Question and answer task processing method and device, equipment, storage medium and program product

Abstract

The invention discloses a question and answer task processing method and device, equipment, a storage medium and a program product, and relates to the technical field of intelligent question and answer, and the question and answer task processing method comprises the following steps: receiving a to-be-processed task question input by a user in a session; determining a task scene type according to the to-be-processed task problem, and extracting task parameters of the to-be-processed task problem; and according to the task parameters, processing the to-be-processed task problem through a target agent corresponding to the task scene type to obtain a processing reply result. According to the method, multi-agent-based question and answer task processing is realized, the problem that an existing question and answer task processing method is low in processing efficiency in a complex task scene is solved, and the efficiency and accuracy of question and answer task processing are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent question-answering technology, and in particular to a question-answering task processing method, apparatus, device, storage medium, and program product. Background Art

[0002] With the popularization of artificial intelligence technology, intelligent question-answering systems usually use a single general model to handle all task problems when dealing with diverse and complex user task requirements, or pre-divide several fixed-function modules based on artificial rules to cover different scenarios.

[0003] When dealing with complex tasks involving multiple fields and multiple parameters, existing technologies may cause inefficient processing or even incorrect responses due to calling mismatched functions, thereby affecting the processing efficiency of the question-answering system when handling complex tasks.

[0004] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide a question-answering task processing method, device, equipment, storage medium and program product, aiming to solve the technical problem of low processing efficiency of existing question-answering task processing methods in complex task scenarios.

[0006] To achieve the above objectives, this application proposes a question-answering task processing method, which is applied to a question-answering task processing platform. The question-answering task processing platform is configured with several intelligent agents, including:

[0007] Receive pending task questions entered by the user in the session;

[0008] Determine the task scenario type according to the task problem to be processed, and extract the task parameters of the task problem to be processed;

[0009] According to the task parameters, the target intelligent agent corresponding to the task scenario type processes the task problem to be processed to obtain a processing response result.

[0010] In one embodiment, before the step of processing the pending task problem by the agent corresponding to the task scenario type according to the task parameters and obtaining a processing response result, the step further includes:

[0011] Determining whether the task parameters meet the necessary task parameter requirements of the target agent;

[0012] If the task parameters do not meet the necessary task parameter requirements, a set of parameters to be supplemented is obtained based on the difference between the necessary task parameter requirements and the task parameters;

[0013] According to the set of parameters to be supplemented, multiple rounds of dialogue are conducted with the user through the session to obtain supplemented task parameters, so that according to the supplemented task parameters, the task problem to be processed is processed by the intelligent agent corresponding to the task scenario type to obtain a processing response result.

[0014] In one embodiment, the step of conducting multiple rounds of dialogue with the user through the conversation based on the set of parameters to be supplemented to obtain supplemented task parameters includes:

[0015] According to the set of parameters to be supplemented, generating a number of parameter supplementation sub-problems;

[0016] Sending the plurality of parameter supplementation sub-questions to the session, and obtaining parameter supplementation answers returned by the user in the session;

[0017] The supplementary task parameters in the parameter supplement answer are extracted, and the supplementary task parameters are added to the task parameters to obtain supplemented task parameters.

[0018] In one embodiment, before the step of processing the pending task problem by the agent corresponding to the task scenario type according to the task parameters and obtaining a processing response result, the step further includes:

[0019] Based on a preset parameter dependency graph, detecting whether there is a parameter conflict among the task parameters, wherein the parameter dependency graph is constructed based on a preset parameter logical dependency relationship;

[0020] If there is a parameter conflict among the task parameters, an alternative resource is allocated to the task parameters through a preset resource pool, and a number of candidate task parameters are generated according to the associated parameters of the alternative resource;

[0021] In the conversation, the candidate task parameters are displayed to the user through visual interactive cards, so that the task problem to be processed is processed by the intelligent agent corresponding to the task scenario type according to the candidate task parameters selected by the user to obtain a processing response result.

[0022] In one embodiment, the target agent is connected to a plurality of functional interfaces, and the step of processing the pending task problem by the target agent corresponding to the task scenario type according to the task parameters to obtain a processing response result includes:

[0023] The task parameters are sent to the target agent for processing as follows:

[0024] Perform task planning according to the task parameters to obtain several subtasks;

[0025] Processing the plurality of subtasks through the plurality of functional interfaces to obtain task processing results;

[0026] Based on the task processing result, a processing reply result is generated.

[0027] In one embodiment, the task scenario type is a conference room reservation scenario, and the target agent is connected to several functional interfaces, which include one or more of a query interface, a reservation interface, and a notification interface.

[0028] The step of processing the pending task problem by the target agent corresponding to the task scenario type according to the task parameters to obtain a processing response result includes:

[0029] The task parameters are sent to the target agent of the conference room predetermined scene for the following processing:

[0030] According to the task parameters, querying whether there is an available conference room through the query interface;

[0031] If the available conference room exists, booking the available conference room through the booking interface to obtain a conference room booking result;

[0032] If the available conference room is successfully reserved, a conference notification is sent to the participants set in the task parameters through the notification interface according to the conference room reservation result to obtain a conference notification result;

[0033] A processing reply result is generated based on the conference room reservation result and the conference notification result.

[0034] In addition, to achieve the above objectives, the present application also proposes a question-answering task processing device, which includes:

[0035] The receiving module is used to receive the pending task questions input by the user in the session;

[0036] An extraction module is used to determine the task scenario type according to the task problem to be processed, and extract the task parameters of the task problem to be processed;

[0037] The processing module is used to process the pending task problem according to the task parameters through the target intelligent agent corresponding to the task scenario type to obtain a processing response result.

[0038] In addition, to achieve the above-mentioned purpose, the present application also proposes a question-answering task processing device, which includes: a memory, a processor, and a computer program stored on the memory and runnable on the processor, and the computer program is configured to implement the steps of the question-answering task processing method described above.

[0039] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium. A computer program is stored on the storage medium, and when the computer program is executed by the processor, the steps of the question-and-answer task processing method described above are implemented.

[0040] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the question-and-answer task processing method described above.

