Service request processing method and system, electronic equipment and computer program product

Abstract

The invention discloses a service request processing method and system, electronic equipment and a computer program product, and relates to the field of artificial intelligence technology and service management. The method comprises the following steps: in response to a service request for a target object, determining at least one target service scene corresponding to the service request; a first target intelligent module corresponding to the target service scene is determined from the multiple first intelligent modules, and different first intelligent modules are used for processing service requests corresponding to different service scenes; processing the service request by using the first target intelligent module to obtain a first processing result corresponding to the first target intelligent module; and generating a service processing response corresponding to the service request by using the management intelligent module and the first processing result. The technical problem that the processing effect of the service request is poor in the related technology is solved.

Description

Technical Field

[0001] This application relates to the fields of artificial intelligence technology and service management, and more specifically, to a method, system, electronic device, and computer program product for processing service requests. Background Technology

[0002] Currently, e-commerce platforms face challenges due to the diversity and complexity of user service requests, requiring them to both follow standardized processes and demonstrate personalized care in their service processing and response.

[0003] In related technologies, attempts are made to balance standardized and personalized services by deploying a single, large, general-purpose intelligent module (Agent) or by using keyword-based static routing to allocate tasks. However, these methods not only struggle to capture subtle user needs but also often exhibit low accuracy and efficiency when handling atypical problems, particularly in processing multi-turn interactions and complex service requests, failing to provide satisfactory service processing responses. Therefore, the technical problem of poor service request processing performance remains.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a method, system, electronic device, and computer program product for processing service requests, in order to at least solve the technical problem of poor processing effect of service requests in related technologies.

[0006] According to one aspect of the embodiments of this application, a method for processing service requests is provided, comprising: in response to a service request for a target object, determining at least one target service scenario corresponding to the service request; determining a first target intelligent module corresponding to the target service scenario from a plurality of first intelligent modules, wherein different first intelligent modules are used to process service requests corresponding to different service scenarios; processing the service request using the first target intelligent module to obtain a first processing result corresponding to the first target intelligent module; and generating a service processing response corresponding to the service request using a management intelligent module and the first processing result.

[0007] According to another aspect of the embodiments of this application, a service request processing apparatus is provided, comprising: a first determining module, configured to determine at least one target service scenario corresponding to a service request in response to a service request for a target object; a second determining module, configured to determine a first target intelligent module corresponding to the target service scenario from a plurality of first intelligent modules, wherein different first intelligent modules are used to process service requests corresponding to different service scenarios; a processing module, configured to process the service request using the first target intelligent module to obtain a first processing result corresponding to the first target intelligent module; and a generating module, configured to generate a service processing response corresponding to the service request using a management intelligent module and the first processing result.

[0008] According to another aspect of the embodiments of this application, a service request processing system is provided, comprising: a scenario routing module, configured to, in response to a service request for a target object, determine at least one target service scenario corresponding to the service request, and determine a first target intelligent module corresponding to the target service scenario from a plurality of first intelligent modules, wherein different first intelligent modules are used to process service requests corresponding to different service scenarios; and a processing module, configured to call the first target intelligent module to process the service request, obtain a first processing result corresponding to the first target intelligent module, and call a management intelligent module to fuse the first processing result to generate a service processing response corresponding to the service request.

[0009] According to another aspect of the embodiments of this application, a computing device is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0010] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory storing an executable program; and a processor connected to the memory via a bus for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0011] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0012] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0013] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0014] According to another aspect of the embodiments of this application, a computer program is also provided, which, when executed by a processor, implements the methods of the various embodiments of this application.

[0015] In this embodiment, if a service request for a target object is detected, at least one target service scenario corresponding to the service request can be determined. From multiple first intelligent modules, the first target intelligent module corresponding to the target service scenario can be determined, and the service request can be processed using the first target intelligent module to obtain a corresponding first processing result. Furthermore, a service processing response corresponding to the service request for the target object can be generated using a management intelligent module and the first processing result. In other words, this embodiment proposes a method for collaboratively processing service requests using multiple intelligent modules (management intelligent module + first target intelligent module). By accurately identifying the target service scenario corresponding to the service request, it ensures that the service request can be directed to a suitable first target intelligent module for processing. The first target intelligent module can provide customized responses for the corresponding target service scenario, compensating for the lack of versatility of a single agent in related technologies. The management intelligent module integrates and optimizes the first processing result generated by the first intelligent module, generating a more comprehensive and accurate service processing response. The above method overcomes the limitations of static routing and single agents in handling complex service requests in related technologies, enhancing the user experience. This achieves the technical effect of improving the processing effect of service requests and solves the technical problem of poor processing effect of service requests in related technologies.

[0016] The above general description and the following detailed description are for illustrative and explanatory purposes only and do not constitute a limitation thereof. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This is a schematic diagram illustrating an application scenario of a service request processing method according to an embodiment of this application;

[0019] Figure 2 This is a flowchart of a service request processing method according to an embodiment of this application;

[0020] Figure 3 This is a schematic diagram of an after-sales service outsourcing solution based on multi-intelligent module collaboration according to an embodiment of this application;

[0021] Figure 4 This is a schematic diagram of a service request processing apparatus according to an embodiment of this application;

[0022] Figure 5 This is a schematic diagram of a service request processing system according to an embodiment of this application;

[0023] Figure 6 This is a structural block diagram of a computing device according to an embodiment of this application;

[0024] Figure 7 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some, not all, of the embodiments of the present application. Other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative effort should fall within the scope of protection of the present application.

[0026] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in other orders. Wherein, "other orders" refers to orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such as a process, method, system, product, or apparatus that comprises a series of steps or units, not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed, or inherent to such processes, methods, products, or apparatus.

[0027] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions, and corresponding access points are provided for users to choose to authorize or refuse.

[0028] The technical solution provided in this application is mainly implemented using a deep learning model. Deep learning models can be widely applied in fields such as Natural Language Processing (NLP), computer vision, and speech processing. Specifically, they can be applied to computer vision tasks such as Visual Question Answering (VQA), Image Captioning (IC), and image generation, as well as to natural language processing tasks such as text-based sentiment classification, text summarization, and machine translation. Therefore, the main application scenarios of this application include, but are not limited to, digital assistants, intelligent robots, search, online education, office software, e-commerce, and intelligent design. In the embodiments of this application, the data processing using a recommendation model in a recommendation scenario is used as an example for explanation.

[0029] First, some nouns or terms appearing in the process of processing service requests in the embodiments of this application shall be interpreted as follows:

[0030] Standard Operating Procedures (SOPs) define fixed steps and rules for handling service requests to ensure service consistency.

[0031] Agent, intelligent module, is a software entity that can independently perform specific tasks, such as handling service requests and providing solutions;

[0032] Multi-agent systems (multi-agent architectures) consist of multiple independent agents that collaborate to handle complex tasks, improving processing flexibility and efficiency.

[0033] The Supervisor Agent module is responsible for coordinating and integrating the outputs of multiple agents, deciding on the final service response, and ensuring the optimization and consistency of the results.

[0034] Specialized sub-Agents are agents customized for specific scenarios or user groups, such as after-sales service for specific products or care for high-value customers, to provide more professional services.

[0035] Scene routing automatically identifies and assigns the most suitable Agent or scene path to the service request based on its content and context, thus achieving precise service.

[0036] According to an embodiment of this application, a method for processing service requests is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or the information may be executed in a different order than that shown here.

[0037] The deep learning model involved in the embodiments of this application can be an artificial intelligence-based language model (LM) or a multimodal model (MM).

[0038] Considering the limited computing resources of mobile terminals, the methods described above in this application embodiment can be applied to, for example... Figure 1 The application scenarios shown are as follows. In, for example... Figure 1 In the application scenario shown, the deep learning model is deployed on server 10. Server 10 can connect to one or more client devices 20 via a local area network (LAN), wide area network (WAN), internet connection, or other types of data network. Client devices 20 may include, but are not limited to, smartphones, tablets, laptops, PDAs, personal computers, smart home devices, and in-vehicle devices. Client devices 20 can interact with users through a graphical user interface to invoke the deep learning model, thereby implementing the service request processing method provided in this embodiment.

[0039] In this embodiment, the system consisting of a client device and a server can perform the following steps: The client device executes a service request for a target object. The server executes a response to the service request for the target object, determining at least one target service scenario corresponding to the service request; determining a first target intelligent module corresponding to the target service scenario from a plurality of first intelligent modules, wherein different first intelligent modules are used to process service requests corresponding to different service scenarios; processing the service request using the first target intelligent module to obtain a first processing result corresponding to the first target intelligent module; and generating a service processing response corresponding to the service request using a management intelligent module and the first processing result.

[0040] With the rapid development of high-performance computing units, the methods provided in this application embodiment can also be applied to model-in-the-loop machines in other application scenarios. In one optional embodiment, the model-in-the-loop machine has multiple built-in models. Users can select a model to adjust as needed to obtain their own model. The high-performance computing unit built into the model-in-the-loop machine can then directly call the adjusted model to execute the methods provided in this application embodiment. In another optional embodiment, the deep learning model-in-the-loop machine has a pre-trained model built-in. The high-performance computing unit built into the model-in-the-loop machine can then directly call this model to execute the methods provided in this application embodiment.

