Question answering method, device, equipment, system and program product

Abstract

The embodiment of the invention provides a question answering method. The method comprises the steps of obtaining a to-be-processed question initiated for an agent group; then complexity judgment is carried out on the to-be-processed problem to obtain complexity information; if the complexity information indicates that the to-be-processed problem is a complex problem, disassembling the to-be-processed problem to obtain a plurality of sub-problems; extracting question features corresponding to the sub-questions to determine target agents adaptive to the sub-questions in the agent group according to the question features; and executing the sub-questions based on the target agent to obtain answer information corresponding to the to-be-processed question. Therefore, a question answering process based on the intelligent agent group is realized, and by decomposing the complex question and adapting the targeted intelligent agent in the intelligent agent group, the professionality of understanding the question by the intelligent agent is ensured, and the accuracy of question answering is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to question answering methods, devices, equipment, systems and program products. Background Art

[0002] With the rapid development of artificial intelligence technology, AI agents have shown broad application potential in many fields, especially in handling complex tasks.

[0003] Generally, the process of intelligent agents handling complex tasks mainly relies on configuring predefined rules or machine learning models to guide their behavior, that is, processing complex tasks through the optimization of model algorithms.

[0004] However, the optimization process of the model algorithm is cumbersome, and due to the limitations of the intelligent agent in various fields, it is impossible to give professional answers immediately, which affects the accuracy of the answers to the questions. Summary of the invention

[0005] In view of this, the present application provides a question answering method to improve the accuracy of question answering.

[0006] According to a first aspect of an embodiment of the present application, a method for answering a question is provided, which can be applied to a system or program including a question answering function in a terminal device, and specifically includes:

[0007] Get pending issues for the group of agents;

[0008] Performing complexity determination on the problem to be processed to obtain complexity information;

[0009] If the complexity information indicates that the problem to be processed is a complex problem, decomposing the problem to be processed to obtain a plurality of sub-problems;

[0010] Extracting problem features corresponding to the sub-problems, so as to determine a target agent in the agent group that is adapted to the sub-problems according to the problem features;

[0011] The sub-problem is processed based on the target agent to obtain answer information corresponding to the problem to be processed.

[0012] Optionally, in some possible implementations, the performing complexity determination on the problem to be processed to obtain complexity information includes:

[0013] Rewriting the problem to be processed to obtain a rewritten problem;

[0014] Performing complexity determination on the rewriting problem based on multiple preset dimensions to obtain complexity scores corresponding to each preset dimension;

[0015] A weighted calculation is performed on the complexity scores corresponding to the respective preset dimensions to obtain the complexity information.

[0016] Optionally, in some possible implementations, performing complexity determination on the rewriting problem based on multiple preset dimensions to obtain complexity scores corresponding to each preset dimension includes:

[0017] Determining the interactive information corresponding to the rewriting question based on the preset dimension;

[0018] determining problem areas indicated by the interaction information;

[0019] determining correlation parameters between the problem areas;

[0020] Complexity determination is performed according to the interaction information and the relevance parameter respectively to obtain complexity scores corresponding to the interaction information and the relevance parameter respectively.

[0021] Optionally, in some possible implementations, if the complexity information indicates that the problem to be processed is a complex problem, the problem to be processed is decomposed to obtain multiple sub-problems, including:

[0022] If the complexity information indicates that the problem to be processed is a complex problem, decomposing the problem to be processed based on preset dimensions to obtain a decomposed problem;

[0023] Determine the correlation between the disassembly problems to obtain serial and parallel information between the disassembly problems;

[0024] The disassembly problem is configured based on the serial-parallel information to obtain a plurality of the sub-problems.

[0025] Optionally, in some possible implementations, extracting the problem feature corresponding to the sub-problem to determine the target agent in the agent group that is adapted to the sub-problem according to the problem feature includes:

[0026] Performing intent recognition on the sub-questions to obtain intent information;

[0027] Determine the task information and the scenario information indicated by the intention information, so as to determine the problem feature corresponding to the sub-problem based on the task information and the scenario information;

[0028] A target intelligent agent that is adapted to the sub-problem is determined from the intelligent agent group according to the problem characteristics.

[0029] Optionally, in some possible implementations, the method further includes:

[0030] Obtaining an assistance tool corresponding to the task information;

[0031] Calling the extended tool configured in the agent group;

[0032] The target agent is enhanced and configured through the assisting tool and the extending tool.

[0033] Optionally, in some possible implementations, the method further includes:

[0034] If there is no target agent suitable for the sub-problem in the agent group, the sub-problem will be distributed to the underlying agent for processing, and the question information corresponding to the sub-problem will be recorded. The underlying agent is used to provide a general answer function, and the question information is used to guide the agent registration process in the agent group.

[0035] Optionally, in some possible implementations, processing the sub-problem based on the target agent to obtain answer information corresponding to the problem to be processed includes:

[0036] Determine the key elements corresponding to the sub-questions, and make integrity judgment on the key elements;

[0037] If the key elements are complete, the sub-problems are processed based on the target agent to obtain processing results;

[0038] The processing result is rendered according to the answer template to obtain the answer information.

[0039] Optionally, in some possible implementations, the method further includes:

[0040] If the key element is incomplete, determining the missing element corresponding to the sub-question according to the key element;

[0041] generating follow-up information according to the missing element;

[0042] Acquire feedback information corresponding to the follow-up information, so as to trigger the target agent to process the sub-question based on the feedback information.

[0043] Optionally, in some possible implementations, if the key element is complete, processing the sub-problem based on the target agent to obtain a processing result includes:

[0044] If the key elements are complete, the complexity of the sub-problems is determined to obtain verification information;

[0045] Comparing the verification information with the complexity information to obtain a convergence parameter;

[0046] If the convergence parameter satisfies the answer condition, the sub-problem is processed based on the target agent to obtain the processing result.