[0041] This application provides a method for processing question-and-answer tasks. First, the task question entered by the user in the conversation is received to ensure that the platform can capture the user's needs. Then, the task scenario of the task question is determined and the task parameters are extracted to quickly classify the user question into the correct task processing scenario. Finally, the target intelligent agent is used for processing. By quickly and accurately identifying user needs, classifying the question into the appropriate scenario, and calling the corresponding intelligent agent for processing, the user is provided with a more timely and effective task processing response. This solves the problem of low processing efficiency of existing question-and-answer task processing methods in complex task scenarios and improves the efficiency and accuracy of question-and-answer task processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0044] Figure 1 A flowchart illustrating the first embodiment of the question-answering task processing method of this application;

[0045] Figure 2 A flowchart of the second embodiment of the question-answering task processing method of this application is provided;

[0046] Figure 3 This is a schematic diagram of the module structure of the question-answering task processing device according to an embodiment of the present application;

[0047] Figure 4 This is a schematic diagram of the device structure of the hardware operating environment involved in the question-and-answer task processing method in the embodiment of this application.

[0048] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0049] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0050] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0051] The main solution of the embodiment of the present application is: receiving the pending task problem input by the user in the conversation; determining the task scenario type based on the pending task problem, and extracting the task parameters of the pending task problem; based on the task parameters, processing the pending task problem through the target intelligent agent corresponding to the task scenario type to obtain a processing reply result.

[0052] Existing technologies typically use a single, universal model to handle all tasks, or pre-divide several fixed-function modules (such as classifiers and semantic parsing modules) based on artificial rules to cover different scenarios. This lacks the ability to dynamically perceive and adapt to task scenarios, leading to rigid resource allocation. For example, a single model struggles to optimize performance for different scenarios, while fixed module divisions prevent dynamic adjustment of processing logic based on task parameters. This is especially true for complex problems involving multiple domains and intersecting multiple parameters. The system may inefficiently process or even provide incorrect responses due to invoking incompatible modules or models.

[0053] In this application, the platform first receives the task questions entered by the user in the conversation, ensuring that the platform can capture the user's needs. Then, the task scenario of the task question is determined and the task parameters are extracted, quickly classifying the user question into the correct task processing scenario. Finally, the target intelligent agent is used for processing. By quickly and accurately identifying user needs, classifying the question into the appropriate scenario, and calling the corresponding intelligent agent for processing, the user is provided with more timely and effective task processing responses. This solves the problem of low processing efficiency of existing question-and-answer task processing methods in complex task scenarios, and improves the efficiency and accuracy of question-and-answer task processing.

[0054] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication, and program execution capabilities, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device capable of performing the above functions, a question-and-answer task processing device, etc. The following uses a question-and-answer task processing device as an example to illustrate this embodiment and the following embodiments.

[0055] Based on this, the embodiment of the present application provides a question-answering task processing method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the question-answering task processing method of this application.

[0056] In this embodiment, the question-answering task processing method is applied to a question-answering task processing platform, which is configured with several intelligent agents, and includes steps S10 to S30:

[0057] Step S10, receiving the pending task problem input by the user in the session;

[0058] It's important to note that a user refers to an individual who interacts with the Q&A task processing platform, potentially accessing the platform through a chat interface, voice assistant, or other means. A conversation is the continuous interaction between a user and the Q&A task processing platform, typically conducted in natural language. This can take the form of a chat window, voice dialogue, or other means, and serves as a platform for users to input questions for tasks. Questions are specific questions entered by users that the Q&A task processing platform needs to address. These questions can cover a variety of scenarios, such as conference room reservations and inquiries, and represent user requests for answers or actions to be performed.

[0059] It can be understood that receiving the task problem input by the user ensures the smooth progress of the entire task processing process, provides processing objects for subsequent steps, and enables the system to perform subsequent operations based on the specific needs of the user.

[0060] Specifically, the question-and-answer task processing platform can receive user input in a variety of ways, such as accepting task questions entered by the user via a text input box on a webpage, or using a voice input module on a mobile app to convert the user's voice into text. The receiving module may need to have corresponding adaptation capabilities for different input methods. For example, text input requires a text parsing interface, while voice input requires voice recognition technology and a corresponding text-to-text processing process. Using different receiving methods can meet user usage habits in different scenarios, improving system usability and convenience.

[0061] For example, suppose a user enters "I want to book a conference room that can accommodate 10 people and has projector equipment at 10 am tomorrow" in the company's internal smart office question-and-answer system. The system receives this text information through the input box on the web page and treats it as a task question to be processed in text form.

[0062] Step S20, determining the task scenario type according to the task problem to be processed, and extracting the task parameters of the task problem to be processed;

[0063] It should be noted that task scenario types refer to the different fields or context categories to which the problem belongs, such as reservation scenarios, query scenarios, etc. Task parameters are the key information involved in the problem, such as time, location, and other elements in the reservation scenario.

[0064] It is understandable that by performing semantic analysis on the problem, classifying it into the corresponding scenario, and extracting key parameters so that the appropriate intelligent agent can be accurately selected for subsequent processing, the mismatch problem that may arise when a single general model handles all tasks can be avoided, the targeted processing can be improved, and thus help solve the problem of low processing efficiency in complex task scenarios.

[0065] Specifically, the task scenario type can be determined by building a scenario classification model and using machine learning algorithms to train a large number of question texts with annotated scenarios. When a new question is input, the model predicts the scenario type to which it belongs, or it can be achieved through the intent recognition model in natural language processing technology. The model has been trained with a large amount of corpus and can understand the intention of the user's question and classify it into the corresponding scenario type. To extract task parameters, rule-based extraction methods can be used, such as regular expressions matching information in a specific format (such as date, time, etc.), or using named entity recognition technology to identify key entities in the question (such as people, place names, etc.). For example, for a conference room reservation scenario, the scenario type can be determined by rule matching keywords such as "reservation" and "conference room", and named entity recognition can be used to extract task parameters such as conference room capacity, meeting time, and equipment configuration.

[0066] For example, in response to the user input "I want to reserve a conference room that can accommodate 10 people at 10 am tomorrow with projector equipment", the system determines the task scenario type as "conference room reservation scenario" through text analysis, and extracts task parameters including the number of participants (10 people), meeting time (tomorrow morning 10 am), and equipment configuration requirements (projector).

[0067] Step S30: According to the task parameters, the target agent corresponding to the task scenario type processes the task problem to be processed to obtain a processing response result.

[0068] It should be noted that the target intelligent agent is an intelligent processing unit configured for a specific task scenario in the question-answering task processing platform, and each intelligent agent has the ability to process tasks in the corresponding scenario.

