt word parsing tree to obtain the final prompt word parsing tree; generating multiple new prompt words according to the final prompt word parsing tree, calculating the comprehensive score of each new prompt word based on the personalized corpus and the pre-set professional field corpus, and outputting the new prompt word corresponding to the highest comprehensive score to the user.

[0023] The related intelligent agents similar to the above technologies have the following problems: 1) Limitations of a single model: It is difficult for a single model to handle the complex requirements of multiple products and multiple scenarios simultaneously, resulting in poor performance of the system when dealing with cross-product and cross-domain problems.

[0024] Lack of coordination mechanism: The existing system lacks an intelligent agent central control mechanism and cannot dynamically allocate tasks according to user demands, resulting in low processing efficiency.

[0025] 2) Lack of goal orientation: The existing system lacks a goal-oriented coordination mechanism and cannot dynamically adjust the behavior of intelligent agents around a common goal (such as improving customer satisfaction), resulting in low customer satisfaction.

[0026] 3) Limited ability to handle complex problems: When dealing with complex problems, the existing system often relies on a single model and rule engine, lacking the ability of multi-agent collaborative processing, resulting in low problem-solving efficiency.

[0027] This application significantly improves the processing ability of the intelligent customer service system for multi-product ordering, usage, and complaint problems by introducing a multi-agent collaboration mechanism and an agent central control module. The specific improvements are as follows: 1) Multi-agent collaboration: Through multiple independent and interrelated agents, the system can efficiently handle problems related to the ordering, usage, and complaint of different products.

[0028] 2) Agent central control mechanism: The agent central control module dynamically assigns tasks according to the user's requests, ensuring that problems are handled by the most suitable agent and improving the processing efficiency.

[0029] 3) Goal-oriented collaboration: All agents have the common goal of improving customer satisfaction and dynamically adjust their behaviors to ensure that customer problems are satisfactorily solved.

[0030] 4) Ability to handle complex problems: Through complex problem collaborative agents, the system can efficiently solve complex problems and improve customer satisfaction.

[0031] The following will detail various non-restrictive implementation manners of this application with reference to the accompanying drawings.

[0032] First, with reference to Figure 1 , a detailed description will be given of the intelligent customer service system based on multi-agent collaboration of this application.

[0033] This application provides an intelligent customer service system based on multi-agent collaboration, which includes a supervisor agent, a processing agent, and an answering agent. The supervisor agent is used to obtain the multi-modal information input by the user, parse the multi-modal information, dispatch target tasks to different processing agents, call the corresponding answering agent according to the processing results of the target tasks fed back by the processing agents, receive the processing results of the answering agent, and generate a multi-modal reply. The processing agent is used to process the target tasks and feed back the processing results of the target tasks to the supervisor agent. The answering agent is used to call the corresponding capabilities and tools according to the call instructions of the supervisor agent, obtain the processing results, and feed back the processing results to the supervisor agent.

[0034] The intelligent customer service system based on multi-agent collaboration provided by this application can efficiently handle problems related to multi-product ordering, usage, and complaint, and through agent central control, realize the independent and collaborative work of agents, improving customer satisfaction.

[0035] In some embodiments, the supervisor Agent includes a multimodal information receiver, an information parser, a task dispatcher, a simple question answerer, and a multimodal generator; among which, The multimodal information receiver is used to receive the multimodal information input by the user; The information parser is used to parse the multimodal information, extract key questions, requirements, and features, and obtain an analysis result; The task dispatcher is used to assign the target task to the corresponding processing Agent according to the analysis result; The simple question answerer is used to directly give an answer to a general question; The multimodal generator is used to generate a multimodal reply according to the processing result provided by the answering Agent.

[0036] In some embodiments, the processing Agent includes: A text processing Agent, an image processing Agent, a video processing Agent, and a file processing Agent.

[0037] Exemplarily, the text processing Agent is responsible for the following functions: Specifically handle text-related tasks, such as text understanding, text summarization, text translation, text generation, etc. The team may include Agent members with different specialties. For example, the text understanding expert Agent is responsible for deeply understanding the text semantics, and the text generation Agent is responsible for generating new text content according to requirements.

[0038] Exemplarily, the tools of the text processing Agent include: • A text parsing tool: perform lexical, syntactic, and semantic analysis on the input text.