[0047] According to a second aspect of an embodiment of the present application, there is provided a question answering device, comprising:

[0048] An acquisition unit, used to acquire pending problems initiated by the group of intelligent agents;

[0049] A processing unit, used for determining the complexity of the problem to be processed to obtain complexity information;

[0050] The processing unit is further configured to decompose the problem to be processed to obtain a plurality of sub-problems if the complexity information indicates that the problem to be processed is a complex problem;

[0051] The processing unit is further used to extract the problem features corresponding to the sub-problem, so as to determine the target agent in the agent group that is adapted to the sub-problem according to the problem features;

[0052] An answering unit is used to process the sub-problem based on the target agent to obtain answer information corresponding to the problem to be processed.

[0053] Optionally, in some possible implementations, the processing unit is used to rewrite the problem to be processed to obtain a rewritten problem when performing a complexity determination on the problem to be processed to obtain complexity information; perform a complexity determination on the rewritten problem based on multiple preset dimensions to obtain a complexity score corresponding to each preset dimension; and perform a weighted calculation on the complexity score corresponding to each preset dimension to obtain the complexity information.

[0054] Optionally, in some possible implementations, the processing unit is used to determine the interaction information corresponding to the rewriting problem based on the preset dimensions when performing complexity judgment on the rewriting problem based on multiple preset dimensions to obtain complexity scores corresponding to each preset dimension; determine the problem domain indicated by the interaction information; determine the correlation parameters between the problem domains; and perform complexity judgments according to the interaction information and the correlation parameters respectively to obtain complexity scores corresponding to the interaction information and the correlation parameters respectively.

[0055] Optionally, in some possible implementations, the processing unit is used to decompose the problem to be processed based on preset dimensions to obtain decomposed problems if the complexity information indicates that the problem to be processed is a complex problem; determine the correlation between the decomposed problems to obtain serial and parallel information between the decomposed problems; and configure the decomposed problems based on the serial and parallel information to obtain multiple sub-problems.

[0056] Optionally, in some possible implementations, the processing unit is used to perform intent recognition on the sub-problem to obtain intention information when extracting the problem features corresponding to the sub-problem to determine the target agent in the agent group that is adapted to the sub-problem based on the problem features; determine the task information and scene information indicated by the intention information to determine the problem features corresponding to the sub-problem based on the task information and the scene information; and determine the target agent that is adapted to the sub-problem from the agent group based on the problem features.

[0057] Optionally, in some possible implementations, the processing unit is used to obtain an assistive tool corresponding to the task information; call an extended tool configured in the agent group; and enhance the configuration of the target agent through the assistive tool and the extended tool.

[0058] Optionally, in some possible implementations, the processing unit is used to distribute the sub-problem to a backup agent for processing if there is no target agent suitable for the sub-problem in the agent group, and record the question information corresponding to the sub-problem. The backup agent is used to provide a general answer function, and the question information is used to guide the agent registration process in the agent group.

[0059] Optionally, in some possible implementations, the answer unit is used to determine the key elements corresponding to the sub-problem when processing the sub-problem based on the target intelligent agent to obtain answer information corresponding to the problem to be processed, and to judge the integrity of the key elements; if the key elements are complete, processing the sub-problem based on the target intelligent agent to obtain a processing result; rendering the processing result according to the answer template to obtain the answer information.

[0060] Optionally, in some possible implementations, the answering unit is used to determine the missing element corresponding to the sub-question based on the key element if the key element is incomplete; generate follow-up information based on the missing element; and obtain feedback information corresponding to the follow-up information to trigger the target intelligent agent to process the sub-question based on the feedback information.

[0061] Optionally, in some possible implementations, the answering unit is used to, when processing the sub-problem based on the target intelligent agent to obtain a processing result if the key element is complete, perform complexity judgment on the sub-problem to obtain verification information; compare the verification information with the complexity information to obtain a convergence parameter; and if the convergence parameter meets the answering condition, process the sub-problem based on the target intelligent agent to obtain the processing result.

[0062] According to a third aspect of an embodiment of the present application, there is provided a question answering device, including an input-output component and a processor;

[0063] The input and output components are used to obtain problems to be processed;

[0064] The processor is used to answer the pending question obtained by the input-output component by executing the question answering method as described in the first aspect or any implementation of the first aspect.

[0065] According to a fourth aspect of an embodiment of the present application, there is provided a question answering system, including an interactive client and a server;

[0066] The interactive client is used to obtain the pending questions and send the pending questions to the server, and to display the answer results of the questions output by the server;

[0067] The server is used to answer the pending questions obtained by the interactive client by executing the question answering method described in the first aspect or any implementation manner of the first aspect.

[0068] According to a fifth aspect of an embodiment of the present application, a computer program product is provided, including: a computer program, which, when executed by a processor, implements the question answering method described in the first aspect or any implementation manner of the first aspect.