[0041] Furthermore, when users need to train their own models, they can upload their own datasets via the client. This dataset is sent from the client to the server. The server can then use this dataset to fine-tune the pre-trained model, resulting in the user's customized model, which can then be deployed to the production environment. To facilitate user adjustments, the server provides complete adjustment tools, development frameworks, and processes, supporting various adjustment strategies. This allows the adjusted model to better adapt to different application domains and achieve a high degree of customization.

[0042] Under the aforementioned operating environment, this application provides the following: Figure 2 The method for handling service requests is shown. Figure 2 This is a flowchart of a service request processing method according to an embodiment of this application. For example... Figure 2 As shown, the method may include the following steps:

[0043] Step S202: In response to a service request for a target object, determine at least one target service scenario corresponding to the service request.

[0044] In the technical solution provided in step S202 of this application, the target object can be an object that needs to be processed and resolved through a service request. For example, if the service request scenario is an e-commerce scenario, the target object can be a specific product or service in the e-commerce scenario, and the user can be a customer of the e-commerce platform. The service request can be a user's need or problem regarding the target object, used to request corresponding services or solutions for the target object. For example, if the service request scenario is an e-commerce scenario, the target request can be various inquiries about the target object. Furthermore, if the e-commerce scenario is further refined to an after-sales scenario, the service request can be a user's needs for returns, repairs, or inquiries about goods or services; that is, the service request can be a product return request, a product inquiry request, etc.

[0045] Optionally, the target service scenario can be a specific task environment or logical framework for processing the aforementioned service request, defined based on the content and background of the service request. For example, if the service request is a product return request, the target service scenario may include return review, refund processing, and logistics arrangement; if the service request is a product consultation request, the target service scenario may include product information inquiry, usage guidance, and answers to frequently asked questions.

[0046] In this embodiment, if a service request is detected in response to a target request, at least one target service scenario corresponding to the service request can be determined.

[0047] Optionally, after detecting a user's service request targeting a specific object, the target object in the service request can be identified. For example, by parsing the content of the service request, specific product identifiers, user account information, etc., can be quickly located, thereby clarifying the target object. The type of service request can also be identified, such as returns, exchanges, repairs, inquiries, and other types. For example, service requests can be categorized using Natural Language Processing (NLP) technology or keyword recognition.

[0048] Optionally, based on an understanding of the target object and the type of service request, the context or task logic of the target request can be further analyzed. This context and task logic can then be matched with various service scenarios in a pre-defined service scenario library to obtain matching results. Each service scenario represents a specific set of processing flows and strategies, such as an e-commerce return scenario or a high-value customer consultation scenario. Based on the matching results, the target service scenario can be determined from the service scenarios.

[0049] For example, user Xiao Wang submits a service request through the customer service portal: "I bought a pair of sneakers last week, but the size doesn't fit. I want to return them and apply for a refund." By analyzing the content of this service request, we can determine that the target service scenario is "product return and refund." This target service scenario may involve sub-tasks such as checking the product status, guiding the return process, and confirming refund conditions.

[0050] Step S204: Determine the first target intelligent module corresponding to the target service scenario from multiple first intelligent modules.

[0051] In the technical solution provided in step S204 of this application, different first intelligent modules are used to process service requests corresponding to different service scenarios.

[0052] Optionally, the first intelligent module can be an intelligent module in a multi-intelligent module architecture used for processing target service scenarios. The first intelligent module can be a basic or specialized processing unit in the multi-intelligent module architecture; different first intelligent modules have different characteristics and expertise, and can provide professional and accurate solutions for service requests in different target service scenarios. The first intelligent module can be a scenario agent.

[0053] Optionally, the first intelligent module in this embodiment may have the following characteristics: Each first intelligent module focuses on handling one or more specific service requests, such as warranty policy explanations, return and exchange process guidance, and product usage instructions. Through in-depth learning and understanding of different service scenarios, it can provide high-quality specialized services for different service scenarios. Each first intelligent module can operate independently and has its own knowledge base, decision model, and interaction strategy. This means that each first intelligent module can efficiently handle corresponding service requests without being affected by other first intelligent modules. The design of the multi-intelligent module architecture makes it easy to add or delete first intelligent modules, improving the flexibility of the multi-intelligent module architecture. The configuration of intelligent modules can be quickly adjusted according to task development needs or new service scenarios without making significant changes to the entire multi-intelligent module architecture.

[0054] Optionally, the first target intelligent module can be an intelligent module selected to handle the corresponding service request for a specific target service scenario in a multi-intelligent module architecture.

[0055] In this embodiment, after determining the target service scenario corresponding to the service request, the first target intelligent module corresponding to the target service scenario can be determined from multiple intelligent modules.

[0056] Optionally, in the process of handling service requests in collaboration among multiple intelligent modules, the first target intelligent module that is most suitable for handling the service request in the target service scenario can be intelligently selected from a number of preset first intelligent modules.

[0057] Optionally, the expertise of each first intelligent module in the multi-intelligent module architecture for a specific service scenario can be evaluated, and its matching degree with the target service scenario can be compared to determine which intelligent modules are capable of handling service requests under that target service scenario. In addition to service scenario matching, the current processing capacity and workload of the first intelligent modules can also be considered to avoid selecting overloaded or inefficient first intelligent modules, ensuring rapid response and high quality for service requests. Furthermore, the historical processing records of each first intelligent module can be referenced, such as success rate and user satisfaction metrics, while also considering specific user preferences, such as language preferences and processing style preferences, to optimize the selection of the first target intelligent module.

[0058] It should be noted that the selection process of the first target intelligent module in the embodiments of this application is dynamic. The selection strategy of the first target intelligent module can be adjusted according to the real-time feedback of service request processing, such as user satisfaction and intelligent module response speed, so as to achieve continuous optimization and improvement.

[0059] In this embodiment, by intelligently determining the first target intelligent module from multiple first intelligent modules, the collaboration of multiple intelligent modules enables flexible responses to complex and ever-changing service scenarios, ensuring that each service request receives professional, efficient, and personalized processing. This mechanism improves the efficiency of service request processing and enhances the user experience.

[0060] Step S206: The service request is processed using the first target intelligent module to obtain the first processing result corresponding to the first target intelligent module.

[0061] In the technical solution provided in step S206 of this application, the first processing result can be a preliminary or specific processing conclusion obtained after the first target intelligent module processes the assigned service request. The first processing result may include analysis of the target request, proposed solutions, and possible action suggestions.

[0062] For example, if the service request is related to product returns, exchanges, or warranties, the initial processing result may include confirmation of the service request, explanation of the return policy, feasibility analysis of exchanges, or specific warranty procedure guidance. For directly actionable service requests, such as providing a refund link, initiating an exchange process, or guiding users to reset products, the initial processing result may include specific instructions or steps to guide the user through the corresponding operation. If the service request involves querying specific information about the target object, such as the target object's order status or product usage instructions, the first target intelligent module will perform a data query and return the corresponding query results as a direct response to the service request. For complex service requests or those exceeding the processing capacity of the first target intelligent module, the initial processing result may include an escalation strategy of transferring the service request to a higher-level intelligent module or suggesting human customer service intervention.

[0063] In summary, the first processing result directly reflects the first target intelligent module's understanding and processing capabilities of the service request. For simple and direct service requests, the first processing result can be the final solution; while for complex and ever-changing service requests, the first processing result can provide key information and alternative solutions to enable higher-level decision-making and integration, ensuring that the final result fed back to the user for the service request not only conforms to the task rules but also meets the user's personalized needs.

[0064] In this embodiment, after determining the first target intelligent module corresponding to the target service scenario from multiple first intelligent modules, the first target intelligent module can be used to process the service request and obtain the corresponding first processing result.

[0065] Optionally, after identifying the first target intelligent module corresponding to the target service scenario, the first target intelligent module can be activated. This module can be used to analyze specific details in the service request, such as the user's problem description, product information, and order status. Through natural language processing technology and scenario matching algorithms, it can gain a deeper understanding of the user's true needs and problems.

[0066] Optionally, the first target intelligence module can search its knowledge base for relevant rules, solutions, and precedents based on the information parsed from the aforementioned service request, perform decision-making logic operations, and generate a first processing result. The process of generating the first processing result can be achieved through querying and following standard operating procedures (SOPs) and through personalized adjustments based on user characteristics (such as membership level and purchase history).

[0067] In this embodiment, for simple service requests, the first processing result can be the final solution, directly addressing the user's needs. For more complex service requests, the first processing result can serve as the basis for subsequent comprehensive decision-making by the intelligent management module, providing users with more comprehensive and personalized services.

[0068] Step S208: Using the management intelligence module and the first processing result, generate the service processing response corresponding to the service request.