[0069] It is understandable that the extracted task parameters are handed over to the target intelligent agent of the corresponding scenario for processing. The intelligent agent performs calculations, queries and other operations based on the parameters and finally obtains the response results. This can give full play to the professionalism of each intelligent agent, because different intelligent agents optimize the processing logic and algorithms for different scenarios, so that the system can solve problems efficiently and accurately in complex task scenarios, thereby improving the user experience.

[0070] Specifically, after receiving the task parameters, the target agent may call the corresponding database query interface, external service API, or perform internal logical operations to process the task. For example, after receiving parameters such as time and room, the conference room reservation agent queries the conference room reservation system database to check whether the corresponding room is available at that time. If it is available, it completes the reservation and generates a response. If it is not available, it may find an alternative solution based on the settings or inform the user that the reservation is unavailable.

[0071] In a feasible embodiment, before the step of processing the pending task problem by the agent corresponding to the task scenario type according to the task parameters and obtaining a processing response result, the step further includes:

[0072] Step S3011, determining whether the task parameters meet the necessary task parameter requirements of the target agent;

[0073] It should be noted that the necessary task parameter requirements refer to the parameter content that the target intelligent agent must have to complete task processing. These parameters are the basis for the normal operation of the intelligent agent and the production of reliable results.

[0074] It is understandable that checking whether the extracted task parameters are complete, accurate and meet the necessary parameter requirements for the intelligent agent to process the task can discover the deficiencies of the task parameters in advance, avoid directly handing over incomplete parameters to the intelligent agent for processing, resulting in errors or unprocessable situations, and ensure the smooth progress of subsequent processing procedures, improve the success rate and efficiency of task processing, and ensure that the answers given by the question-and-answer system are well-founded and accurate, thereby improving the reliability of the system in complex task processing.

[0075] Specifically, the judgment process can be implemented through parameter verification rules, which can be simple existence checks (i.e., whether the parameter exists) or checks on the parameter format, range, etc. For example, for a time parameter, check whether its format conforms to the standard format of date and time and whether it is within a reasonable time range.

[0076] Step S3012: If the task parameters do not meet the necessary task parameter requirements, a set of parameters to be supplemented is obtained based on the difference between the necessary task parameter requirements and the task parameters.

[0077] It should be noted that the difference set refers to the parameter set that exists in the necessary task parameter requirements but is missing in the task parameters. For example, if the necessary task parameter requirements include three parameters A, B, and C, and the task parameters have been extracted to A and B, then the difference set is C, and the parameter set to be supplemented includes C.

[0078] It can be understood that through set operations, the necessary task parameter requirement set is compared with the extracted task parameter set, the difference set is calculated, and the parameter set to be supplemented is obtained, which clarifies the supplementary information that needs to be obtained, and provides a basis for subsequently asking the user for supplementary parameters. It can collect missing parameters in a targeted manner, avoid blind questions, make multi-round conversations more efficient and purposeful, reduce unnecessary user operations, and improve the interaction efficiency and user experience of the question-and-answer system.

[0079] Step S3013: Based on the set of parameters to be supplemented, multiple rounds of dialogue are conducted with the user through the conversation to obtain supplemented task parameters, and based on the supplemented task parameters, the task problem to be processed is processed by the intelligent agent corresponding to the task scenario type to obtain a processing response result.

[0080] It's important to note that multi-round dialogue refers to the process of multiple exchanges between the question-answering system and the user around the same task, with the goal of obtaining more information or clarifying questions. The supplemented task parameters are the complete set of parameters formed by combining the missing parameters added by the user after multiple rounds of dialogue with the original parameters.

[0081] It is understandable that the system asks the user questions multiple times based on the set of parameters to be supplemented. After the user answers one by one, the system extracts and integrates the supplemented parameters, and then hands them over to the target intelligent agent for processing to obtain the final answer. Supplementing parameters through multiple rounds of dialogue can gradually improve the task parameters, which can ensure that the intelligent agent has enough information to accurately handle the task, avoid incorrect answers due to missing parameters, and improve the accuracy and effectiveness of the question-answering system in complex task processing. At the same time, it can also guide users to complete information and improve the naturalness and fluency of interaction.

[0082] Specifically, during multiple rounds of conversation, the system can ask users questions in various ways, such as directly asking questions or providing options. For example, if the number of people in a reservation is missing, the system can ask, "How many people do you need to reserve?" or provide common options for the user to choose from. After the user responds, the system uses natural language understanding technology to extract supplementary parameters and merge them with the original parameters.

[0083] In this implementation, the judgment of the integrity of task parameters and the supplementation of parameters through multiple rounds of dialogue can ensure that the parameters handed over to the target intelligent agent for processing are complete and meet the requirements, avoiding processing errors or failures due to missing parameters, further improving the accuracy and success rate of complex task processing, and optimizing the interactive experience between users and the question-answering system, so that the system can solve problems more intelligently and accurately when facing complex task scenarios.

[0084] In a feasible implementation, the step of conducting multiple rounds of dialogue with the user through the conversation based on the set of parameters to be supplemented to obtain supplemented task parameters includes:

[0085] Step S30131: generating a number of parameter supplementation sub-problems according to the set of parameters to be supplemented;

[0086] It's important to note that the parameter supplement sub-question is a specific question generated for each parameter in the set of parameters to be supplemented, guiding the user to complete the corresponding parameter information. For example, if the set of parameters to be supplemented is the meeting time and attendees, the generated sub-questions might be "Please tell me the specific time of the meeting" and "Please tell me the list of attendees."

[0087] As you can see, generating parameter-supplementing sub-questions based on missing parameters can transform them into questions that are easy for users to understand and answer, allowing them to accurately provide the required information. This improves the efficiency and accuracy of parameter supplementation, avoids user confusion caused by not knowing what to supplement, ensures smooth multi-round conversations, and ultimately safeguards the coherence and effectiveness of the entire task processing process.

[0088] Specifically, parameter-supplementing sub-questions can be generated using pre-set templates. Based on the parameter type and characteristics, the corresponding templates are filled in to generate questions. For example, for a time-based parameter, the template might be "Please inform me of the specific time for [parameter name]"; for a personnel list-based parameter, the template might be "Please list [parameter name]." Natural language generation technology can also be used to generate more natural and fluent questions based on the semantic information of the parameters. Different generation methods can be combined to ensure the standardization of questions while improving readability, making them easier for users to understand and answer.