[0039] • A text summarization tool: generate a concise summary of the text.

[0040] • A text translation tool: implement text translation between different languages.

[0041] • A text generation tool: generate new text according to a given topic, template, or instruction.

[0042] Exemplarily, the image processing Agent is responsible for the following functions: Focus on image-related tasks, such as image classification, object detection, image annotation, etc. The team members include an image classification Agent for identifying the image category, and an object detection Agent responsible for detecting specific target objects in the image.

[0043] Exemplarily, the tools of the image processing Agent include: • Image classifier: Classifies the input image and determines its category.

[0044] • Object detector: Detects specific target objects in the image.

[0045] • Image annotation tool: Annotates objects or regions in the image.

[0046] Exemplarily, the video processing Agent is responsible for the following functions: Responsible for analyzing video content, extracting key information such as scenes and objects in the video. There is a video analysis Agent in the team for extracting key video information.

[0047] Exemplarily, the tools of the video processing Agent include: • Video content analysis tool: Analyzes video content, extracts key frames, audio information, scene changes, etc.

[0048] • Video object recognition tool: Recognizes objects in the video.

[0049] • Video action recognition tool: Recognizes actions of people or objects in the video.

[0050] Exemplarily, the file processing Agent is responsible for the following functions: Responsible for processing various file types, such as document files (Word, PDF, etc.), spreadsheet files (Excel), etc. Team members include a document parsing Agent for extracting text and format information from documents, and a spreadsheet processing Agent for analyzing and processing spreadsheet data.

[0051] Exemplarily, the tools of the file processing Agent include: • Document parsing tool: Reads and parses document files, extracts text content, format information, etc.

[0052] • Spreadsheet processing tool: Performs operations such as data extraction, analysis, and calculation on spreadsheet files.

[0053] • File conversion tool: Converts between different file formats.

[0054] In some embodiments, the answering Agent includes: Product ordering Agent, product usage Agent, and product after-sales Agent.

[0055] Exemplarily, the answering Agent is mainly responsible for the following functions: According to the instructions of the supervisor Agent, call the capabilities and tools of each group Agent, integrate the processing results, form the answer content for specific problems (such as product ordering, product use, product after-sales, etc.), and return it to the supervisor Agent.

[0056] The tools of the answering Agent include: • Result integrator: Integrate and optimize the processing results of each group Agent.

[0057] • Information coordinator: Transmit and coordinate information among multiple group Agents to ensure the smoothness of the answer processing flow.

[0058] This application significantly improves the processing ability of the intelligent customer service system for multi-product ordering, use, and after-sales complaint problems by introducing a multi-agent collaboration mechanism, and aims to improve customer satisfaction, with broad application prospects.

[0059] In a second aspect, this application provides an intelligent customer service method based on multi-agent collaboration, which is executed by the intelligent customer service system based on multi-agent collaboration described in any one of the above. The method includes: The supervisor Agent obtains the multi-modal information input by the user, after parsing the multi-modal information, assigns target tasks to different processing Agents, calls the corresponding answering Agent according to the target task processing results fed back by the processing Agents, receives the processing results of the answering Agent, and generates a multi-modal reply; The processing Agent processes the target task and feeds back the target task processing results to the supervisor Agent; The answering Agent calls the corresponding capabilities and tools according to the call instructions of the supervisor Agent, obtains the processing results, and feeds back the real-time processing results to the supervisor Agent.

[0060] In some embodiments, the supervisor Agent obtains the multi-modal information input by the user, after parsing the multi-modal information, assigns target tasks to different processing Agents, calls the corresponding answering Agent according to the target task processing results fed back by the processing Agents, receives the processing results of the answering Agent, and generates a multi-modal reply, including: The multi-modal information receiver of the supervisor Agent receives the multi-modal information input by the user; The information parser of the supervisor Agent parses the multi-modal information, extracts key problems, requirements, and features, and obtains the parsing result; Based on the parsing result, the task dispatcher of the supervisor Agent assigns the target task to the corresponding processing Agent, receives the processing result of the target task fed back by the processing Agent, calls the corresponding answering Agent according to the processing result of the target task, and receives the processing result of the answering Agent; The multi-modal generator of the supervisor Agent is used to generate a multi-modal response according to the processing result provided by the answering Agent.