[0069] In the technical solution provided in step S208 of this application, the management intelligent module (Supervisor Agent), also known as the supervisory intelligent module or the coordinating intelligent module, serves as the core control unit for processing service requests collaboratively by multiple intelligent modules. The management intelligent module can coordinate and manage the outputs (first processing results) of multiple first intelligent modules in the entire multi-intelligent module architecture, and analyze, synthesize, and optimize the first processing results to generate the final service processing response. The functions of the management intelligent module may include: collecting the processing results (including the first processing result and the processing results of other intelligent modules) of the intelligent modules participating in the service request processing (including the first target intelligent module and other potentially involved intelligent modules), summarizing and analyzing them. Based on preset task rules, policy priorities, and user historical data, the management intelligent module determines how to integrate the above processing results to generate the most suitable service processing response for the current service request. The management intelligent module can also monitor and evaluate the efficiency and quality of the service request processing flow, obtain evaluation results, and adjust the configuration, priority, and workflow of the intelligent modules in real time based on the evaluation results to achieve good service results.

[0070] Optionally, the service processing response can be the final reply from the management intelligent module to the user's service request. The service processing response can integrate the initial processing results from the first target intelligent module and the processing results from other potentially involved intelligent modules, forming a comprehensive, accurate, and user-relevant solution based on the decision-making logic of the management intelligent module. The service processing response can also be referred to as a decision result. Its characteristics include: containing information and suggestions from multiple intelligent modules, such as product knowledge, return and exchange processes, and technical support, providing users with a one-stop solution. Based on the user's needs, preferences, and historical interaction records, the service processing response is personalized by the management intelligent module to better suit the user's specific situation and improve the user experience. The service processing response can comprehensively cover the core requirements of the service request and present them to the user in a clear and easy-to-understand manner, ensuring that the user can accurately obtain the required information and operational guidance.

[0071] In this embodiment, after processing the service request using the first target intelligent module and obtaining the first processing result corresponding to the first target intelligent module, the management intelligent module and the first processing result can be used to generate a service processing response corresponding to the service request.

[0072] Optionally, the management intelligence module can collect processing results from various intelligence modules related to the service request, not limited to the primary target intelligence module, but also including other involved tool proxies, scenario intelligence modules, etc. After receiving the processing results from each intelligence module, the management intelligence module will compare and verify the quality and consistency of these results. Based on the user's specific needs, preferences, and current context (e.g., urgency, service history), the management intelligence module intelligently integrates the processing results received from different intelligence modules to form a comprehensive yet personalized service response. For example, if two intelligence modules suggest a refund and an exchange respectively, the management intelligence module can weigh the user's needs and the company's cost-effectiveness, selecting the appropriate suggestion or providing a combination of both.

[0073] Optionally, during the decision-making and response processing of the management intelligent module, pre-defined task rules and strategies can be followed, such as priority settings, user classification and processing rules, and cost control strategies, to ensure that the service processing response meets both user needs and the company's goals and standards. Based on the above analysis, the management intelligent module generates the final service processing response. This response may include, but is not limited to, processing steps, expected results, follow-up action suggestions, and any necessary customer communication content. Furthermore, the management intelligent module can optimize the generated service processing response, for example, by simplifying language and increasing the clarity of operational instructions, making it easier for users to understand and execute.

[0074] Optionally, after the management intelligent module generates a service processing response, it can collect user feedback on the service processing response to evaluate the efficiency and quality of the service processing process, continuously optimize its own decision-making logic and the performance of the entire multi-intelligent module architecture, and form a feedback loop of continuous learning and improvement.

[0075] For example, if user Xiaohua buys a piece of clothing, her service request for that clothing might be: "Upon receiving the item, I found the size does not match the description; I request an immediate return and refund." The target service scenarios corresponding to this request could include: "Return Process Scenario," "Return Strategy Scenario," and "User Emotion Management Scenario." The first target intelligent module corresponding to each of these three target service scenarios could be a Return Process Agent, a Refund Strategy Agent, and a User Emotion Management Agent, respectively. The first processing result from the Return Process Agent might be: confirming that Xiaohua's item is within the return period, providing a download link for a one-time return label, and instructing Xiaohua on how to pack and mail the returned item. The first processing result from the Refund Strategy Agent might be: based on Xiaohua's membership level and purchase history, recommending a direct refund instead of an exchange, and informing Xiaohua of the estimated refund time. The first processing result from the User Emotion Management Agent might be: through sentiment analysis, detecting that Xiaohua is dissatisfied with the shopping experience, and proposing additional compensation.

[0076] For example, the management intelligence module, combining the three primary processing results mentioned above, generates the following service response: "Dear user Xiaohua, we have received your return request and confirmed that your clothing is eligible for return. Please use this link to download the return label and follow the instructions to complete the return process. Considering your inconvenience, we will process your refund directly, which is expected to be completed within 7 business days. We sincerely apologize and will offer additional compensation to make up for your negative experience. Thank you for your understanding and support!"

[0077] In this embodiment, by leveraging the advanced decision-making and coordination capabilities of the management intelligent module, the processing results from different intelligent modules are comprehensively transformed into a user-oriented service processing response, aiming to provide an efficient, personalized, and compliant after-sales service experience. This process fully demonstrates the advantages of a multi-intelligent module architecture in flexibly responding to complex scenarios and improving service quality.

[0078] Through steps S202 to S208 of this application, if a service request for a target object is detected, at least one target service scenario corresponding to the service request can be determined. From multiple first intelligent modules, the first target intelligent module corresponding to the target service scenario can be determined, and the service request can be processed using the first target intelligent module to obtain a corresponding first processing result. Furthermore, the management intelligent module and the first processing result can be used to generate a service processing response corresponding to the service request for the target object. In other words, this embodiment proposes a method for collaboratively processing service requests using multiple intelligent modules (management intelligent module + first target intelligent module). By accurately identifying the target service scenario corresponding to the service request, it ensures that the service request can be directed to a suitable first target intelligent module for processing. The first target intelligent module can provide customized responses for the corresponding target service scenario, compensating for the lack of versatility of a single agent in related technologies. The management intelligent module integrates and optimizes the first processing result generated by the first intelligent module, generating a more comprehensive and accurate service processing response. The above method overcomes the limitations of static routing and single agents in handling complex service requests in related technologies, enhancing the user experience. This achieves the technical effect of improving the processing effect of service requests and solves the technical problem of poor processing effect of service requests in related technologies.

[0079] The method described in this embodiment will be further described below.

[0080] As an optional implementation, step S202, determining at least one target service scenario corresponding to the service request, includes: determining the target user corresponding to the service request; performing intent recognition on the service request to determine the request type corresponding to the service request; and determining the target service scenario based on the target object, request type, and target user.

[0081] In this embodiment, the target user can be the user who initiated the service request. The target user's identity can be identified through their account information on the platform where the service request was initiated. Through account information, in-depth user profiling can be performed to understand the user's behavioral patterns, preferences, historical transaction records, communication style, and other comprehensive information, resulting in a user profile. The user profile can be a data structure containing the following information: basic information, purchase history, communication records, user level, and sentiment analysis. Basic information can include the user's age, gender, and geographical location. Purchase history represents the user's purchase records over a historical period, including purchase frequency, category preferences, and price sensitivity. Communication records represent the user's interactions with the platform, including interactions with customer service, complaint history, and inquiry topics. Membership level represents the user's membership status on the platform where the service request was initiated, reflecting the user's activity and loyalty on the platform. Sentiment analysis represents the emotional state expressed by the user in the service request, such as dissatisfaction, anxiety, or satisfaction.

[0082] Optionally, the request type can be obtained by classifying the content of the user's service request, and can be divided based on the subject, purpose, or nature of the problem. Different service request processing scenarios correspond to different request types.

[0083] For example, in e-commerce scenarios, request types can include: returns and refunds, order status inquiries, product inquiries, payment issues, account issues, and complaints and suggestions. Returns and refunds allow users to request a return and refund, possibly due to quality problems, incorrect sizes, or non-receipt of the item. Order status inquiries allow users to check the order status of the item, such as whether it has been shipped and its estimated arrival time. Product inquiries allow users to seek detailed answers to questions about product functions, usage, and compatibility. Payment issues allow users to resolve difficulties encountered during payment, such as payment failures or duplicate charges. Account issues allow users to address account irregularities, such as login problems or account theft. Complaints and suggestions allow users to file complaints or provide improvement suggestions when dissatisfied with the product or customer service.

[0084] Optionally, in determining at least one target service scenario corresponding to a service request, the target user corresponding to the service request can be identified, or the service request can be subjected to intent recognition to determine the request type corresponding to the service request. Thus, the target service scenario can be determined based on the aforementioned target object, request type, and target user.

[0085] Optionally, upon receiving a service request from a user, a user profile can be retrieved or created based on the user's account information. If the user's account information changes, the user profile can be automatically updated. In-depth analysis of user profiles can identify factors such as user purchasing habits, complaint tendencies, and service sensitivity, preparing for subsequent personalized service request processing.

[0086] Optionally, during the process of intent recognition and request type determination for service requests, natural language processing technology can be used to interpret keywords and context in the service requests, understand the user's true needs and emotional state, and obtain intent recognition results. For example, from the service request "My newly purchased phone has a problem, I want to return it," it can be identified that the user's intent is to return the product. Based on the intent recognition results, service requests can be automatically classified to obtain request types, such as returns, exchanges, product inquiries, account issues, payment questions, etc.