[0069] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0070] By obtaining the pending problem initiated by the agent group, the complexity of the pending problem is determined to obtain complexity information. If the complexity information indicates that the pending problem is a complex problem, the pending problem is decomposed to obtain multiple sub-problems, and the problem features corresponding to the sub-problems are extracted to determine the target agent in the agent group that is adapted to the sub-problem according to the problem features. Then, the sub-problem is executed based on the target agent to obtain the answer information corresponding to the pending problem. Thus, the question answering process based on the agent group is realized. By decomposing complex problems and adapting targeted agents in the agent group, the professionalism of the agent in understanding the problem is guaranteed, and the accuracy of the question answering is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0072] Figure 1 A diagram of the network architecture that runs the question answering system;

[0073] Figure 2 A process architecture diagram of answering questions provided in an embodiment of the present application;

[0074] Figure 3 A flowchart of a question answering method provided in an embodiment of the present application;

[0075] Figure 4 A scenario flow chart of a question answering method provided in an embodiment of the present application;

[0076] Figure 5 A scenario flow chart of a problem solving process provided in an embodiment of the present application;

[0077] Figure 6 A scenario flow chart of an agent scheduling process provided in an embodiment of the present application;

[0078] Figure 7 A scenario flow chart of an agent execution process provided in an embodiment of the present application;

[0079] Figure 8 A schematic diagram of the structure of a question answering device provided in an embodiment of the present application;

[0080] Fig. 9 A schematic diagram of the structure of a terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0081] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0082] It should be understood that the question answering method provided in the present application can be applied to a system or program including a question answering function in a terminal device, such as a question answering application. Specifically, the question answering system can be run on a system such as Figure 1 In the network architecture shown in Figure 1 As shown in FIG. 1 , it is a network architecture diagram of the question answering system. As can be seen from the figure, the question answering system can provide a question answering process with multiple information sources, that is, the question to be processed is determined through the acquisition operation on the terminal side, and then sent to the server to trigger the target agent in the agent group to answer the question to be processed; it can be understood that Figure 1 A variety of terminal devices are shown in FIG. 1 . The terminal devices may be computer devices. In actual scenarios, more or fewer types of terminal devices may participate in the process of answering questions. The specific number and type depend on the actual scenario and are not limited here. In addition, Figure 1 One server is shown in the figure, but in actual scenarios, multiple servers may be involved, especially in multi-disciplinary answer scenarios, and the specific number of servers depends on the actual scenario.

[0083] In this embodiment, the server is a server directly or indirectly associated with the group of intelligent entities. It can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be a smart phone, tablet computer, laptop computer, desktop computer, smart speaker, smart watch, etc., but is not limited to this. The terminal and the server can be directly or indirectly connected by wired or wireless communication, and the terminal and the server can be connected to form a blockchain network, which is not limited in this application.

[0084] It can be understood that the above-mentioned question answering system can run on a personal mobile terminal, for example: as an application such as a smart assistant, it can also run on a server, and it can also run on a third-party device to provide answers to questions to obtain the question answering processing results of the information source; the specific question answering system can be run in the above-mentioned device in the form of a program, and it can also be run as a system component in the above-mentioned device, and it can also be used as a cloud service program. The specific operation mode depends on the actual scenario and is not limited here.

[0085] With the rapid development of artificial intelligence technology, AI agents have shown broad application potential in many fields, especially in handling complex tasks.

[0086] Generally, the process of intelligent agents handling complex tasks mainly relies on configuring predefined rules or machine learning models to guide their behavior, that is, processing complex tasks through the optimization of model algorithms.

[0087] However, the optimization process of the model algorithm is cumbersome, and due to the limitations of the intelligent agent in various fields, it is impossible to give professional answers immediately, which affects the accuracy of the answers to the questions.

[0088] In order to solve the above problems, this application proposes a question answering method, which is applied to Figure 2 In the process framework of question answering shown in Figure 2 As shown, a process architecture diagram of question answering provided in an embodiment of the present application is provided. Through interactive operations on the terminal side, the questions to be processed are determined, thereby triggering the server to execute an efficient agent intelligent recommendation distribution process from the associated agent group. This process determines the target agent by making a complexity judgment after rewriting the question, so that users can easily select the agent service that best suits their needs from a large number of agent options.

[0089] It can be understood that the question answering method provided by the present application can be written as a program to serve as a processing logic in a hardware system, or as a question answering device to implement the above processing logic in an integrated or external manner. As an implementation method, the question answering device obtains the problem to be processed initiated by the intelligent agent group; then determines the complexity of the problem to be processed to obtain complexity information; if the complexity information indicates that the problem to be processed is a complex problem, the problem to be processed is disassembled to obtain multiple sub-problems; and the problem features corresponding to the sub-problems are extracted to determine the target intelligent agent in the intelligent agent group that is adapted to the sub-problem according to the problem features; and then executes the sub-problem based on the target intelligent agent to obtain the answer information corresponding to the problem to be processed. Thereby, the question answering process based on the intelligent agent group is realized, and by decomposing complex problems and adapting targeted intelligent agents in the intelligent agent group, the professionalism of the intelligent agent in understanding the problem is guaranteed, and the accuracy of answering the question is improved.

[0090] In combination with the above process architecture, the following will introduce the method of answering questions in this application. Please refer to Figure 3 , Figure 3 A flowchart of a question answering method provided in an embodiment of the present application, the embodiment of the present application at least includes the following steps:

[0091] 301. Obtain pending issues initiated for the intelligent agent group.

[0092] In this embodiment, the agent group is composed of multiple specialized agents, each of which is responsible for executing a specific subtask or functional module. Agents collaborate and share information and resources through efficient communication protocols.

[0093] In addition, the pending questions are the search questions initiated by the user, the questions generated by the task execution, or the processing processes corresponding to other interactive operations. Among them, the pending questions can be a collection of one or more questions, and the user can adjust the questions at any time during the answering process, that is, flexibly respond to dynamically changing task requirements and environmental conditions, and enhance the adaptability and robustness of the system. Therefore, this embodiment uses the intelligent agent group to answer the pending questions, which can effectively coordinate and manage multiple intelligent agents to improve the exposure rate and task execution efficiency of the intelligent agents and reduce the cost of communication between intelligent agents. Therefore, in a massive Agents system, the management process in the intelligent agent group allows users to obtain suitable Agents without feeling, and effectively meet user needs.