[0061] In some embodiments, after the information parser of the supervisor Agent obtains the parsing result, it determines whether it belongs to a routine question based on the parsing result. If it belongs to a routine question, the simple question solver of the supervisor Agent directly gives an answer to the routine question; if it does not belong to a routine question, the parsing result is sent to the task dispatcher of the supervisor Agent for the assignment of the target task.

[0062] In some embodiments, the processing Agents include: a text processing Agent, a picture processing Agent, a video processing Agent, and a file processing Agent; the answering Agents include: a product ordering Agent, a product using Agent, and a product after-sales Agent. The answering Agents here are exemplary, and more corresponding Agents can also be determined according to actual needs.

[0063] In some embodiments, the multi-modal information includes: text, pictures, videos, and files.

[0064] The Agent in this application refers to an intelligent agent.

[0065] Exemplarily, the collaborative processing flow steps of the multi-Agent hierarchical structure: 1. User input: The user inputs multi-modal information such as text, pictures, videos, and files through the multi-modal information receiver of the supervisor Agent, that is, the supervisor Agent obtains the input.

[0066] 2. Information parsing: The supervisor Agent uses the information parser to comprehensively parse the input multi-modal information, extract the key questions, user requirements, and feature information related to each modality.

[0067] 3. Simple question judgment: The supervisor Agent determines whether the question is a routine question (what kind of question). If it is, it directly gives an answer and feeds it back to the user; if not, it proceeds to the next step.

[0068] 4. Task Assignment: Based on the analysis results, the supervisor Agent accurately assigns the target tasks to the corresponding processing Agents through the task dispatcher, such as text processing Agents, image processing Agents, etc.

[0069] 5. Processing Agents: After receiving the target tasks, each processing Agent processes the tasks using its own tools and capabilities. For example, the text processing Agent performs text analysis and generation, and the image processing Agent team performs image recognition and annotation, etc.

[0070] 6. Answer Agent Invocation and Integration: The supervisor Agent invokes the appropriate answer Agent (product ordering agent, product usage agent, product after-sales agent, or other relevant Agent, etc.) according to the question type (such as product ordering, product usage, product after-sales, or other related questions, etc.). The answer Agent collects the processing results of each processing Agent and uses the result integrator and information coordinator for integration and optimization to form a complete answer content.

[0071] 7. Multimodal Response Generation: After receiving the content returned by the answer Agent, the supervisor Agent invokes the multimodal generator and combines the modal information input by the user to generate the corresponding multimodal response content, such as text, picture display, video demonstration, etc.

[0072] 8. Respond to the User: The supervisor Agent feeds back the generated multimodal response content to the user.

[0073] The intelligent customer service system based on multi-agent collaboration of this application has the following advantages: 1. Supervisor Agent Assigns Tasks: The supervisor intelligent agent, as the intelligent agent central control module, analyzes the user's demands and determines the domain to which the problem belongs. According to the question type, tasks are dynamically assigned to the corresponding intelligent agents. For example, if the user asks about the ordering problem of product A, the product ordering intelligent agent is assigned to handle it; if the user feedbacks the usage problem of product A, the product usage intelligent agent is assigned to handle it; if the user complains about product C, the after-sales intelligent agent is assigned to handle it, improving the processing efficiency.

[0074] 2. Multi-agent Independent and Collaborative Processing: Each intelligent agent independently processes the assigned tasks. For example, the product ordering intelligent agent processes payment problems, the product usage intelligent agent solves technical failures, and the complaint handling intelligent agent processes after-sales problems.

[0075] For complex problems involving multiple products, the intelligent agent central control module invokes the complex problem collaborative intelligent agent to coordinate relevant intelligent agents to jointly solve them.

[0076] 3. Large model support: When each agent processes tasks, it calls the large model support module to obtain necessary knowledge and computing resources.

[0077] 4. Multi-product support: Through multiple independent agents, the system can efficiently handle product ordering, usage, and complaint issues, meeting multi-scenario requirements.

[0078] 5. Goal-oriented collaboration: All agents take improving customer satisfaction as a common goal and dynamically adjust their behaviors to ensure that customer problems are satisfactorily solved.

[0079] 6. User experience improvement: The system can quickly and accurately solve user problems, provide personalized services, and significantly improve the user experience.