[0087] Optionally, in determining the target service scenario based on the target object, request type, and target user, relevant entity information related to the target object, such as specific products and order numbers, can be extracted from the service request. Each combination of request type and target object may correspond to a different service scenario, and the relevant target service scenario can be determined based on the above request type and target object. For example, "return" + "mobile phone" can be matched to the target service scenario of "electronic product return process". Based on user profiles, the target service scenarios determined by the above request type and target object can be personalized and fine-tuned to obtain target service scenarios that better fit the user profile, that is, considering the user's historical satisfaction, membership level, etc., to provide a service experience that is closer to the user's needs. For example, for users who frequently complain, the target service scenario of "advanced user complaint handling" can be prioritized.

[0088] In this embodiment, the detailed analysis of service requests and the matching process with service scenarios ensure that every service request is accurately understood and efficiently processed. From identifying the target user, to recognizing the request intent and type, and then determining the most suitable service scenario based on the request type, target object, and target user, the entire process demonstrates the intelligence, personalization, and flexibility of multi-intelligent module collaboration in service processing. The automation and intelligence of these operations significantly shorten problem-solving time, improve user experience, and also reduce the workload of service personnel (e.g., human customer service representatives), thereby improving service efficiency and quality.

[0089] As an optional implementation, the service request is processed using the first target intelligence module to obtain a first processing result corresponding to the first target intelligence module, including: using the first target intelligence module to determine a target node that matches the service request from multiple operation process nodes, wherein the target node includes: response data and action text data of the target operation process, and the target operation process is the operation process represented by the target node; using the first target intelligence module to make an action decision on the response data and action text data to generate a target action; and using the first target intelligence module to execute the target action to obtain the first processing result.

[0090] In this embodiment, the operation flow nodes can be the various processing steps or decision points defined in the SOP process. For example, if the application scenario of the operation flow nodes is an intelligent after-sales service scenario, the operation flow nodes can be specific customer service tasks or decision-making steps such as inquiring about user details, checking order status, providing answers to frequently asked questions, executing refund operations, and transferring to human customer service. The operation flow nodes are used to guide the first target intelligent module to perform orderly and standardized processing.

[0091] Optionally, the target node can be a specific processing step or decision point determined by the first target intelligent module from multiple operation process nodes based on the characteristics of the service request, which can effectively resolve the current service request. For example, if the user's service request is about product returns, the target node may include an explanation of the return policy, guidance on the return process, and execution of the return operation. Within the SOP framework, the target node represents the most direct and critical response step for a certain type of service request. The response data can be information contained within the target node used to directly reply to the user, and can be text, images, links, or other forms of data. The response data aims to clearly and accurately answer the user's questions or provide solutions, such as providing the user with instructions on how to use the product, detailed information on the return address, and answers to frequently asked questions. This response data can also be referred to as the response answer.

[0092] Optionally, action text data can be a textual description of instructions or operations associated with the target node, which can be used to guide the first intelligent module to perform specific task operations. For example, this text data can instruct the intelligent module to "change the order status to 'return processing'" or "send a refund confirmation email to the user." Action text data is the basis for the first intelligent module to execute tasks. Unlike response data, action text data focuses more on the internal actions of the multi-intelligent module system. Action text data can also be called action text (text + action form). The target operation flow can be a standardized task flow represented by the target node for handling specific types of service requests. The target operation flow includes the complete operation steps and decision logic from receiving the service request to the first target intelligent module solving the problem. The target operation flow is a subset of SOP, specifically designed to handle service requests in specific scenarios, ensuring that the processing is both standardized and efficient.

[0093] Optionally, the target action can be the final operation instruction generated by the first target intelligent module through analysis and decision-making based on response data and action text data. The target action can be an immediate reply to the user, an operation performed within the multi-intelligent module system (e.g., updating order status, initiating a refund process), or a transfer request to other intelligent modules or human customer service. The target action is the actual response of the intelligent module to the service request, directly impacting the resolution of the user's problem or advancing the processing flow. The target action can also be referred to as an answer plus an execution action.

[0094] Optionally, during the process of processing the service request using the first target intelligence module to obtain the first processing result corresponding to the first target intelligence module, the first target intelligence module can be used to determine the target node that matches the current service request from multiple operation process nodes. The first target intelligence module can be used to make corresponding action decisions on the response data and action text data in the target node, generating a target action. For example, the action node of the SOP leaf node can be modified to transform the response answer + execution action into a text + action format. This text + action format is then submitted to the first target intelligence module, which executes the target action corresponding to the text + action format to obtain the first processing result.

[0095] Optionally, during the analysis of service requests and determination of target nodes, if a user initiates a service request, the first target intelligence module can receive the request. The first target intelligence module uses built-in NLP technology to analyze keywords, context, and user intent in the service request to quickly understand the user's specific needs. After analyzing the user's specific needs from the service request, the first target intelligence module can filter out target nodes matching the current service request from a set of preset operation flow nodes. These operation flow nodes constitute part of the Standard Operating Procedure (SOP), and each operation flow node represents a set of rules or steps for handling a specific type of service request. The determination of target nodes can be based on the degree of matching between the user request and the characteristics of the operation flow nodes in the SOP. For example, if the user's service request is related to "return," the first target intelligence module will identify operation flow nodes related to the return process as target nodes.

[0096] Optionally, during the action decision-making and target action generation process, the first target intelligence module can further analyze and make decisions on the response data and action text data in the target node to determine the appropriate response method for the service request. For example, it can understand the depth of user needs, assess the priority of the operation, and evaluate the potential impact from multiple dimensions such as the response data and action text data. The first target intelligence module can use machine learning algorithms to predict which response will meet user needs based on historical data and current strategies. Based on the action decision results for the response data and action text data, the first target intelligence module generates a specific target action. The target action could be sending a message to the user, updating a field in the database of the multi-intelligent module system, or calling an external service to perform a refund operation, etc.

[0097] Optionally, during the execution of the target action and the feedback of the first processing result, the target action generated in the previous step is executed through the first target intelligence module. For example, the first target intelligence module can be a front-end operation that interacts with the user, such as sending a message or email; it may also include back-end system operations, such as updating the order status or initiating a refund process in the financial system. After the target action is executed, the first processing result can be obtained.

[0098] In this embodiment, the first target intelligent module performs detailed request analysis, intelligent decision-making, and precise operations to efficiently and personally resolve user service requests. This process relies on a highly flexible and intelligent multi-agent architecture, combining the standardization of standard operating procedures (SOPs) with the dynamic response capabilities of intelligent modules, aiming to provide a superior, more accurate, and more efficient user experience than traditional intelligent customer service.

[0099] As an optional implementation, the method further includes: determining at least one second target intelligent module corresponding to the service request from a plurality of second intelligent modules based on the target object and the target user corresponding to the service request, wherein the second target intelligent module is a second intelligent module that matches the target object and / or the target user; processing the service request using the second target intelligent module to obtain a second processing result corresponding to the second target intelligent module; step S208, generating a service processing response corresponding to the service request using the management intelligent module and the first processing result, including: fusing the first processing result and the second processing result using the management intelligent module to generate a service processing response.

[0100] In this embodiment, the second intelligent module can refer to a specialized sub-Agent in a multi-intelligent module architecture, which can be used to handle service requests in specific domains, types, or for specific user groups. The second intelligent module can possess deeper knowledge and a more powerful ability to handle specific situations, enabling it to provide more accurate and personalized services. The second intelligent module focuses on handling specific types of service requests, such as different service categories like returns, exchanges, repairs, and inquiries, or issues related to a specific product category. Based on user profiles and historical interaction information, the second intelligent module can provide services tailored to the specific user characteristics, such as customized solutions based on membership level, shopping habits, and communication style. The second intelligent module can be based on deep learning technology, continuously improving its processing efficiency and accuracy in specific domains through continuous learning and optimization.

[0101] Optionally, the second target intelligent module can be a second intelligent module selected from multiple second intelligent modules after determining the target scenario and target user of the service request. The process of selecting the second target intelligent module from the second intelligent modules can be based on a deep understanding of the request content, target object, and user characteristics to ensure that the service request can be processed professionally and effectively.

[0102] Optionally, the second processing result may refer to the processing suggestions, solutions, or direct responses provided by the second target intelligent module in response to the service request. The second processing result can reflect the second intelligent module's deep understanding and service capabilities in a specific domain or for a specific user. The second processing result can directly address the specific domain or user characteristics of the service request, and may include return guidelines, repair appointments, product recommendations, account problem resolution suggestions, etc. When generating the second processing result, the second target intelligent module can integrate information from multiple aspects such as product information, user history, and service policies to provide a comprehensive solution that meets user expectations. Due to the specialized capabilities of the second intelligent module and its deep understanding of the target user, the second processing result often provides a more accurate service solution, improving user satisfaction.