[0094] Therefore, this embodiment can be applied to the execution logic of large language models, speech synthesis algorithms, and speech recognition algorithms, and the specific usage scenario depends on actual needs.

[0095] Specifically, in order to realize the management process in the above-mentioned intelligent group, Figure 4 As shown, Figure 4 A scenario flow chart of a question answering method provided in an embodiment of the present application; the figure shows the management process in the intelligent agent group, that is, for the acquired pending questions, the problem pre-processing is first performed, and then the scheduling module is triggered to schedule the intelligent agent, which adopts a central controller using a dynamic scheduling algorithm to adjust the task allocation and priority of the intelligent agent according to real-time feedback to ensure that the system can maintain efficient operation under different loads.

[0096] Specifically, the scheduling process involves decomposing complex problems. After decomposing simple problems, intelligent agent adaptation is performed under the multi-level classification system to perform executor logic and obtain corresponding response information.

[0097] Optionally, during the multi-level classification process, a fault-tolerance and recovery strategy is also configured, that is, when a single agent fails, the system can automatically reallocate tasks to ensure the continuity and reliability of the overall task.

[0098] Optionally, during the execution process, an adaptive learning mechanism is also configured, that is, the system has self-learning capabilities and can continuously optimize the decision-making process and the behavior patterns of the intelligent agent based on historical data and real-time feedback.

[0099] It can be seen that this embodiment is a design for problem distribution for a group of intelligent agents. Compared with the general personalized recommendation process, the recommendation system often only considers information such as the user's historical behavior and preferences when making recommendations, but ignores the user's current contextual information and real needs. In the context of solving actual needs, the recommendation effect of personalized recommendations will also be limited. In this embodiment, the intelligent scheduling process within the intelligent agent can perform fine-grained problem understanding and targeted intelligent agent matching by decomposing the problem.

[0100] Compared with the general subscription system, the intelligent agent matching process has the problem that the subscription scope is limited to the scope perceived by the subscriber. The subscription system cannot function in areas that the subscriber cannot reach. For example, student A wants to inquire about the contract for renting a house, but does not understand the relevant legal knowledge. Then A cannot know that he should filter and subscribe to the sub-fields under the legal field. The intelligent agent matching process of this embodiment automatically matches by decomposing the problem, without the need for the user to make a selection operation.

[0101] Next, the modules and algorithm configurations in the above scenario flow are explained in conjunction with the specific implementation process.

[0102] 302. Perform complexity determination on the problem to be processed to obtain complexity information.

[0103] In this embodiment, the complexity determination process is as follows: Figure 4The process of pre-processing the problem is used to judge the complexity of the problem so as to carry out subsequent intelligent agent scheduling in a targeted manner.

[0104] Specifically, for the process of determining complexity information, the problem to be processed can be first rewritten to obtain a rewritten problem; then the complexity of the rewritten problem can be determined based on multiple preset dimensions to obtain the complexity score corresponding to each preset dimension; and the complexity score corresponding to each preset dimension can be weighted calculated to obtain complexity information.

[0105] Among them, the process of rewriting the user's question is to rewrite the problem to be processed into a multi-round dialogue or thinking chain for problem processing. For example, when the user asks what lunch recommendations are there for the meeting I am going to attend, the question is rewritten into the content of relevant intelligent agents such as address query and lunch recommendation, so that relevant content can be recommended to the user based on the venue of the meeting and the restaurants at the location.

[0106] Furthermore, the setting of multiple preset dimensions in the process of determining the complexity score can be determined by the interactive information required by the problem within the agent, and the interactive information can be determined by the agents involved in the multiple rounds of dialogue after the rewriting. That is, firstly, the interactive information corresponding to the rewritten problem is determined based on the preset dimensions; then the problem domain indicated by the interactive information is determined; and the correlation parameters between the problem domains are determined; and then the complexity is determined according to the interactive information and the correlation parameters respectively, so as to obtain the complexity scores corresponding to the interactive information and the correlation parameters respectively.

[0107] Therefore, in the above process of judging whether a problem is a complex problem based on its complexity, the determination of a complex problem will be made based on a comprehensive judgment based on the weighted scores of multiple dimensions, and a problem will be determined to be a complex problem if the score exceeds the threshold; the multidimensional factors include the following: number of steps (interaction information), interaction frequency (interaction information), domain span (correlation parameter) and other related factors for complexity scoring.

[0108] 303. If the complexity information indicates that the problem to be processed is a complex problem, the problem to be processed is decomposed to obtain multiple sub-problems.

[0109] In this embodiment, the process of decomposing a complex problem is a scheduling process adapted to the scheduling module. The scheduling module will collaborate according to an efficient communication protocol based on the decomposed sub-problems to share information and resources.

[0110] Specifically, in the process of decomposing a complex problem, the problem to be processed can be first decomposed based on preset dimensions to obtain decomposed problems; then the correlation between the decomposed problems can be determined to obtain serial and parallel information between the decomposed problems; and the decomposed problems can be configured based on the serial and parallel information to obtain multiple sub-problems.

[0111] In a possible scenario, the disassembly process of the above complex problem is as follows: Figure 5 As shown, Figure 5 A scenario flow chart of a problem decomposition process provided in an embodiment of the present application; the figure shows that for complex problems, componentization (reAct) will be considered, and the problem will be decomposed according to the above factors, that is, interaction information and relevance parameters, such as the number of steps, interaction frequency, domain span, etc.; after the decomposition, the serial or parallel situation of sub-agents will be considered; so that complex problems will eventually be decomposed into simple problems; and multiple complex problems will be connected in series and processed in sequence to ensure the final response effect.