[0080] It should be noted that the method of one or more embodiments of this application can be executed by a single device, such as a computer or a server. The method of this embodiment can also be applied to a distributed scenario, where multiple devices cooperate with each other to complete it. In this case of a distributed scenario, one of these multiple devices can only execute one or more steps of the method of one or more embodiments of this application, and these multiple devices will interact with each other to complete the described method.

[0081] It should be noted that the above specifically describes certain embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be executed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the specific order or sequential order shown to achieve the desired result. In certain implementations, multitasking and parallel processing are also possible or may be advantageous.

[0082] Based on the same inventive concept, corresponding to the method of any of the above embodiments, this application also discloses an electronic device; Specifically, Figure 2 FIG. shows a schematic hardware structure diagram of an electronic device for an intelligent customer service method based on multi-agent collaboration provided in this embodiment. The device may include: a processor 410, a memory 420, an input / output interface 430, a communication interface 440, and a bus 450. Among them, the processor 410, the memory 420, the input / output interface 430, and the communication interface 440 are communicatively connected to each other inside the device through the bus 450.

[0083] The processor 410 can be implemented in the form of a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0084] The memory 420 can be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, etc. The memory 420 can store an operating system and other application programs. When implementing the technical solutions provided in the embodiments of the present application through software or firmware, the relevant program codes are stored in the memory 420 and are called and executed by the processor 410.

[0085] The input / output interface 430 is used to connect to the input / output module to implement information input and output. The input / output module can be configured as a component in the device (not shown in the figure) or can be externally connected to the device to provide corresponding functions. Among them, the input device can include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device can include a display, a speaker, a vibrator, an indicator light, etc.

[0086] The communication interface 440 is used to connect to a communication module (not shown in the figure) to implement communication interaction between this device and other devices. Among them, the communication module can implement communication in a wired manner (for example, USB, network cable, etc.) or can implement communication in a wireless manner (for example, mobile network, WIFI, Bluetooth, etc.).

[0087] The bus 450 includes a path for transmitting information between various components of the device (for example, the processor 410, the memory 420, the input / output interface 430, and the communication interface 440).

[0088] It should be noted that although the above device only shows the processor 410, the memory 420, the input / output interface 430, the communication interface 440, and the bus 450, in the specific implementation process, this device may also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device may also only include the components necessary to implement the solutions of the embodiments of the present application, and do not have to include all the components shown in the figure.

[0089] The electronic device in the above embodiment is used to implement the corresponding intelligent customer service method based on multi-agent collaboration in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0090] Based on the same inventive concept, corresponding to the method of any of the above embodiments, one or more embodiments of the present application further provide a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the multi-agent collaborative-based intelligent customer service method as described in any of the above embodiments.

[0091] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0092] The computer instructions stored in the storage medium of the above embodiment are used to cause the computer to execute the multi-agent collaborative-based intelligent customer service method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0093] Those of ordinary skill in the art should understand that the discussion of any of the above embodiments is only exemplary and is not intended to imply that the scope of the present application (including the claims) is limited to these examples; under the concept of the present application, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations in different aspects of one or more embodiments of the present application as described above, which are not provided in detail for the sake of brevity.

[0094] Additionally, for simplicity of explanation and discussion, and so as not to render one or more embodiments of the present application difficult to understand, well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Further, the devices may be shown in block diagram form to avoid rendering one or more embodiments of the present application difficult to understand, and this also takes into account the fact that details regarding the implementation of these block diagram devices are highly dependent on the platform on which one or more embodiments of the present application are to be implemented (i.e., these details should be entirely within the understanding of those skilled in the art). In cases where specific details (such as circuits) are set forth to describe exemplary embodiments of the present application, it will be apparent to those skilled in the art that one or more embodiments of the present application may be implemented without these specific details or with variations of these specific details. Accordingly, these descriptions should be considered illustrative rather than restrictive.

[0095] Although the present application has been described in connection with specific embodiments thereof, many alternatives, modifications, and variations will be apparent to those of ordinary skill in the art based on the foregoing description. For example, other memory architectures (such as dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0096] One or more embodiments of the present application are intended to cover all such alternatives, modifications, and variations that fall within the broad scope of the appended claims. Accordingly, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of the present application shall be included within the scope of protection of the present application.