[0103] Optionally, based on the target object and the target user corresponding to the service request, at least one second target intelligent module corresponding to the service request can be determined from multiple second intelligent modules. The second target intelligent module can then be used to process the service request to obtain a corresponding second processing result. During the process of generating a service processing response using the management intelligent module and the first processing result, the management intelligent module can be used to fuse the aforementioned first and second processing results to generate the service processing response.

[0104] Optionally, by deeply analyzing the user profiles of the target object and target user in the service request, at least one second target intelligent module can be intelligently matched from a set of pre-defined second intelligent modules (specialized sub-Agents / specialized Agents) based on the target object and user profile, resulting in a matching result. By confirming the matching result, it is ensured that the selected second target intelligent module can meet the professional and personalized requirements of the service request. Once the second target intelligent module is determined, it will be activated or assigned to begin processing the service request. Each second target intelligent module utilizes its knowledge base and processing logic to perform in-depth analysis and processing of the service request, generating a second processing result.

[0105] Optionally, the management intelligence module can collect the second processing results provided by each second target intelligence module, as well as the first processing result provided by the first target intelligence module. The first and second processing results can be combined to formulate a corresponding service processing response. Alternatively, a corresponding service processing response can be formulated by selecting one of the two processing results.

[0106] In this embodiment, by introducing a second intelligent module that matches a specific target object and target user, the processing power and personalized response of the multi-intelligent module architecture are enhanced. The management intelligent module not only plays a coordinating and supervisory role but is also responsible for integrating and optimizing the processing results of different intelligent modules, ensuring that the final service response not only solves the user's immediate problems but also meets the user's deeper experience expectations.

[0107] As an optional implementation, based on the target object and the target user corresponding to the service request, the second target intelligent module corresponding to the service request is determined from multiple second intelligent modules, including: obtaining the filtering conditions corresponding to the second intelligent module; matching the target object and the target user with multiple filtering conditions to obtain the target filtering conditions that match the target object and / or the target user; and determining the second intelligent module corresponding to the target filtering conditions as the second target intelligent module.

[0108] In this embodiment, the filtering criteria can be rules or standards used to differentiate and define the responsibilities and processing permissions of each intelligent module in a multi-intelligent module architecture. The filtering criteria can be set based on the content of the service request, the target object, and the target user. The purpose of the filtering criteria is to improve the efficiency and targeting of service request processing, avoid resource waste, and ensure that users' service requests are professionally resolved. The filtering criteria can also be called admission criteria.

[0109] Optionally, the target filtering criteria can be a set of conditions found from the filtering criteria of multiple second intelligent modules when processing a service request, matching the current service request, target object, and target user. Target filtering criteria can be used to narrow down the candidate range of second intelligent modules, ensuring that the subsequently assigned second target intelligent module can effectively process the service request.

[0110] Optionally, in the process of determining the second target intelligent module based on the target object and target user, the filtering conditions corresponding to the second intelligent module can be obtained. The target object and target user can be matched with multiple filtering conditions to obtain the target filtering conditions that match the target object and / or target user, thereby determining the second target intelligent module corresponding to the target filtering conditions.

[0111] Optionally, a series of second intelligent modules can be pre-configured in the multi-intelligent module architecture. Each second intelligent module has specific filtering conditions, which define the processing scope and entry threshold of the intelligent module. For example, one second intelligent module can be used to handle return and exchange requests for high-value goods, while another can be used to answer inquiries related to a specific brand. By parsing the target object and target user in the service request and comparing them one by one with the pre-configured filtering conditions of the second intelligent modules, the degree of fit between the target object and target user and the filtering conditions is evaluated. The above matching logic not only considers the target object but also comprehensively evaluates the target user, ensuring that the selected second target intelligent module can simultaneously meet the requirements of both aspects.

[0112] Optionally, in the process of determining the second intelligent module corresponding to the target filtering conditions as the second target intelligent module, after the matching process between the target object, target user, and filtering conditions is completed, one or more target filtering conditions matching the target object and target user can be generated. Based on the target filtering conditions, it can be determined which second intelligent modules are suitable for handling the current service request. The above decision can be based on various factors such as the priority of the filtering conditions, the availability of the second intelligent modules, and past processing performance indicators. Finally, the service request is assigned to the second target intelligent module determined through the above decision-making process.

[0113] For example, if a VIP user submits a return request for a recently purchased high-end smartwatch, multiple secondary smart modules can be selected based on the smartwatch's product attributes, return service type, and the VIP user's profile to match a target secondary smart module with the appropriate professional processing capabilities and priority service. This target secondary smart module not only understands the relevant return and exchange policies for smartwatches but can also provide a faster response and a higher level of service based on the VIP user's status.

[0114] In this embodiment, the core of the method lies in intelligently matching target users, target objects, and filtering conditions to guide service requests to a suitable second intelligent module for processing. The design and execution of the filtering conditions ensure the accuracy and efficiency of the service request processing flow. By using filtering conditions that match specific objects and user attributes, a second target intelligent module with specialized expertise can be invoked to improve the accuracy of service request processing. Considering the specific needs and preferences of target users, the above mechanism can provide more personalized service responses and improve user experience.

[0115] As an optional implementation, the target object and target user are matched with multiple filtering conditions to obtain target filtering conditions that match the target object and / or target user. This includes: matching the object information of the target object with the object information in the filtering conditions to obtain a first filtering condition that matches the target object; matching the user information of the target user with the user information in the filtering conditions to obtain a second filtering condition that matches the target user; and obtaining target filtering conditions based on the first filtering condition and / or the second filtering condition.

[0116] In this embodiment, the object information of the target object can be used to represent the specific attributes and details of the target object. The object information in the filtering conditions can be used to represent the specific attributes and details of the object that matches the filtering conditions. For example, in an e-commerce scenario, the object information may include information such as the product model, purchase time, usage status, and order number. The first filtering condition may refer to the set of object information conditions used by the second intelligent module when processing service requests related to a specific object. The first filtering condition aims to identify the characteristics and classification of objects in the service request, thereby guiding the service request to the most appropriate second target intelligent module for processing. For example, a second intelligent module that specifically handles electronic device failures may have first filtering conditions that include elements such as "object is an electronic device" and "report failure".

[0117] Optionally, user information can be used to represent multi-dimensional information such as the background information, preferences, historical behavior, and current needs of the user initiating the service request. This information can assist the intelligent module in understanding the user's needs hierarchy and prioritizing request processing. For example, it can include the user's VIP level, historical complaint records, shopping frequency, and satisfaction rating of their most recent purchase. The second filtering condition can be based on user information, defining the rules or strategies followed by the second intelligent module when handling service requests involving specific user groups. The second filtering condition can be used to determine which second intelligent module can better understand and meet the user's specific needs, especially for users with special statuses (e.g., VIP users) or behavioral patterns. For example, a second intelligent module providing rapid response and priority processing for VIP users could have second filtering conditions such as "user is a VIP" or "user's need is urgent."

[0118] Optionally, in the process of matching target objects, target users, and filtering conditions to obtain target filtering conditions, the object information of the target object can be matched with the object information in the filtering conditions to obtain the corresponding first filtering condition. The user information of the target object can be matched with the user information in the filtering conditions to obtain the second filtering condition. At least one of the above two filtering conditions can be used as the target filtering condition.

[0119] Optionally, object information of the target object can be extracted from the service request. The extracted object information is compared with object information in preset filtering conditions across multiple secondary intelligent modules to identify which filtering conditions match the target object. Once a filtering condition with a high degree of matching is found, it can be marked as the first filtering condition, indicating which secondary intelligent modules possess the professional qualifications to handle requests related to the target object.

[0120] Optionally, the target user's user information can be parsed. This information can be matched against the user information in the filtering criteria of the second intelligent module to determine which filtering criteria match the user information. During this matching process, filtering criteria highly relevant to the current target user's user information can be selected and marked as second filtering criteria to guide how to locate services based on the user's specific needs.

[0121] Optionally, the first and second screening criteria can be comprehensively considered to assess which criteria are important or have a high impact on the current service request. The first and second screening criteria can also be prioritized according to task requirements. Based on priority and matching degree, at least one of the first and second screening criteria can be selected as the target screening criterion to guide subsequent intelligent module selection and request processing.

[0122] For example, if the service request involves a high-end brand of electronic products, and is initiated by a VIP user who frequently purchases high-end electronic products and has an urgent service need, the core focus can be determined from multiple filter criteria such as "high-end brand electronic products," "VIP user," and "urgent need," thereby generating target filter criteria for "high-end brand electronic products" and "VIP user's urgent service request."

[0123] In this embodiment, the above method effectively identifies a second intelligent module suitable for handling service requests from the corresponding target object and target user. This process not only improves service response speed and accuracy but also ensures service personalization and professionalism, enhancing the user experience.

[0124] As an optional implementation, the management intelligence module is used to fuse the first processing result and the second processing result to generate a service processing response, including: obtaining the first priority of the first target intelligence module and the second priority of the second target intelligence module; determining at least one candidate intelligence module from the first target intelligence module and the second target intelligence module based on the first priority, the second priority and a preset processing strategy, wherein the candidate intelligence module satisfies the preset processing strategy and the priority of the candidate intelligence module is greater than the preset priority; and using the management intelligence module to fuse the processing results corresponding to the candidate intelligence module to generate a service processing response.