[0112] It can be understood that the sub-problem configuration of this embodiment realizes the combined processing process of intelligent agents, that is, solving problems through the combination of agents. For example, when a user asks me what lunch recommendations I want to attend a meeting, the agents should call the content of relevant intelligent agents such as address query and lunch recommendation, and recommend relevant content to the user based on the venue of the meeting and the restaurants at the location, thereby improving the accuracy of the answer to the question.

[0113] 304. Extract the problem features corresponding to the sub-problems, and determine the target intelligent agent in the intelligent agent group that is suitable for the sub-problems according to the problem features.

[0114] In this embodiment, the problem feature corresponding to the sub-question is the intention understanding process of the problem description, and its feature dimensions may include a literal understanding dimension and a scene understanding dimension.

[0115] Therefore, for the process of determining the target intelligent agent, we can first identify the intent of the sub-problem to obtain the intent information; then determine the task information and scene information indicated by the intent information, so as to determine the problem features corresponding to the sub-problem based on the task information and scene information; and determine the target intelligent agent that is adapted to the sub-problem from the intelligent agent group based on the problem features.

[0116] It can be seen that through the above configuration process, the multi-level classification process will automatically dispatch to the most suitable Agents for execution according to the user's problem; among them, the task information can include the information of the user's needs, abstractly sorting out the user's most common usage needs; and the scenario information can include the type of sub-scenario field, or the usage scenario and nature of the task, etc., thereby improving the accuracy of sub-problem classification.

[0117] In a possible scenario, the configuration of the target agent can also be enhanced with related tools, such as Figure 6 As shown, Figure 6A scenario flow chart of an agent scheduling process provided in an embodiment of the present application; the figure shows Figure 4 The execution process of multi-level classification in the figure is that for the intelligent agent after successful classification, it can obtain the auxiliary tools corresponding to the task information through the concrete task segmentation; and call the extended tools configured in the intelligent agent group, such as encyclopedia tools (One Hundred Thousand Whys, etc.); then enhance the target intelligent agent through the auxiliary tools and extended tools.

[0118] In addition, if Figure 6 As shown in , in the multi-level classification process, a fault-tolerance and recovery strategy is also configured, that is, if there is no target agent suitable for the subproblem in the agent group, the subproblem will be distributed to the backup agent for processing, and the problem information corresponding to the subproblem will be recorded.

[0119] Among them, the bottom-up intelligent agent is used to provide general answering functions, such as a general large language model, etc.; and the question information is used to guide the registration process of the agents in the agent group, that is, this embodiment also configures an agent pool for the agent group, in which the agents to be registered into the agent group are configured. The question information can trigger the agent pool to perform function matching, thereby calling the corresponding agent to register to provide corresponding function support.

[0120] It is understandable that in addition to solving problems through the agent pool, other agents in the agent group that can execute sub-problems can also be called. That is, when a single agent fails, the system can automatically reallocate tasks to ensure the continuity and reliability of the overall task.

[0121] 305. Process the sub-problems based on the target intelligent agent to obtain answer information corresponding to the problem to be processed.

[0122] In this embodiment, the process of the target agent processing sub-problems, that is, the process of multiple target agents processing their own sub-problems respectively, can be configured in a corresponding order according to the serial and parallel situations of the problems to improve the processing efficiency.

[0123] In addition, during the execution process, this embodiment is also configured with an adaptive learning mechanism, that is, the system has self-learning capabilities and can continuously optimize the decision-making process and the behavior pattern of the intelligent agent based on historical data and real-time feedback.

[0124] Therefore, for Figure 4 The configuration of the actuator shown can be found in Figure 7 , Figure 7A scenario flow chart of an agent execution process provided for an embodiment of the present application; the figure shows the answer to a sub-question, which will first determine the key elements corresponding to the sub-question and judge the integrity of the key elements; if the key elements are complete, the sub-question is processed based on the target agent to obtain the processing result; and the processing result is rendered according to the answer template to obtain the answer information, and the answer information can also be search enhanced based on selective search technology and retrieval enhancement technology to enrich the answer content.

[0125] In addition, in the case where the key elements are incomplete, the missing elements corresponding to the sub-problem are determined based on the key elements; then follow-up information is generated based on the missing elements; and feedback information corresponding to the follow-up information is obtained to trigger the target intelligent agent to process the sub-problem based on the feedback information.

[0126] It can be seen that for the execution process of Agents, preprocessing will be performed. If relevant elements are missing, a reply will be constructed based on the missing elements and returned to the customer for the user to complete. This process will be repeated according to the above logic until all the elements of the question are complete. Finally, post-processing will be performed to output the template to the user side. The template matching process will also judge the quality of the answer, and the answer content will be configured in a targeted manner to match the user's question and answer habits, thereby improving the quality of the answer.

[0127] Optionally, when the key elements are complete, a secondary judgment of the complexity can be performed, that is, the complexity of the sub-problem is judged to obtain verification information; and the verification information is compared with the complexity information to obtain the convergence parameter; if the convergence parameter meets the answer condition, the sub-problem is processed based on the target agent to obtain the processing result. Among them, the process of secondary judgment of complexity (verification information) refers to the description of step 302 above, and the convergence parameter is the comparison of the complexity score of the verification information and the complexity information, that is, whether the complexity score corresponding to the verification information is reduced. If it is reduced or reaches a certain value, the answer condition is met. The specific value setting depends on the actual needs to avoid the omission of items in the scheduling process from affecting the answer process, thereby improving the accuracy of the question answer.

[0128] In combination with the above embodiments, it can be known that by obtaining the pending problem initiated for the agent group; then determining the complexity of the pending problem to obtain complexity information; if the complexity information indicates that the pending problem is a complex problem, the pending problem is decomposed to obtain multiple sub-problems; and the problem features corresponding to the sub-problems are extracted to determine the target agent in the agent group that is adapted to the sub-problem according to the problem features; and then executing the sub-problem based on the target agent to obtain the answer information corresponding to the pending problem. Thus, the question answering process based on the agent group is realized, and by decomposing the complex problem and adapting the targeted agent in the agent group, the professionalism of the agent in understanding the problem is guaranteed, and the accuracy of the question answering is improved.