Claims

1. An intelligent customer service system based on multi-agent collaboration, characterized in that, The system includes a supervisor Agent, a processing Agent, and an answering Agent; The supervisor Agent is used to obtain the multimodal information input by the user. After parsing the multimodal information, it dispatches target tasks to different processing Agents, calls the corresponding answering Agent according to the processing results of the target tasks fed back by the processing Agents, receives the processing results of the answering Agent, and generates a multimodal reply; The processing Agent is used to process the target tasks and feed back the processing results of the target tasks to the supervisor Agent; The answering Agent is used to call the corresponding capabilities and tools according to the call instructions of the supervisor Agent, obtain the processing results, and feed back the processing results to the supervisor Agent.

2. The intelligent customer service system based on multi-agent collaboration according to claim 1, wherein The supervisor Agent includes a multimodal information receiver, an information parser, a task dispatcher, a simple question solver, and a multimodal generator; among them, The multimodal information receiver is used to receive the multimodal information input by the user; The information parser is used to parse the multimodal information, extract key questions, requirements, and features, and obtain the parsing results; The task dispatcher is used to dispatch the target tasks to the corresponding processing Agents according to the parsing results; The simple question solver is used to directly give answers to routine questions; The multimodal generator is used to generate a multimodal reply according to the processing results provided by the answering Agent.

3. The intelligent customer service system based on multi-agent collaboration according to claim 1 or 2, characterized in that The processing Agent includes: A text processing Agent, a picture processing Agent, a video processing Agent, and a file processing Agent.

4. The intelligent customer service system based on multi-agent collaboration according to claim 3, wherein The answering Agent includes: A product ordering Agent, a product usage Agent, and a product after-sales Agent.

5. An intelligent customer service method based on multi-agent collaboration, characterized in that, The method is executed by the intelligent customer service system based on multi-agent collaboration described in any one of claims 1-4. The method includes: The supervisor Agent obtains the multimodal information input by the user. After parsing the multimodal information, it dispatches target tasks to different processing Agents, calls the corresponding answering Agent according to the processing results of the target tasks fed back by the processing Agents, receives the processing results of the answering Agent, and generates a multimodal reply; The processing Agent processes the target tasks and feeds back the processing results of the target tasks to the supervisor Agent; The answering Agent calls the corresponding capabilities and tools according to the call instructions of the supervisor Agent, obtains the processing results, and feeds back the real-time processing results to the supervisor Agent.

6. The intelligent customer service method based on multi-agent collaboration according to claim 5, characterized in that, The supervisor Agent obtains the multimodal information input by the user. After parsing the multimodal information, it dispatches target tasks to different processing Agents, calls the corresponding answering Agent according to the processing results of the target tasks fed back by the processing Agents, receives the processing results of the answering Agent, and generates a multimodal reply, including: The multimodal information receiver of the supervisor Agent receives the multimodal information input by the user; The information parser of the supervisor Agent parses the multimodal information, extracts key questions, requirements, and features, and obtains the parsing results; Based on the parsing result, the task dispatcher of the supervisor Agent assigns the target task to the corresponding processing Agent, receives the processing result of the target task fed back by the processing Agent, calls the corresponding answering Agent according to the processing result of the target task, and receives the processing result of the answering Agent; The multi-modal generator of the supervisor Agent is used to generate a multi-modal reply according to the processing result provided by the answering Agent.

7. The intelligent customer service method based on multi-agent collaboration according to claim 5, wherein After the information parser of the supervisor Agent obtains the parsing result, it determines whether it belongs to a routine question based on the parsing result. If it belongs to a routine question, the simple question solver of the supervisor Agent directly gives an answer to the routine question; If it does not belong to a routine question, the parsing result is sent to the task dispatcher of the supervisor Agent for dispatching the target task.

8. The intelligent customer service method based on multi-agent collaboration according to claim 7, wherein The processing Agent includes: a text processing Agent, a picture processing Agent, a video processing Agent, and a file processing Agent; the answering Agent includes: a product ordering Agent, a product using Agent, and a product after-sales Agent.

9. The intelligent customer service method based on multi-agent collaboration according to claim 5, characterized in that, The multi-modal information includes: text, pictures, videos, and files.

10. An electronic device, the electronic device comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, the method described in any one of claims 5 to 9 is implemented.

Images