[0125] In this embodiment, the first priority can be used to represent the priority level of the first target intelligent module when processing service requests, and can reflect the priority of the first target intelligent module in service requests under different scenarios. The setting of the first priority can be based on various factors such as the professional capabilities, processing efficiency, and historical performance of the first target intelligent module, and may also be related to the urgency, complexity, or specific task requirements of the service request. For example, if the first target intelligent module is a first intelligent module specifically handling electronic product returns and exchanges, then the first priority of the first target intelligent module can be higher, especially when the service request involves electronic product returns and exchanges.

[0126] Optionally, the second priority can be used to represent the priority level of the second target intelligent module when processing service requests. Similar to the first priority, the second priority can reflect the processing priority of the second target intelligent module when processing service requests from a specific target object or a specific target user. By setting the second priority, the relative importance of different second target intelligent modules under a specific service request can be evaluated, thereby making corresponding processing decisions for the service request.

[0127] For example, if the second target intelligence module is a second intelligence module focused on VIP customer service, the second priority of the second target intelligence module can take precedence over the second intelligence module for ordinary user service, especially when the service request comes from a VIP user.

[0128] Optionally, the preset processing strategy can refer to the task rules or strategies followed when generating service processing responses. That is, it can be a task strategy that instructs the management intelligent module how to select and integrate the processing results of different intelligent modules based on first priority, second priority, and the content of the service request. The preset processing strategy may include prioritizing urgent requests, prioritizing the output of intelligent modules with good historical performance, and adjusting the intelligent module selection logic based on user satisfaction feedback. For example, for complex service requests, the preset processing strategy may tend to integrate the processing results of multiple intelligent modules to provide a comprehensive solution. When processing service requests from VIP users, the preset strategy may consider the processing results of the VIP service intelligent module to ensure service quality and user satisfaction.

[0129] Optionally, a candidate intelligent module can refer to an intelligent module that may provide a relatively accurate service processing response, selected by comparing a first priority, a second priority, and a preset processing strategy before the first processing result and the second processing result are merged.

[0130] Optionally, in the process of fusing the first and second processing results to generate a service processing response, candidate intelligent modules can be determined from the first and second target intelligent modules based on the first priority of the first target intelligent module, the second priority of the second target intelligent module, and a preset processing strategy. The management intelligent module can then be used to fuse the processing results corresponding to the candidate intelligent modules to generate a service processing response.

[0131] Optionally, after parsing the service request, the service request can be assigned to a first target intelligent module and a second target intelligent module through scenario routing. These two intelligent modules generate a first processing result and a second processing result respectively based on their respective specialized capabilities and the characteristics of the service request. After obtaining the first and second processing results, the management intelligent module can compare the priorities of the first and second target intelligent modules and the matching degree between the two processing results and the policy requirements according to a preset processing strategy, and select intelligent modules that meet the conditions as candidate intelligent modules. The management intelligent module collects the processing results of the candidate intelligent modules, including processing suggestions, solutions, customer service scripts, etc., and performs fusion processing based on a preset strategy. The fusion process can adjust the weights according to the preset processing strategy requirements. For example, if the strategy requires "maximizing customer satisfaction," the management intelligent module can give greater weight to the processing results of intelligent modules that have historically brought more positive feedback. Based on the fused processing results, the management intelligent module generates a comprehensive service processing response that meets the requirements, which simultaneously considers problem solving, user experience, and the achievement of task objectives.

[0132] In this embodiment, the combination of prioritization and preset processing strategies ensures that service requests are processed by the most suitable intelligent modules, thereby improving the relevance and professionalism of the response. Prioritizing intelligent modules helps allocate resources more rationally, ensuring that high-value or urgent requests are processed first. Integrating the outputs of multiple intelligent modules provides more comprehensive and personalized solutions, enhancing user satisfaction and trust. The execution of preset processing strategies guarantees the achievement of task objectives.

[0133] As an optional implementation, the processing results corresponding to candidate intelligent modules are fused using a management intelligent module to generate a service processing response, including: using at least one tool to process the service request and obtain service information corresponding to the service request; using the management intelligent module and the service information to fuse the processing results corresponding to candidate intelligent modules to obtain fused information; and using the management intelligent module and the fused information to generate a service processing response.

[0134] In this embodiment, a tool can refer to a software component or data processing module used to assist the intelligent module in analyzing, making decisions, or executing specific tasks during service processing. Tools can enhance the functionality and processing efficiency of the intelligent module. Tools include order tracking tools, user behavior analysis tools, and sentiment recognition tools. Service information can be additional information generated by the intelligent module during the process of using tools to process service requests. Service information can provide the intelligent module with a deeper understanding and more accurate decision-making basis. Service information can include: object status information, user background information, tool output information, and processing suggestions. Object status information can be used to indicate the current status of the target object, such as whether it is under warranty or has inventory. User background information can be used to indicate the user's purchase history, service preferences, membership level, etc. Tool output information can be auxiliary information provided by the aforementioned tools, such as logistics status prediction and user satisfaction prediction. Processing suggestions can be suggestions proposed by the intelligent module based on preliminary analysis, such as refunds, exchanges, or repairs. Optionally, fused information can be formed based on the processing results and service information provided by candidate intelligent modules integrated by the management intelligent module, and can be a processing solution generated from a comprehensive understanding of the service request.

[0135] Optionally, during the process of fusing the processing results corresponding to candidate intelligent modules using the management intelligent module, at least one tool can be used to process the service request and obtain the service information corresponding to the service request. Then, by fusing the processing results corresponding to candidate intelligent modules using the management intelligent module and the service information, fused information is obtained. Finally, using the management intelligent module and the fused information, a service processing response is generated.

[0136] Optionally, a tool call can be invoked to deeply analyze the service request and collect relevant service information to better understand the background of the service request and user needs. Based on the tool's analysis results, the intelligent module generates a series of service information related to the service request. The management intelligent module receives processing results from candidate intelligent modules, as well as comprehensive service information obtained in the first step. The management intelligent module compares the processing results of different intelligent modules, analyzes the advantages and disadvantages of each result, and considers details of service information such as user satisfaction, processing cost, and efficiency. The management intelligent module can apply specific fusion algorithms, such as weighted average, rule-based decision, and learning-based decision, to synthesize the outputs of candidate intelligent modules and the service information provided by the tool. Through the fusion algorithm, the management intelligent module generates fused information. Based on the fused information, the management intelligent module formulates the final service processing response strategy.

[0137] In this embodiment, the above method effectively integrates the professional capabilities of different intelligent modules and external information provided by tools, enabling a deep understanding and efficient processing of complex service request scenarios. This process not only improves the accuracy and personalization of service responses but also effectively enhances the flexibility and response speed of the entire service processing. It serves as a mechanism for automating and optimizing complex task flows within a multi-intelligent module architecture.

[0138] As an optional implementation, the method further includes: in response to a smart module configuration request, adding or deleting a plurality of second smart modules based on the smart module configuration request.

[0139] In this embodiment, the intelligent module configuration request can refer to an instruction issued by the manager or operator of the multi-intelligent module architecture, used to adjust the multi-intelligent module architecture and functions. The intelligent module matching value request can be manually triggered or automatically generated when the performance of the aforementioned multi-intelligent module architecture is insufficient, the workload fluctuates, or other task requirements change.

[0140] Optionally, if a configuration request for a second intelligent module is detected, multiple second intelligent modules can be added or removed.

[0141] Optionally, if a smart module configuration request is detected, the specific requirements in the smart module configuration request can be analyzed. For example, whether a corresponding second smart module needs to be added due to a surge in a certain type of service request, or whether certain second smart modules need to be reduced due to adjustments in task strategies.

[0142] Optionally, for the addition of intelligent modules, pre-trained new intelligent modules can be deployed through the platform or backend management interface, or entirely new intelligent module instances can be created. Necessary operating parameters should be set for the new intelligent module, such as priority, processing permissions, and task scope. Ensure that the new intelligent module can seamlessly integrate into the multi-intelligent module architecture and collaborate with existing intelligent modules and tools within the architecture.

[0143] Optionally, in the event of a reduction in intelligent modules, without affecting the normal operation of the multi-intelligent module architecture, the intelligent module to be deleted is removed from the processing queue and service chain. Data related to the intelligent module to be deleted is cleaned up, and the computing resources occupied by the intelligent module are released. The workload and scope of responsibility of other intelligent modules in the multi-intelligent module architecture are adjusted to fill the gap left by the deleted intelligent module, ensuring that task continuity and service quality are not affected.

[0144] Optionally, operation logs for adding or removing intelligent modules are recorded, and the intelligent module directory and status database are updated. The operational status after changes to the multi-intelligent module architecture is closely monitored, including processing speed, response quality, and resource utilization, to verify whether the execution effect of intelligent module configuration requests has achieved the expected goals. Based on the above monitoring results, the intelligent module configuration can be further adjusted, or the intelligent modules can be trained and upgraded to continuously optimize processing efficiency and service quality.