[0129] In order to better implement the above solution of the embodiment of the present application, the following also provides related devices for implementing the above solution. Figure 8 , Figure 8 This is a schematic diagram of the structure of a question answering device provided in an embodiment of the present application. The answering device 800 includes:

[0130] An acquisition unit 801 is used to acquire a pending problem initiated by an agent group;

[0131] The processing unit 802 is used to determine the complexity of the problem to be processed to obtain complexity information;

[0132] The processing unit 802 is further configured to decompose the problem to be processed to obtain a plurality of sub-problems if the complexity information indicates that the problem to be processed is a complex problem;

[0133] The processing unit 802 is further configured to extract a problem feature corresponding to the sub-problem, so as to determine a target agent in the agent group that is adapted to the sub-problem according to the problem feature;

[0134] The answering unit 803 is used to process the sub-problem based on the target agent to obtain answer information corresponding to the problem to be processed.

[0135] Optionally, in some possible implementations, the processing unit 802 is used to rewrite the problem to be processed to obtain a rewritten problem when performing a complexity determination on the problem to be processed to obtain complexity information; perform a complexity determination on the rewritten problem based on multiple preset dimensions to obtain a complexity score corresponding to each preset dimension; and perform a weighted calculation on the complexity score corresponding to each preset dimension to obtain the complexity information.

[0136] Optionally, in some possible implementations, the processing unit 802 is used to determine the interaction information corresponding to the rewriting problem based on the preset dimensions when performing complexity judgment on the rewriting problem based on multiple preset dimensions to obtain complexity scores corresponding to each preset dimension; determine the problem domain indicated by the interaction information; determine the correlation parameters between the problem domains; and perform complexity judgments according to the interaction information and the correlation parameters respectively to obtain complexity scores corresponding to the interaction information and the correlation parameters respectively.

[0137] Optionally, in some possible implementations, the processing unit 802 is used to decompose the problem to be processed based on preset dimensions to obtain decomposed problems if the complexity information indicates that the problem to be processed is a complex problem, to obtain multiple sub-problems; determine the correlation between the decomposed problems to obtain serial and parallel information between the decomposed problems; and configure the decomposed problems based on the serial and parallel information to obtain multiple sub-problems.

[0138] Optionally, in some possible implementations, the processing unit 802 is used to perform intent recognition on the sub-problem when extracting the problem features corresponding to the sub-problem to determine the target agent in the agent group that is adapted to the sub-problem based on the problem features to obtain intention information; determine the task information and scene information indicated by the intention information to determine the problem features corresponding to the sub-problem based on the task information and the scene information; and determine the target agent that is adapted to the sub-problem from the agent group based on the problem features.

[0139] Optionally, in some possible implementations, the processing unit 802 is used to obtain an assistive tool corresponding to the task information; call an extended tool configured in the agent group; and enhance the configuration of the target agent through the assistive tool and the extended tool.

[0140] Optionally, in some possible implementations, the processing unit 802 is used to distribute the sub-problem to a backup agent for processing if there is no target agent suitable for the sub-problem in the agent group, and record the question information corresponding to the sub-problem. The backup agent is used to provide a general answer function, and the question information is used to guide the agent registration process in the agent group.

[0141] Optionally, in some possible implementations, the answer unit 803 is used to determine the key elements corresponding to the sub-problem when processing the sub-problem based on the target intelligent agent to obtain the answer information corresponding to the problem to be processed, and to judge the integrity of the key elements; if the key elements are complete, processing the sub-problem based on the target intelligent agent to obtain a processing result; rendering the processing result according to the answer template to obtain the answer information.

[0142] Optionally, in some possible implementations, the answer unit 803 is used to determine the missing element corresponding to the sub-problem based on the key element if the key element is incomplete; generate follow-up information based on the missing element; and obtain feedback information corresponding to the follow-up information to trigger the target intelligent agent to process the sub-problem based on the feedback information.

[0143] Optionally, in some possible implementations, the answer unit 803 is used to, when the key element is complete, process the sub-problem based on the target intelligent agent to obtain a processing result, if the key element is complete, perform complexity judgment on the sub-problem to obtain verification information; compare the verification information with the complexity information to obtain a convergence parameter; if the convergence parameter meets the answer condition, process the sub-problem based on the target intelligent agent to obtain the processing result.

[0144] By obtaining the pending problem initiated by the agent group, the complexity of the pending problem is determined to obtain complexity information. If the complexity information indicates that the pending problem is a complex problem, the pending problem is decomposed to obtain multiple sub-problems, and the problem features corresponding to the sub-problems are extracted to determine the target agent in the agent group that is adapted to the sub-problem according to the problem features. Then, the sub-problem is executed based on the target agent to obtain the answer information corresponding to the pending problem. Thus, the question answering process based on the agent group is realized. By decomposing complex problems and adapting targeted agents in the agent group, the professionalism of the agent in understanding the problem is guaranteed, and the accuracy of the question answering is improved.

[0145] The question answering device provided in this embodiment belongs to the same application concept as the method provided in the above embodiments of this application, can execute the question answering method provided in any of the above embodiments of this application, and has the corresponding functional modules and beneficial effects of executing the question answering method. For technical details not fully described in this embodiment, please refer to the specific processing content of the question answering method provided in the above embodiments of this application, which will not be repeated here.