[0089] For example, the set of parameters to be supplemented is device configuration and meeting type. In the conference room reservation scenario, the generated parameter supplement sub-questions may be "What equipment do you need in the conference room?" and "Is your meeting an online or offline meeting?"

[0090] Step S30132: sending the plurality of parameter supplementation sub-questions to the session, and obtaining parameter supplementation answers returned by the user in the session;

[0091] It should be noted that the parameter supplement answer is the user's response to the parameter supplement sub-question sent by the system, which may contain supplementary parameter information.

[0092] It can be understood that by communicating with the user in a conversation, sending the generated multiple parameter supplement sub-questions to the user, and receiving the user's answers to the parameter supplement questions, the user can provide the required information in a timely manner and obtain supplementary parameters, so that the system can dynamically collect the complete parameters required for task processing, enhance the interactivity and flexibility of the system, adapt to the changing parameters in complex task scenarios, ensure that task processing can continue to advance, and improve the adaptability and practicality of the system.

[0093] For example, in a conference room reservation scenario, the system sends two parameter-supplemented sub-questions to the user: "What equipment do you need in the conference room?" and "Is your meeting an online or offline meeting?" After seeing them, the user answers the equipment requirements and meeting type-related content in the conversation.

[0094] Step S30133: extract the supplementary task parameters in the parameter supplement answer, and add the supplementary task parameters to the task parameters to obtain the supplemented task parameters.

[0095] It should be noted that the supplementary task parameters are parameter contents extracted from the user's parameter supplementation answer and corresponding to the parameter set to be supplemented.

[0096] It is understandable that accurately extracting the valid parameter information in the user's supplementary answers and integrating it into the task parameters to form a complete parameter set, so that the parameters for task processing are complete and meet the requirements, can avoid errors in the intelligent agent's task processing due to parameter errors or improper integration, and ensure that the question-answering system can give correct answers based on accurate parameters.

[0097] Specifically, natural language processing techniques such as entity recognition and keyword extraction can be used to extract supplementary task parameters. For example, from the user's response "I need a projector and microphone," entity recognition can be used to extract "projector" and "microphone" as supplementary parameters for device configuration; and from the response "It's an offline meeting," "offline meeting" can be extracted as a supplementary parameter for the meeting type. When integrating these parameters into task parameters, ensure that the format and naming of the parameters are consistent with the original parameters so that the target agent can correctly recognize and use them.

[0098] In this implementation, by generating parameter supplementation sub-questions, sending questions to obtain user answers, and extracting and integrating supplementary parameters, the parameter supplementation process is made more standardized, efficient, and accurate, thereby improving the integrity of task parameters, ensuring that the intelligent agent can accurately handle problems in complex task scenarios, and enhancing the performance of the question-answering system in complex task processing.

[0099] In a feasible embodiment, before the step of processing the pending task problem by the agent corresponding to the task scenario type according to the task parameters and obtaining a processing response result, the step further includes:

[0100] Step S3021: detecting whether there is a parameter conflict among the task parameters based on a preset parameter dependency graph, wherein the parameter dependency graph is constructed based on a preset parameter logical dependency relationship;

[0101] It should be noted that the parameter dependency graph is a structure that describes the logical dependency relationship between parameters and stipulates the rules for mutual influence and mutual restriction between parameters. For example, in the conference room reservation scenario, the parameter logical dependency relationship may include the association constraints between conference room capacity and equipment configuration, and the conflict detection rules between time range and available resources.

[0102] It is understandable that by detecting parameter conflicts through the parameter dependency graph, that is, checking whether the extracted task parameters violate the logical dependency relationship between the parameters, it is possible to discover unreasonable aspects of the task parameters in advance, avoid handing over conflicting parameters to the intelligent agent for processing, and thus avoid incorrect results or situations where processing cannot be carried out. This can improve the accuracy and reliability of task processing, ensure that the question-answering system gives correct answers based on reasonable parameters, enhance the quality control of the system in complex task processing, and reduce errors caused by parameter logic problems.

[0103] Specifically, constructing a parameter dependency graph involves analyzing the business rules and logic within the task scenario, representing the dependencies between parameters in the form of a graph. Each node represents a parameter, and edges represent the dependencies and rules between parameters. For example, in a conference room reservation scenario, there's an edge between the number of attendees node and the conference room capacity node. The rule on this edge is that the number of attendees cannot exceed the conference room capacity. When detecting parameter conflicts, each task parameter is checked one by one according to the rules in the graph.

[0104] For example, in a conference room reservation scenario, the number of participants in the task parameters is 20, while the conference room capacity is 15 people. According to the rule in the parameter dependency graph that the number of participants cannot exceed the conference room capacity, the system can detect that there is a parameter conflict.

[0105] Step S3022: If there is a parameter conflict in the task parameters, allocate alternative resources for the task parameters through a preset resource pool, and generate a number of candidate task parameters based on the associated parameters of the alternative resources;

[0106] It's important to note that a resource pool is a collection of alternative resources, such as conference rooms and equipment. Alternative resources are available resources selected to address conflicting parameters. Associated parameters are other parameters associated with the alternative resource. For example, if a larger conference room is selected as an alternative resource, its associated parameters might include its location and equipment configuration. Candidate task parameters are several possible task parameter combinations generated based on the alternative resources and their associated parameters, for users to choose from.

[0107] Understandably, when parameter conflicts arise, the system selects alternative resources from the resource pool and generates multiple candidate task parameters based on their associated parameters, providing users with more reasonable processing options. By allocating alternative resources and generating candidate task parameters, users can be provided with personalized solutions, avoiding task completion failures due to parameter conflicts. This improves the system's flexibility and adaptability, meeting the diverse needs of users in complex task scenarios and enhancing user satisfaction. It also demonstrates the system's intelligence, proactively providing solutions when problems arise.

[0108] Specifically, the resources in the resource pool can be pre-classified and tagged to facilitate quick search for alternative resources. For example, in a conference room reservation scenario, the conference rooms in the resource pool can be classified according to attributes such as capacity, location, and equipment configuration. When alternative resources are needed, the system can filter out conference rooms with larger capacity from the resource pool as alternative resources based on conflicting parameters (such as insufficient conference room capacity). Then, based on the associated parameters of the alternative conference room (such as location in another building, more advanced equipment configuration, etc.), different candidate task parameter combinations are generated, such as Candidate 1: Conference Room XX (capacity 30 people, location: Building B, equipment configuration: projector, computer, camera), Candidate 2: Conference Room YY (capacity 25 people, location: Building A, equipment configuration: projector, computer), etc.