[0145] In this embodiment, by dynamically adjusting the composition of intelligent modules, the system can quickly adapt to changing task environments, whether dealing with seasonal service peaks or new special products or user groups. Adding high-efficiency, high-performance intelligent modules in a timely manner, and reducing redundant or inefficient modules, helps improve the speed and quality of after-sales service, reduces user waiting time, and increases user satisfaction. Reasonably increasing or decreasing the number of intelligent modules avoids unnecessary resource waste and helps enterprises better control operating costs, especially when task volume fluctuates significantly. The intelligent module configuration request mechanism supports the continuous development and optimization of tasks, ensuring that the multi-intelligent module architecture always maintains effective service capabilities by continuously introducing new intelligent modules with new functions or phasing out older versions.

[0146] The embodiments of this application will be further explained below for the after-sales service scenario of e-commerce platforms.

[0147] Currently, after-sales service demands are becoming increasingly diversified, but the rigid Standard Operating Procedures (SOPs) of related technologies struggle to address users' changing needs, resulting in low user interaction efficiency and poor user experience. With advancements in artificial intelligence, multi-intelligent module architectures, as a new technological direction in the field of intelligent customer service, can achieve task division, specialized breakthroughs, and flexible interactive responses. Existing intelligent customer service systems typically employ single agents or static task allocation methods, making it difficult to balance standard SOP processes with flexible responses. Key drawbacks include: general agents have limited ability to handle specific types of problems and special user groups, resulting in poor personalization; static SOP configurations lead to low flexibility in task scenarios, complex scheduling, and difficulty adapting to new demands; a lack of multi-layered intelligent module collaboration results in weak result fusion capabilities, making it difficult to provide comprehensive and suitable solutions; and poor response to complex user work orders and multi-round dynamic interactions makes it difficult to provide high-quality answers. Therefore, the technical problem of poor service request processing remains.

[0148] However, this application proposes a multi-intelligent module collaborative after-sales service hosting solution. By accurately identifying the target service scenario corresponding to the service request, it ensures that the service request can be directed to a suitable first target intelligent module for processing. The first target intelligent module can provide customized responses for the corresponding target service scenario, compensating for the lack of versatility of single agents in related technologies. The management intelligent module integrates and optimizes the first processing result generated by the first intelligent module, generating a more comprehensive and accurate service processing response. The above method overcomes the limitations of static routing and single agents in handling complex service requests in related technologies, enhancing the user experience. This achieves the technical effect of improving the processing effect of service requests and solves the technical problem of poor service request processing effect in related technologies.

[0149] In this embodiment, Figure 3 This is a schematic diagram illustrating a multi-intelligent module collaborative after-sales service outsourcing solution according to an embodiment of this application, such as... Figure 3 As shown, the user sends an after-sales request to the system entry point. Users initiate after-sales service requests via voice chat (such as telephone, instant messaging software) or text chat (such as online customer service chat windows, social media private messages). Service requests can include various types such as returns, exchanges, inquiries, and complaints. Upon receiving such a service request, the Standard Operating Procedure (SOP) process can be executed. During the execution of the SOP process, keywords can be extracted based on the request content to determine the type of service request. Once the target service request is received, the processing flow of reply answer + action can be executed sequentially according to the preset SOP process. This means that the multi-intelligent module architecture will reply to the user according to fixed rules or answer templates in the database and perform corresponding processing operations, such as order status updates and refund processing. If the service request requires multiple rounds of interaction to resolve, the above reply answer + action process can be repeated until the service request is resolved or the processing limit is reached. In the SOP process, corresponding generated answer + action text can be generated based on the above reply answer + action. The answer refers to the direct reply content prepared based on the service request, which can be in various forms such as text, links, and images. Action text can describe the operational instructions that should be taken within the multi-intelligent module architecture, providing clear guidance for the Agent to perform subsequent actions, such as updating order status or sending email notifications.

[0150] Optionally, such as Figure 3As shown, the above answer + action text can be passed to a specific Agent, which then makes intelligent decisions based on real-time task rules and user characteristics. Upon receiving the answer + action text, the Agent executes the corresponding action, such as sending a message to the user or updating the backend system status. The scenario routing module parses the user's intent and routes the request to the corresponding scenario intelligent module (Scenario Intelligent Module 1, Scenario Intelligent Module 2, Scenario Intelligent Module 3, ...) based on different products, question types, and user profiles. The admission module controls whether access to a specific intelligent module (Specialized Intelligent Module a, Specialized Intelligent Module b, Specialized Intelligent Module c) is granted. These specialized intelligent modules provide solutions for specific product pools and specific user groups. Each specialized intelligent module / scenario intelligent module outputs its own suggestions, processing solutions, and response content. During request processing, the intelligent modules can invoke a series of tools (Tools 1 to 4), such as order information queries, user credit assessments, and logistics status tracking, to assist in decision-making and provide more comprehensive services. The management intelligent module is responsible for collecting and integrating the outputs of various specialized intelligent modules and scenario intelligent modules, dynamically selecting / merging answers based on priority, task strategies, etc., and ultimately replying to the user.

[0151] In summary, the after-sales service hosting solution based on multi-intelligent module collaboration described in this application proposes a parallel architecture that separates task logic and intent. The task policy agent is subject to fully parallel admission based on admission conditions; the scenario agent is driven by user intent, dynamically routing to the corresponding sub-agent based on the intent. Furthermore, based on the multi-intelligent module architecture, a Standard Operating Procedure (SOP) process is introduced. By modifying the action nodes of the SOP leaf nodes into a text + action format, the decision-making and execution are handled by the agents within the multi-intelligent module architecture.

[0152] In this embodiment, the Agent module is highly scalable, allowing operations teams to add or remove specialized sub-Agents and configure rules independently without affecting the overall SOP process. It supports parallel collaboration among multiple agents, and the Supervisor Agent can dynamically merge data, not being limited to single-rule decisions. That is, the Supervisor Agent combines answers from different sources for comprehensive analysis to arrive at the final answer. The intelligent routing algorithm automatically assigns tasks to the appropriate Agents based on real-time task / user characteristics.

[0153] It should be noted that, for the sake of simplicity, the aforementioned method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited by the order of the actions described, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0154] Based on the information from the above embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, they can also be implemented by hardware. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software object. This computer software object is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0155] According to an embodiment of this application, a service request processing apparatus for implementing the above-described service request processing method is also provided. Figure 4 This is a schematic diagram of a service request processing apparatus according to an embodiment of this application, such as... Figure 4 As shown, the device 400 includes: a first determining module 402, a second determining module 404, a processing module 406, and a generating module 408.

[0156] The first determining module 402 is used to determine at least one target service scenario corresponding to the service request in response to the service request for the target object; the second determining module 404 is used to determine the first target intelligent module corresponding to the target service scenario from a plurality of first intelligent modules, wherein different first intelligent modules are used to process service requests corresponding to different service scenarios; the processing module 406 is used to process the service request using the first target intelligent module to obtain a first processing result corresponding to the first target intelligent module; and the generating module 408 is used to generate a service processing response corresponding to the service request using the management intelligent module and the first processing result.

[0157] It should be noted that the first determining module 402, the second determining module 404, the processing module 406, and the generating module 408 mentioned above correspond to steps S202 to S208 in the above embodiments. The four modules and their corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules or units can be hardware or software components stored in memory and processed by one or more processors. The above modules can also run as part of the device in the server 10 provided in the above embodiments.

[0158] It should be noted that the preferred implementation schemes involved in the above embodiments of this application are the same as the schemes, application scenarios and implementation processes provided in the above embodiments, but are not limited to the schemes provided in the above embodiments.

[0159] According to embodiments of this application, a service request processing system for implementing the above-described service request processing method is also provided. Figure 5 This is a schematic diagram of a service request processing system according to an embodiment of this application, such as... Figure 5 As shown, the system 500 includes a routing module 502 and a processing module 504.

[0160] The routing module 502 is used to respond to a service request for a target object, determine at least one target service scenario corresponding to the service request, and determine the first target intelligent module corresponding to the target service scenario from multiple first intelligent modules; the processing module 504 is used to call the first target intelligent module to process the service request, obtain the first processing result corresponding to the first target intelligent module, call the management intelligent module to merge the first processing result, and generate a service processing response corresponding to the service request.

[0161] The system may further include a filtering module, which is used to determine at least one second target intelligent module corresponding to the service request from multiple second intelligent modules based on the target object and the target user corresponding to the service request. The second target intelligent module is a second intelligent module that matches the target object and / or the target user. The system may further include a processing module, which is used to call the second target intelligent module to process the service request, obtain a second processing result corresponding to the second target intelligent module, and call the management intelligent module to merge the first processing result and the second processing result to generate a service processing response.

[0162] Embodiments of this application may provide a computing device. Figure 6 This is a structural block diagram of a computing device according to an embodiment of this application. Figure 6As shown, the computing device 600 may include: one or more (one shown in the figure) processors 602, memory 604, memory controller, and peripheral interfaces.