[0146] The functions implemented by the above acquisition unit 801, processing unit 802 and answering unit 803 may be implemented by the same or different processors respectively, which is not limited in the embodiment of the present application.

[0147] It should be understood that each unit in the above device can be implemented in the form of a processor calling software. For example, the device includes a processor, the processor is connected to a memory, and instructions are stored in the memory. The processor calls the instructions stored in the memory to implement any of the above methods or realize the functions of each unit of the device, wherein the processor can be a general-purpose processor, such as a CPU or a microprocessor, etc., and the memory can be a memory in the device or a memory outside the device. Alternatively, the unit in the device can be implemented in the form of a hardware circuit, and the functions of some or all units can be realized by designing the hardware circuit. The hardware circuit can be understood as one or more processors; for example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all units above are realized by designing the logical relationship of the components in the circuit; for another example, in another implementation, the hardware circuit can be implemented by PLD, taking FPGA as an example, which can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to realize the functions of some or all units above. All units of the above device can be implemented in the form of a processor calling software, or in the form of a hardware circuit, or in the form of a processor calling software, and the remaining part can be implemented in the form of a hardware circuit.

[0148] In an embodiment of the present application, a processor is a circuit with the ability to process signals. In one implementation, the processor may be a circuit with the ability to read and run instructions, such as a CPU, a microprocessor, a GPU, or a DSP; in another implementation, the processor may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or reconfigurable, such as a hardware circuit implemented by an ASIC or PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document to implement the hardware circuit configuration can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, DPU, etc.

[0149] It can be seen that each unit in the above device can be one or more processors (or processing circuits) configured to implement the above method, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0150] In addition, all or part of the units in the above device can be integrated together, or can be implemented independently. In one implementation, these units are integrated together and implemented in the form of a SOC. The SOC may include at least one processor for implementing any of the above methods or implementing the functions of each unit of the device. The type of the at least one processor may be different, for example, including a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.

[0151] The embodiment of the present application also provides a question answering device; the question answering device includes a processor and an input-output component. The input-output component is used to obtain a question to be processed;

[0152] The processor is used to answer the pending question obtained by the input-output component by executing any one of the question answering methods in the above embodiments.

[0153] The above-mentioned interface circuit can be any interface circuit that can realize the data communication function, for example, it can be a USB interface circuit, a Type-C interface circuit, a serial port circuit, a PCIE circuit, etc.

[0154] Optionally, the embodiment of the present application further provides a system, wherein the interactive client is used to obtain the pending questions and send the pending questions to the server, and to display the answer results of the questions output by the server;

[0155] The server is used to answer the pending questions obtained by the interactive client by executing the question answering method described in any one of the above embodiments.

[0156] Another embodiment of the present application also provides a question answering device, see Fig. 9 As shown, the device includes:

[0157] Memory 900 and processor 910;

[0158] The memory 900 is connected to the processor 910 and is used to store programs;

[0159] The processor 910 is configured to implement the question answering method disclosed in any of the above embodiments by running the program stored in the memory 900 .

[0160] Specifically, the question answering device may further include: a bus, a communication interface 920 , an input device 930 and an output device 940 .

[0161] The processor 910, the memory 900, the communication interface 920, the input device 930 and the output device 940 are connected to each other via a bus.

[0162] A bus may include a pathway that transfers information between components of a computer system.

[0163] The processor 910 may be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the scheme of the present invention. It may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0164] The processor 910 may include a main processor, and may also include a baseband chip, a modem, and the like.

[0165] The memory 900 stores a program for executing the technical solution of the present invention, and may also store an operating system and other key services. Specifically, the program may include a program code, and the program code includes a computer operation instruction. More specifically, the memory 900 may include a read-only memory (ROM), other types of static storage devices that can store static information and instructions, a random access memory (RAM), other types of dynamic storage devices that can store information and instructions, a disk storage, a flash, and the like.

[0166] The input device 930 may include a device for receiving data and information input by a user, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor.

[0167] Output device 940 may include devices that allow information to be output to a user, such as a display screen, printer, speaker, etc.

[0168] The communication interface 920 may include any transceiver or the like to communicate with other devices or communication networks, such as Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc.

[0169] The processor 910 executes the program stored in the memory 900 and calls other devices, which can be used to implement each step of any question answering method provided in the above embodiments of the present application.

[0170] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the question answering method according to various embodiments of the present application described in any of the above embodiments of this specification.

[0171] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0172] In addition, the embodiment of the present application may also be a storage medium on which a computer program is stored. The computer program is executed by a processor to execute the steps of the question answering method according to various embodiments of the present application described in any of the above embodiments of this specification, and specifically the following steps may be implemented:

[0173] Get pending issues for the group of agents;

[0174] Performing complexity determination on the problem to be processed to obtain complexity information;

[0175] If the complexity information indicates that the problem to be processed is a complex problem, decomposing the problem to be processed to obtain a plurality of sub-problems;

[0176] Extracting problem features corresponding to the sub-problems, so as to determine a target agent in the agent group that is adapted to the sub-problems according to the problem features;

[0177] The sub-problem is processed based on the target agent to obtain answer information corresponding to the problem to be processed.

[0178] For the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0179] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0180] The steps in the methods of each embodiment of the present application can be adjusted in order, combined and deleted according to actual needs, and the technical features recorded in each embodiment can be replaced or combined.

[0181] The modules and sub-modules in the devices and terminals of the various embodiments of the present application can be combined, divided and deleted according to actual needs.