[0109] Step S3023, in the conversation, the candidate task parameters are displayed to the user through a visual interactive card, so that the task problem to be processed is processed by the intelligent agent corresponding to the task scenario type according to the candidate task parameters selected by the user to obtain a processing response result.

[0110] It should be noted that a visual interactive card is a carrier that displays information to users in an intuitive and friendly interface during a conversation. It usually contains elements such as text, pictures, buttons, etc., which are convenient for users to view and operate.

[0111] As you can understand, by presenting candidate task parameters to users through visual interactive cards, users can select a solution based on their needs. The system then performs subsequent processing based on the user's selected parameters to obtain a response result. Using visual interactive cards to present candidate task parameters allows users to more intuitively understand the available options, facilitates comparison and selection, and improves the convenience and user experience of user interaction. This can make communication between the system and users more efficient, ensuring that users can choose the appropriate solution based on their actual situation, and improving the system's convenience, flexibility, and user satisfaction in handling complex task scenarios.

[0112] Specifically, the design of visual interactive cards can be diverse. For example, they can display detailed information about each candidate task parameter in a list format, including parameter content and corresponding resource characteristics, and set a selection button for each candidate solution. Users can click the button to select the corresponding solution. At the same time, you can also add brief explanatory text to the card to help users quickly understand the advantages and disadvantages of each candidate solution.

[0113] For example, in the conference room reservation scenario, the system presents two candidate task parameter combinations to the user through visual interactive cards. The cards clearly list the capacity, location, equipment configuration and other information of each candidate conference room. After the user clicks to select one of them, the system will process the conference room reservation based on the parameters and reply to the user with the reservation result.

[0114] In this implementation, parameter conflicts are detected based on the parameter dependency graph, and alternative resources are allocated to generate candidate parameters for users to choose from. This can effectively solve the problem of logical conflicts in task parameters and avoid erroneous processing. By providing users with reasonable alternatives to screen and allowing users to make their own choices, the system's intelligence level and user experience in complex task processing are further improved.

[0115] This embodiment provides a method for processing question-and-answer tasks. First, the platform receives task questions entered by users in a conversation, ensuring that the platform captures the user's needs. Then, the task scenario of the task question is determined and task parameters are extracted, quickly categorizing the user's question into the correct task processing scenario. Finally, the method utilizes a target agent for processing. By quickly and accurately identifying user needs, categorizing the question into the appropriate scenario, and invoking the corresponding agent for processing, the method provides users with more timely and effective task processing responses. This method addresses the low processing efficiency of existing question-and-answer task processing methods in complex task scenarios, improving the efficiency and accuracy of question-and-answer task processing.

[0116] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the first embodiment can be referred to the above introduction and will not be described in detail later. Figure 2 , Figure 2This is a flow chart of the second embodiment of the question-answering task processing method of this application.

[0117] In this embodiment, the target agent is connected to a plurality of functional interfaces, and the step of processing the pending task problem by the target agent corresponding to the task scenario type according to the task parameters to obtain a processing response result includes:

[0118] The task parameters are sent to the target agent for processing as follows:

[0119] Step S3031, performing task planning according to the task parameters to obtain a number of subtasks;

[0120] It's important to note that functional interfaces are channels through which agents interact with other external systems, services, or internal functional modules. Each functional interface has specific functions, such as query, reservation, and notification. Connecting a target agent to several functional interfaces means that the target agent can call corresponding functions through these interfaces to complete specific operations in the task processing process. For example, in a conference room reservation scenario, the target agent connects to the conference room query functional interface, the conference room reservation functional interface, and the personnel notification functional interface.

[0121] In addition, it should be noted that the target agent can integrate the capabilities of different functional modules by connecting to multiple functional interfaces, and realize the decomposition and processing of complex tasks. The functional interface can be in the form of an application programming interface (API) and the target agent calls these interfaces by sending requests and receives the returned results. For example, the conference room query functional interface can provide conference room availability information, the conference room reservation functional interface can complete the actual reservation operation, and the personnel notification functional interface can send meeting notifications to participants. Different functional interface combinations can meet the diverse needs of different task scenarios. For example, in the conference room reservation scenario, these interfaces work together to complete the entire process from query to reservation to notification, thereby improving the functional integrity and task processing capabilities of the system.

[0122] It is understandable that after sending the task parameters to the target intelligent body, the target intelligent body will carry out task planning, and according to the task parameters, decompose the complex task into multiple relatively simple sub-tasks, decompose the complex task problem into multiple easy-to-handle parts, clarify the goals and processes of each part, and make subsequent processing more organized and efficient, avoiding processing confusion or omission of key links due to the complexity of the task.

[0123] Step S3032: Process the plurality of subtasks through the plurality of functional interfaces to obtain task processing results;

[0124] It should be noted that subtasks are specific task units derived from task planning. Each subtask has a clear goal and processing content, such as querying a conference room or reserving a conference room. Task processing results are the information obtained after each subtask is processed. These results comprehensively reflect the execution status of the entire task.

[0125] It can be understood that each subtask is assigned to the corresponding functional interface for actual operation, and the functional interface is used to complete the specific processing of the subtask to obtain the corresponding task processing results. Processing subtasks through functional interfaces can give full play to the professional functional advantages of each interface, ensure that each subtask is processed efficiently and accurately, enhance the stability and reliability of the system in complex task processing, and also facilitate the management and optimization of the processing process of each subtask.

[0126] Step S3033: Generate a processing reply result based on the task processing result.

[0127] It can be understood that converting the results of the target intelligent agent's task processing into user-oriented replies and generating processing reply results can present the complex processing process and results within the system to the user in a concise and clear manner, allowing the user to understand the task processing status in a timely manner and achieve effective information feedback. This can improve the user's right to know and satisfaction, enable the question-and-answer system to form a complete task processing closed loop, and enhance the system's user experience and practicality in complex task processing.

[0128] Specifically, the generated response can be based on pre-set response templates and rules, or through natural language generation technology. For example, key information from the task processing results can be incorporated into a response template to generate a response; or natural language generation technology can be used to integrate the results of multiple subtasks into a coherent and natural text response. Different generation methods can be selected and optimized based on specific task scenarios and user needs to ensure accurate, clear, and user-friendly responses.