[0163] The aforementioned computing device can be understood as an integrated intelligent terminal, including but not limited to servers, desktop computers, personal computers (PCs), and all-in-one model machines. Furthermore, the computing device may have the model described in the above embodiments of this application pre-installed.

[0164] Specifically, this computing device can pre-install various types of models, including but not limited to models in fields such as natural language processing, visual processing, speech processing, code processing, and multimodal task processing, thus providing diverse model choices. In different product forms, this computing device can support one or more model usage methods, including but not limited to model training, model invocation, model fine-tuning, model deployment, model inference, and application. In some product forms, this computing device also supports model management, including but not limited to multi-type model management (supporting the management of discriminative, generative, and other model types), model version control (supporting the control of different model versions), and model evaluation (evaluating model performance and effectiveness based on model evaluation tools). In other product forms, this computing device can also create applications based on models, providing API calling capabilities. Models can be called into created applications through API interfaces, and application management tools are provided to achieve application control.

[0165] Furthermore, this computing device can also include data management (supporting the creation and management of model tuning datasets), a training center (providing abundant training resources to help users learn and master AI technologies), and basic control capabilities (providing enterprise-level basic control capabilities to ensure system security and efficient operation). Through these functions, it provides a comprehensive, integrated device for AI development, training, deployment, and application.

[0166] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the methods in the above embodiments. The memory may include high-speed random access memory (RAM) and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memories. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to terminal A via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks (LANs), mobile communication networks, and combinations thereof.

[0167] The processor can invoke an executable program stored in memory via a transmission device to execute any of the methods described in the above embodiments.

[0168] Embodiments of this application may provide an electronic device. Figure 7 This is a structural block diagram of an electronic device according to an embodiment of this application, such as... Figure 7 As shown, the electronic device may include: an input / output device 72; a memory 74; and a processor 76, wherein the processor 76 is connected to the input / output device 72 and the memory 74 via a bus 78.

[0169] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the methods in the above embodiments. The memory may include high-speed random access memory (RAM) and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memories. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to terminal A via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks (LANs), mobile communication networks, and combinations thereof.

[0170] The processor can invoke an executable program stored in memory via a transmission device to execute any of the methods described in the above embodiments.

[0171] Those skilled in the art will understand that, Figure 7 The structure shown is illustrative. Electronic devices can also be smartphones (such as Android phones, iOS phones, etc.), tablets, PDAs, mobile internet devices (MIDs), PADs, and other terminal devices. This diagram does not limit the structure of the aforementioned electronic devices. For example, electronic devices may include more or fewer components (such as network interfaces, display devices, etc.) than shown in the diagram, or have a different configuration than shown in the diagram.

[0172] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. This program can be stored in a computer-readable storage medium, which may include: a flash drive, read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0173] Embodiments of this application also provide a computer-readable storage medium. Optionally, in this embodiment, the aforementioned computer-readable storage medium can be used to store program code executed by the method provided in the above embodiments.

[0174] Optionally, in this embodiment, the storage medium may be located in a computing device or an electronic device.

[0175] Optionally, in this embodiment, the computer-readable storage medium is configured to store an executable program. When the executable program runs, it controls the device where the computer-readable storage medium is located to perform any of the methods described in the above embodiments.

[0176] Embodiments of this application also provide a computer program product. Optionally, in this embodiment, the computer program product may include a computer program. When executed by a processor, the computer program implements the methods provided in the above embodiments.

[0177] Embodiments of this application also provide a computer program product. Optionally, the computer program product may include a non-volatile computer-readable storage medium. The non-volatile computer-readable storage medium can be used to store a computer program. When the computer program is executed by a processor, it implements the method provided in the above embodiments.

[0178] Embodiments of this application also provide a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, it implements the method provided in the above embodiments.

[0179] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0180] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are illustrative; for example, the division of units is a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined, integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling, direct coupling, or communication connection shown or discussed may be through some interfaces, indirect coupling of units or modules, or communication connection, and may be electrical or other forms.

[0181] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of this embodiment.

[0182] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0183] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, ROM, RAM, portable hard drives, magnetic disks, or optical disks.

[0184] The above description represents the preferred embodiments of this application. For those skilled in the art, various improvements and modifications can be made without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for processing service requests, characterized in that, include: In response to a service request for a target object, at least one target service scenario corresponding to the service request is determined; The first target intelligent module corresponding to the target service scenario is determined from a plurality of first intelligent modules, wherein different first intelligent modules are used to process service requests corresponding to different service scenarios; The service request is processed using the first target intelligence module to obtain a first processing result corresponding to the first target intelligence module; Using the management intelligence module and the first processing result, a service processing response corresponding to the service request is generated.

2. The method according to claim 1, characterized in that, Determining at least one target service scenario corresponding to the service request includes: Identify the target user corresponding to the service request; The service request is subjected to intent recognition to determine the request type corresponding to the service request; The target service scenario is determined based on the target object, the request type, and the target user.

3. The method according to claim 1, characterized in that, The step of processing the service request using the first target intelligence module to obtain a first processing result corresponding to the first target intelligence module includes: The first target intelligence module is used to determine the target node that matches the service request from multiple operation process nodes. The target node includes: response data and action text data of the target operation process. The target operation process is the operation process represented by the target node. The first target intelligence module is used to make action decisions on the response data and the action text data to generate a target action; The first target intelligent module is used to execute the target action to obtain the first processing result.

4. The method according to claim 1, characterized in that, The method further includes: Based on the target object and the target user corresponding to the service request, at least one second target intelligent module corresponding to the service request is determined from a plurality of second intelligent modules, wherein the second target intelligent module is a second intelligent module that matches the target object and / or the target user; The service request is processed using the second target intelligence module to obtain a second processing result corresponding to the second target intelligence module; The step of generating a service processing response corresponding to the service request using the management intelligence module and the first processing result includes: The management intelligence module is used to fuse the first processing result and the second processing result to generate the service processing response.

5. The method according to claim 4, characterized in that, The step of determining the second target intelligent module corresponding to the service request from multiple second intelligent modules based on the target object and the target user corresponding to the service request includes: Obtain the filtering conditions corresponding to the second intelligent module; The target object and the target user are matched with multiple filtering conditions to obtain target filtering conditions that match the target object and / or the target user; The second intelligent module corresponding to the target screening condition is determined to be the second target intelligent module.

6. The method according to claim 5, characterized in that, The step of matching the target object and the target user with multiple filtering conditions to obtain target filtering conditions that match the target object and / or the target user includes: The object information of the target object is matched with the object information in the filtering conditions to obtain the first filtering condition that matches the target object; The user information of the target user is matched with the user information in the filtering conditions to obtain a second filtering condition that matches the target user; The target filtering conditions are obtained based on the first filtering condition and / or the second filtering condition.

7. The method according to claim 4, characterized in that, The step of using the management intelligence module to fuse the first processing result and the second processing result to generate the service processing response includes: Obtain the first priority of the first target intelligent module and the second priority of the second target intelligent module; Based on the first priority, the second priority, and the preset processing strategy, at least one candidate intelligent module is determined from the first target intelligent module and the second target intelligent module, wherein the candidate intelligent module satisfies the preset processing strategy and the priority of the candidate intelligent module is greater than the preset priority; The management intelligence module is used to fuse the processing results corresponding to the candidate intelligence modules to generate the service processing response.

8. The method according to claim 7, characterized in that, The step of using the management intelligence module to fuse the processing results corresponding to the candidate intelligence modules and generate the service processing response includes: At least one tool is used to process the service request and obtain the service information corresponding to the service request. By using the management intelligence module and the service information, the processing results corresponding to the candidate intelligence modules are fused to obtain fused information; The service processing response is generated using the management intelligence module and the fused information.

9. The method according to any one of claims 4 to 8, characterized in that, The method further includes: In response to the intelligent module configuration request, the plurality of second intelligent modules are added or removed based on the intelligent module configuration request.

10. A service request processing system, characterized in that, include: The scenario routing module is used to respond to a service request for a target object, determine at least one target service scenario corresponding to the service request, and determine the first target intelligent module corresponding to the target service scenario from a plurality of first intelligent modules, wherein different first intelligent modules are used to process service requests corresponding to different service scenarios; The processing module is used to call the first target intelligent module to process the service request, obtain the first processing result corresponding to the first target intelligent module, and call the management intelligent module to merge the first processing result to generate the service processing response corresponding to the service request.

11. The system according to claim 10, characterized in that, The system also includes: A filtering module is used to determine at least one second target intelligent module corresponding to the service request from a plurality of second intelligent modules based on the target object and the target user corresponding to the service request, wherein the second target intelligent module is a second intelligent module that matches the target object and / or the target user; The processing module is further configured to call the second target intelligent module to process the service request, obtain the second processing result corresponding to the second target intelligent module, and call the management intelligent module to merge the first processing result and the second processing result to generate the service processing response.

12. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor, connected to a memory via a bus, is used to run the program, wherein the program, when running, executes the method described in any one of claims 1 to 9.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1 to 9.

14. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 9.

Images