[0182] In the several embodiments provided in the present application, it should be understood that the disclosed terminals, devices and methods can be implemented in other ways. For example, the terminal embodiments described above are only schematic, for example, the division of modules or submodules is only a logical function division, and there may be other division methods in actual implementation, for example, multiple submodules or modules can be combined or integrated into another module, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0183] The modules or submodules described as separate components may or may not be physically separated, and the components of the modules or submodules may or may not be physical modules or submodules, that is, they may be located in one place, or they may be distributed on multiple network modules or submodules. Some or all of the modules or submodules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0184] In addition, each functional module or submodule in each embodiment of the present application may be integrated into one processing module, or each module or submodule may exist physically separately, or two or more modules or submodules may be integrated into one module. The above-mentioned integrated modules or submodules may be implemented in the form of hardware or in the form of software functional modules or submodules.

[0185] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0186] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly by hardware, software units executed by a processor, or a combination of the two. The software units may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0187] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0188] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A question answering method, characterized in that: include: Get pending issues for the group of agents; Performing complexity determination on the problem to be processed to obtain complexity information; If the complexity information indicates that the problem to be processed is a complex problem, decomposing the problem to be processed to obtain a plurality of sub-problems; Extracting problem features corresponding to the sub-problems, so as to determine a target agent in the agent group that is adapted to the sub-problems according to the problem features; The sub-problem is processed based on the target agent to obtain answer information corresponding to the problem to be processed.

2. The method according to claim 1, characterized in that: The determining the complexity of the problem to be processed to obtain complexity information includes: Rewriting the problem to be processed to obtain a rewritten problem; Performing complexity determination on the rewriting problem based on multiple preset dimensions to obtain complexity scores corresponding to each preset dimension; A weighted calculation is performed on the complexity scores corresponding to the respective preset dimensions to obtain the complexity information.

3. The method according to claim 2, characterized in that The determining the complexity of the rewriting problem based on multiple preset dimensions to obtain a complexity score corresponding to each preset dimension includes: Determining the interactive information corresponding to the rewriting question based on the preset dimension; determining problem areas indicated by the interaction information; determining correlation parameters between the problem areas; Complexity determination is performed according to the interaction information and the relevance parameter respectively to obtain complexity scores corresponding to the interaction information and the relevance parameter respectively.

4. The method according to claim 1, characterized in that: If the complexity information indicates that the problem to be processed is a complex problem, the problem to be processed is decomposed to obtain multiple sub-problems, including: If the complexity information indicates that the problem to be processed is a complex problem, decomposing the problem to be processed based on a preset dimension to obtain a decomposed problem; Determine the correlation relationship between the disassembly problems to obtain serial and parallel information between the disassembly problems; The disassembly problem is configured based on the serial-parallel information to obtain a plurality of the sub-problems.

5. The method according to claim 1, characterized in that The extracting the problem features corresponding to the sub-problems to determine the target agent in the agent group adapted to the sub-problems according to the problem features includes: Performing intent recognition on the sub-questions to obtain intent information; Determine the task information and the scenario information indicated by the intention information, so as to determine the problem feature corresponding to the sub-problem based on the task information and the scenario information; A target intelligent agent that is adapted to the sub-problem is determined from the intelligent agent group according to the problem characteristics.

6. The method according to claim 5, characterized in that The method further comprises: Obtaining an assistance tool corresponding to the task information; Calling the extended tool configured in the agent group; The target agent is enhanced and configured through the assisting tool and the extending tool.

7. The method according to claim 5, characterized in that The method further comprises: If there is no target agent suitable for the sub-problem in the agent group, the sub-problem will be distributed to the underlying agent for processing, and the question information corresponding to the sub-problem will be recorded. The underlying agent is used to provide a general answer function, and the question information is used to guide the agent registration process in the agent group.

8. The method according to claim 1, characterized in that The processing of the sub-problem based on the target intelligent agent to obtain answer information corresponding to the problem to be processed includes: Determine the key elements corresponding to the sub-questions, and make integrity judgment on the key elements; If the key elements are complete, the sub-problems are processed based on the target agent to obtain processing results; The processing result is rendered according to the answer template to obtain the answer information.

9. The method according to claim 8, characterized in that The method further comprises: If the key element is incomplete, determining the missing element corresponding to the sub-question according to the key element; generating follow-up information according to the missing element; Acquire feedback information corresponding to the follow-up information, so as to trigger the target agent to process the sub-question based on the feedback information.

10. The method according to claim 8, characterized in that If the key elements are complete, processing the sub-problems based on the target agent to obtain processing results includes: If the key elements are complete, the complexity of the sub-problems is determined to obtain verification information; Comparing the verification information with the complexity information to obtain a convergence parameter; If the convergence parameter satisfies the answer condition, the sub-problem is processed based on the target agent to obtain the processing result.

11. A question answering device, characterized in that: include: An acquisition unit, used to acquire pending problems initiated by the group of intelligent agents; A processing unit, used for determining the complexity of the problem to be processed to obtain complexity information; The processing unit is further configured to decompose the problem to be processed to obtain a plurality of sub-problems if the complexity information indicates that the problem to be processed is a complex problem; The processing unit is further used to extract the problem features corresponding to the sub-problem, so as to determine the target agent in the agent group that is adapted to the sub-problem according to the problem features; An answering unit is used to process the sub-problem based on the target agent to obtain answer information corresponding to the problem to be processed.

12. A question answering device comprising an input-output component and a processor; The input and output components are used to obtain problems to be processed; The processor is used to answer the pending question obtained by the input-output component by executing the question answering method as described in any one of claims 1 to 10.

13. A question answering system, comprising an interactive client and a server; The interactive client is used to obtain the pending questions and send the pending questions to the server, and to display the answer results of the questions output by the server; The server is used to answer the pending questions obtained by the interactive client by executing the question answering method described in any one of claims 1 to 10.

14. A computer program product, characterized in that include: A computer program, wherein when the computer program is executed by a processor, the question answering method according to any one of claims 1 to 10 is implemented.

Images