[0129] Optionally, the task scenario type is a conference room reservation scenario, the functional interface includes one or more of a query interface, a reservation interface and a notification interface, the task parameters include task requirement parameters and task entity parameters, the task requirement parameters include one or more of task setting, task query or task modification, the task entity parameters include one or more of participants, conference room capacity, meeting type, meeting time and equipment configuration, and the meeting type includes online meetings and offline meetings.

[0130] It should be noted that the conference room reservation scenario is a specific type of task scenario that involves the processing of conference room-related business. The query interface is used to query the availability information of the conference room, such as whether the conference room is free and the equipment configuration of the conference room. The reservation interface is used to complete the actual reservation operation of the conference room. The notification interface is used to send meeting notifications and other messages to the participants. The task requirement parameters reflect the user's specific operational requirements for the task, such as whether to reserve a conference room (task setting), query the status of a conference room (task query), or modify the information of a reserved conference room (task modification). Task entity parameters are entity information related to conference room reservation, such as the list of participants, the capacity of the conference room, the type of meeting (online or offline), the specific meeting time, and the required equipment configuration.

[0131] In a feasible implementation, the task scenario type is a conference room reservation scenario, the target agent is connected to a plurality of functional interfaces, and the plurality of functional interfaces include one or more of a query interface, a reservation interface, and a notification interface. The step of processing the pending task problem by the target agent corresponding to the task scenario type according to the task parameters to obtain a processing response result includes:

[0132] The task parameters are sent to the target agent of the conference room predetermined scene for the following processing:

[0133] Step S3041: querying whether there is an available conference room through the query interface according to the task parameters;

[0134] It's important to note that the conference room reservation scenario is a specific task scenario type, involving the processing of conference room-related tasks. The query interface is used to query conference room availability information, such as availability and equipment configuration. It searches the conference room reservation system database based on task parameters (such as room capacity, meeting type, meeting time, and equipment configuration) and returns a list of meeting rooms that meet these criteria.

[0135] Understandably, the target agent uses the query interface to check whether a conference room meets the reservation requirements, such as availability at the specified time and availability of the required capacity and equipment, according to the task parameters. Using the query interface to query available conference rooms can quickly and accurately filter out conference room resources that meet user needs, avoiding blind reservations that lead to wasted resources or unmet needs. This can improve the success rate and efficiency of conference room reservations and ensure that users can obtain available conference room resources in a timely manner.

[0136] Specifically, after receiving a query request from the target agent, the query interface queries the conference room reservation system database. For example, it searches for all unbooked conference rooms within a meeting timeframe (tomorrow at 10:00 AM). It then filters for meeting rooms with a capacity of 20 people. Finally, it further filters based on equipment configuration (projectors) to obtain a list of available conference rooms that meet the timeframe, capacity, and equipment requirements. The query interface can employ a variety of query algorithms and optimization strategies, such as indexed queries and caching mechanisms, to improve query speed and accuracy, ensuring timely return of query results.

[0137] Step S3042: If the available conference room exists, the available conference room is reserved through the reservation interface to obtain a conference room reservation result;

[0138] It should be noted that the reservation interface is a functional interface for completing the conference room reservation operation, which can send a reservation request to the conference room reservation system and receive reservation result feedback.

[0139] It is understood that after querying available conference rooms, the target agent selects one (perhaps based on certain rules, such as prioritizing conference rooms with matching capacity and complete equipment), sends a reservation request through the reservation interface, completes the conference room reservation operation, and obtains a result indicating whether the reservation was successful or failed due to a conflict. Reserving conference rooms through the reservation interface allows the queried available conference room resources to be actually allocated to the user, ensuring that the user's needs are met. This allows for the effective management and allocation of conference room resources and avoids resource waste. It also provides timely feedback on the reservation results to the user, allowing them to understand the reservation status.

[0140] Specifically, after receiving a reservation request from the target agent, the reservation interface sends the request to the conference room reservation system's backend service. The backend service updates the conference room's reservation status, marking it as reserved, and returns the reservation result to the reservation interface. The reservation interface then feeds the result back to the target agent. For example, after the reservation interface sends a request to reserve conference room A, the backend service checks whether conference room A is still available at the specified time. If so, the reservation is completed and a successful reservation result is returned. If a conflict arises during the reservation process (e.g., conference room A has already been reserved by someone else), a failed reservation result is returned.

[0141] Step S3043: If the available conference room is successfully reserved, a meeting notification is sent to the participants set in the task parameters through the notification interface according to the conference room reservation result to obtain a meeting notification result;

[0142] It should be noted that the notification interface is a functional interface for sending meeting notifications to participants. It can interact with messaging services such as email systems and SMS platforms to send meeting-related information (such as time, location, etc.) to participants.

[0143] It's understandable that after a conference room is successfully reserved, the target agent sends a meeting notification to the attendees listed in the task parameters through the notification interface and obtains the notification result, such as whether the notification was sent successfully or failed due to incorrect attendee contact information. Sending meeting notifications through the notification interface can promptly inform attendees of the conference room reservation, ensuring that they can attend the meeting on time, improving meeting organization efficiency, and avoiding problems such as absenteeism caused by attendees being unaware of the meeting arrangements.

[0144] Specifically, after receiving the notification request sent by the target agent, the notification interface will select the appropriate notification method based on the contact information of the participants (such as email, mobile phone number, etc.). For example, for participants with email addresses, the meeting notification is sent through the email system; for participants with mobile phone numbers, the notification is sent through the SMS platform. The content of the notification includes information such as the meeting time, location, and conference room name. The notification interface will record the status of the notification sending and feedback the result to the target agent. For example, if the email is successfully sent and not returned, the notification result is successful; if the mobile phone number is incorrect and the SMS sending fails, the notification result is failed.

[0145] Step S3044: Generate a processing reply result based on the conference room reservation result and the conference notification result.

[0146] It is understandable that the target intelligent agent integrates the conference room reservation results and the meeting notification results, organizes them into a clear and accurate reply content, and feeds back to the user, informing the user of the specific circumstances of the conference room reservation and notification sending, such as the reservation is successful and the notification has been sent, the reservation is successful but some notifications failed, the reservation failed, etc. The comprehensive results of the conference room reservation and notification sending can be presented to the user in a concise and clear manner, allowing the user to understand the final status of the entire task processing in a timely manner.

[0147] Specifically, the generated response can be based on pre-set response templates or rules. For example, if the conference room reservation is successful and all participant notifications are sent successfully, the response will read, "Your conference room has been successfully reserved, and all participant notifications have been successfully sent." If the conference room reservation is successful but some notifications fail, the response will read, "Your conference room has been successfully reserved, but some participant notifications failed to be sent. Please verify the contact information of the relevant personnel and send them again." If the conference room reservation fails, the response will include the reason for the reservation failure and relevant suggestions.

[0148] For example, in a conference room reservation scenario, after receiving task parameters such as meeting time, attendees, and equipment configuration, the target agent begins task planning. For example, the task can be broken down into three subtasks: searching for meeting rooms that meet the requirements, reserving a meeting room, and notifying attendees.

[0149] The target agent then processes each subtask through the corresponding functional interface. For the conference room query subtask, the conference room query functional interface is called to send a query request to the conference room reservation system database to obtain a list of available conference rooms. For the conference room reservation subtask, the conference room reservation functional interface is called to send the reservation information to the reservation system and obtain the reservation result. For the attendee notification subtask, the attendee notification functional interface is called to send a meeting notification to the attendees via email or text message and obtain the notification delivery result.

[0150] Finally, a response is generated based on the queried room availability information, reservation results, and notification delivery results. For example, if the room reservation is successful and the notification is delivered successfully, a response such as "Your reserved room has been successfully reserved, and the notifications to the participants have been sent" is generated. If the reservation fails, a response is generated with the reason for the failure and any relevant suggestions.

[0151] In this implementation, for the conference room reservation scenario, the target intelligent agent completes the processing of conference room query, reservation and notification sending in sequence through the query interface, reservation interface and notification interface, and generates a comprehensive reply result, realizing the full process from resource query to allocation to notification, effectively solving the problem in the complex task scenario of conference room reservation, improving the user's usage efficiency in the conference room reservation scenario, and enhancing the application ability and intelligence level in specific complex task scenarios.

[0152] In this embodiment, the target intelligent agent is connected to the functional interface of multiple task processing, and the target intelligent agent plans and decomposes tasks, calls the functional interface to process subtasks and generates response results. This can give full play to the role of the functional interface, decompose and process complex tasks, improve the efficiency and accuracy of task processing, and enable the question-answering system to solve problems more systematically and efficiently in complex task scenarios.

[0153] It should be noted that the above examples are only used to understand this application and do not constitute a limitation on the question-and-answer task processing method of this application. More simple transformations based on this technical concept are all within the scope of protection of this application.

[0154] This application also provides a question-answering task processing device, please refer to Figure 3 , the question-answering task processing device includes:

[0155] Receiving module 10, for receiving the pending task problem input by the user in the session;

[0156] An extraction module 20 is used to determine the task scenario type according to the task problem to be processed, and extract the task parameters of the task problem to be processed;

[0157] The processing module 30 is used to process the pending task problem according to the task parameters through the target agent corresponding to the task scenario type to obtain a processing response result.

[0158] Optionally, the processing module 30 is further configured to:

[0159] Determining whether the task parameters meet the necessary task parameter requirements of the target agent;

[0160] If the task parameters do not meet the necessary task parameter requirements, a set of parameters to be supplemented is obtained based on the difference between the necessary task parameter requirements and the task parameters;

[0161] According to the set of parameters to be supplemented, multiple rounds of dialogue are conducted with the user through the session to obtain supplemented task parameters, so that according to the supplemented task parameters, the task problem to be processed is processed by the intelligent agent corresponding to the task scenario type to obtain a processing response result.

[0162] Optionally, the processing module 30 is further configured to:

[0163] According to the set of parameters to be supplemented, generating a number of parameter supplementation sub-problems;

[0164] Sending the plurality of parameter supplementation sub-questions to the session, and obtaining parameter supplementation answers returned by the user in the session;

[0165] The supplementary task parameters in the parameter supplement answer are extracted, and the supplementary task parameters are added to the task parameters to obtain supplemented task parameters.

[0166] Optionally, the processing module 30 is further configured to:

[0167] Based on a preset parameter dependency graph, detecting whether there is a parameter conflict among the task parameters, wherein the parameter dependency graph is constructed based on a preset parameter logical dependency relationship;

[0168] If there is a parameter conflict among the task parameters, an alternative resource is allocated to the task parameters through a preset resource pool, and a number of candidate task parameters are generated according to the associated parameters of the alternative resource;

[0169] In the conversation, the candidate task parameters are displayed to the user through visual interactive cards, so that the task problem to be processed is processed by the intelligent agent corresponding to the task scenario type according to the candidate task parameters selected by the user to obtain a processing response result.

[0170] Optionally, the target agent is connected to several functional interfaces, and the processing module 30 is further configured to:

[0171] The task parameters are sent to the target agent for processing as follows:

[0172] Perform task planning according to the task parameters to obtain several subtasks;

[0173] Processing the plurality of subtasks through the plurality of functional interfaces to obtain task processing results;

[0174] Based on the task processing result, a processing reply result is generated.

[0175] Optionally, the task scenario type is a conference room reservation scenario, and the target agent is connected to a plurality of functional interfaces, which include one or more of a query interface, a reservation interface, and a notification interface.

[0176] The step of processing the pending task problem by the target agent corresponding to the task scenario type according to the task parameters to obtain a processing response result includes:

[0177] The task parameters are sent to the target agent of the conference room predetermined scene for the following processing:

[0178] According to the task parameters, querying whether there is an available conference room through the query interface;

[0179] If the available conference room exists, booking the available conference room through the booking interface to obtain a conference room booking result;

[0180] If the available conference room is successfully reserved, a conference notification is sent to the participants set in the task parameters through the notification interface according to the conference room reservation result to obtain a conference notification result;

[0181] A processing reply result is generated based on the conference room reservation result and the conference notification result.

[0182] The question-and-answer task processing device provided in this application, which utilizes the question-and-answer task processing method described in the aforementioned embodiment, can address the technical issue of low processing efficiency in complex task scenarios associated with existing question-and-answer task processing methods. Compared to the prior art, the beneficial effects of the question-and-answer task processing device provided in this application are the same as those of the question-and-answer task processing method described in the aforementioned embodiment. The other technical features of the question-and-answer task processing device are the same as those disclosed in the aforementioned embodiment and are not further elaborated upon here.

[0183] The present application provides a question-and-answer task processing device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the question-and-answer task processing method in the above-mentioned first embodiment.

[0184] Reference below Figure 4 , which shows a schematic diagram of the structure of a question-answering task processing device suitable for implementing the embodiments of the present application. The question-answering task processing device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The question-and-answer task processing device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0185] like Figure 4As shown, the question-answering task processing device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the question-answering task processing device. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the question-answering task processing device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a question-answering task processing device with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems can be implemented or provided instead.

[0186] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a read-only memory 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are performed.

[0187] The question-and-answer task processing device provided in this application, which utilizes the question-and-answer task processing method described in the aforementioned embodiment, can address the technical issue of low processing efficiency in complex task scenarios associated with existing question-and-answer task processing methods. Compared to the prior art, the beneficial effects of the question-and-answer task processing device provided in this application are the same as those of the question-and-answer task processing method described in the aforementioned embodiment. The other technical features of this question-and-answer task processing device are the same as those disclosed in the aforementioned embodiment and are not further elaborated upon here.

[0188] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0189] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0190] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the question-and-answer task processing method in the above-mentioned embodiment.

[0191] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0192] The above-mentioned computer-readable storage medium may be included in the question-answering task processing device; or it may exist independently without being assembled into the question-answering task processing device.

[0193] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the question-and-answer task processing device, the question-and-answer task processing device: D exclusive D.

[0194] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0195] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of code, and the module, program segment or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, as well as the combination of boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0196] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0197] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., a computer program) for executing the above-mentioned question-and-answer task processing method. This computer-readable storage medium can solve the technical problem of low processing efficiency of existing question-and-answer task processing methods in complex task scenarios. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the question-and-answer task processing method provided in the above-mentioned embodiment, and will not be elaborated here.

[0198] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned question-answering task processing method when executed by a processor.

[0199] The computer program product provided in this application can address the technical issue of low processing efficiency in complex task scenarios caused by existing question-and-answer task processing methods. Compared to the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the question-and-answer task processing methods provided in the aforementioned embodiments, and are not further elaborated here.

[0200] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A question-answering task processing method, characterized in that: The question-answering task processing method is applied to a question-answering task processing platform, which is configured with several intelligent agents, including: Receive pending task questions entered by the user in the session; Determine the task scenario type according to the task problem to be processed, and extract the task parameters of the task problem to be processed; According to the task parameters, the target intelligent agent corresponding to the task scenario type processes the task problem to be processed to obtain a processing response result.

2. The question-answering task processing method according to claim 1, wherein: Before the step of processing the pending task problem by the agent corresponding to the task scenario type according to the task parameters and obtaining a processing response result, the method further includes: Determining whether the task parameters meet the necessary task parameter requirements of the target agent; If the task parameters do not meet the necessary task parameter requirements, a set of parameters to be supplemented is obtained based on the difference between the necessary task parameter requirements and the task parameters; According to the set of parameters to be supplemented, multiple rounds of dialogue are conducted with the user through the session to obtain supplemented task parameters, so that according to the supplemented task parameters, the task problem to be processed is processed by the intelligent agent corresponding to the task scenario type to obtain a processing response result.

3. The question-answering task processing method according to claim 2, wherein: The step of conducting multiple rounds of dialogue with the user through the conversation according to the set of parameters to be supplemented to obtain supplemented task parameters includes: According to the set of parameters to be supplemented, generating a number of parameter supplementation sub-problems; Sending the plurality of parameter supplementation sub-questions to the session, and obtaining parameter supplementation answers returned by the user in the session; The supplementary task parameters in the parameter supplement answer are extracted, and the supplementary task parameters are added to the task parameters to obtain supplemented task parameters.

4. The question-answering task processing method according to claim 1, wherein: Before the step of processing the pending task problem by the agent corresponding to the task scenario type according to the task parameters and obtaining a processing response result, the method further includes: Based on a preset parameter dependency graph, detecting whether there is a parameter conflict among the task parameters, wherein the parameter dependency graph is constructed based on a preset parameter logical dependency relationship; If there is a parameter conflict among the task parameters, an alternative resource is allocated to the task parameters through a preset resource pool, and a number of candidate task parameters are generated according to the associated parameters of the alternative resource; In the conversation, the candidate task parameters are displayed to the user through visual interactive cards, so that the task problem to be processed is processed by the intelligent agent corresponding to the task scenario type according to the candidate task parameters selected by the user to obtain a processing response result.

5. The question-answering task processing method according to claim 1, wherein: The target agent is connected to a plurality of functional interfaces, and the step of processing the pending task problem by the target agent corresponding to the task scenario type according to the task parameters to obtain a processing response result includes: The task parameters are sent to the target agent for processing as follows: Perform task planning according to the task parameters to obtain several subtasks; Processing the plurality of subtasks through the plurality of functional interfaces to obtain task processing results; Based on the task processing result, a processing reply result is generated.

6. The question-answering task processing method according to claim 1, wherein: The task scenario type is a conference room reservation scenario, the target agent is connected to a number of functional interfaces, and the several functional interfaces include one or more of a query interface, a reservation interface, and a notification interface. The step of processing the pending task problem by the target agent corresponding to the task scenario type according to the task parameters to obtain a processing response result includes: The task parameters are sent to the target agent of the conference room predetermined scene for the following processing: According to the task parameters, querying whether there is an available conference room through the query interface; If the available conference room exists, booking the available conference room through the booking interface to obtain a conference room booking result; If the available conference room is successfully reserved, a conference notification is sent to the participants set in the task parameters through the notification interface according to the conference room reservation result to obtain a conference notification result; A processing reply result is generated based on the conference room reservation result and the conference notification result.

7. A question-answering task processing device, characterized in that: The question-answering task processing device includes: The receiving module is used to receive the pending task questions input by the user in the session; An extraction module is used to determine the task scenario type according to the task problem to be processed, and extract the task parameters of the task problem to be processed; The processing module is used to process the pending task problem according to the task parameters through the target intelligent agent corresponding to the task scenario type to obtain a processing response result.

8. A question-answering task processing device, characterized in that: The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the question-answering task processing method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the question-answering task processing method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the question-answering task processing method according to any one of claims 1 to 6 are implemented.

Images