Knowledge question-answering method and system, electronic equipment and storage medium

Abstract

The invention discloses a knowledge question and answer method and system, electronic equipment and a storage medium. The method comprises the steps of obtaining question information of an object; performing intention recognition processing on the question information through a decision maker agent, and determining intention information; according to the intention information, target sub-agents needing to be called are determined, directed paths between the target sub-agents and the decision maker agent and communication paths between the target sub-agents are established, and a question and answer system is formed; and answering the intention information based on the question and answer system to obtain a target answer corresponding to the question information. According to the method, the directed path between the sub-agents and the decision maker agent and the communication path between the sub-agents are established, so that the agents work cooperatively, and the answering accuracy of the question and answer system is improved. The method can be widely applied to the technical field of large models.

Description

Technical Field

[0001] The present application relates to the field of large model technology, and in particular to a knowledge question and answer method, system, electronic device and storage medium. Background Art

[0002] Large language models are mainly used to solve single-agent task planning problems. However, due to the misalignment between large model knowledge and specific embodied environments, the plans generated by large models are often difficult to execute in the environment. In a multi-agent environment, multiple agents are usually controlled using large models; when multiple agents work together to complete a task, in addition to encountering environmental incompatibility issues similar to single agents, there is also the problem of how multiple agents can communicate and collaborate efficiently. The method of directly collaborating through multi-agent dialogue and negotiation is inefficient. On the one hand, it is difficult to derive an effective collaborative strategy entirely through dialogue; on the other hand, it is impossible to measure the contribution of a single agent to the overall task goal, and it is difficult to drive each agent to improve its strategy to increase overall benefits. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to provide an efficient knowledge question answering method, system, electronic device and storage medium.

[0004] To achieve the above-mentioned purpose, one aspect of an embodiment of the present application proposes a knowledge question-answering method, the method comprising: obtaining the question information of an object; performing intent recognition processing on the question information by a decision-maker agent to determine the intent information; determining the target sub-agent to be called according to the intent information, and establishing a directed path between the target sub-agent and the decision-maker agent, and a communication path between the target sub-agents, to form a question-answering system; answering the intent information based on the question-answering system, obtaining the target answer corresponding to the question information, and feeding back the target answer to the object. The embodiment of the present application enables the agents to work together by establishing a directed path between the sub-agent and the decision-maker agent, and a communication path between the sub-agents, thereby improving the answer accuracy of the question-answering system.

[0005] In some embodiments, the method provided in the embodiments of the present application determines the target sub-agent to be called according to the intention information, and establishes a communication path between the target sub-agents; including:

[0006] According to all sub-agents, determining the weighted adjacency matrix corresponding to the target sub-agent and the coupling strength of each sub-agent;

[0007] The communication path between the target sub-agents is determined according to the weighted adjacency matrix and the coupling strength.

[0008] In some embodiments, the method provided in the embodiments of the present application determines the target sub-agent to be called according to the intention information, and establishes a directed path between the target sub-agent and the decision maker agent, including:

[0009] A directed path is established from the decision maker agent to the target sub-agent, and the target sub-agent is set as a pinning node to synchronize the target sub-agent with the decision maker agent.

[0010] In some embodiments, the method provided by the embodiments of the present application, determining the target sub-agent according to the intention information, includes:

[0011] Decomposing the intention information to determine segment information;

[0012] Based on the domain information of the segmented information, a comparative analysis is performed with the applicable domain information of each of the sub-agents to determine the target sub-agent.

[0013] In some embodiments, the method provided by the embodiments of the present application, answering the intention information based on the question-answering system to obtain a target answer corresponding to the question information, includes:

[0014] Input the intention information into each of the target sub-agents to obtain a number of answer information;

[0015] The target answer is obtained by screening and integrating the plurality of answer information through the decision maker intelligent agent.

[0016] In some embodiments, the method provided by the embodiments of the present application, wherein the decision maker agent screens and integrates the plurality of answer information to obtain the target answer, comprises:

[0017] Obtaining trajectory information of the target sub-agent during interaction;

[0018] Determine the report information of each track in the track information, and determine the state-action pair corresponding to each report information;

[0019] Averaging the reward information with the same state-action pair to obtain a first function;

[0020] The second function is obtained by averaging the first function values ​​of all actions in each state.

[0021] constructing an advantage function according to the first function and the second function;

[0022] Based on the advantage function, the plurality of answer information are screened and integrated to obtain the target answer.

[0023] In some embodiments, in the method provided by the embodiments of the present application, the pinning node is established by the following formula:

[0024]

[0025] Among them, x i is the target sub-agent node, t is the time, s(t) is the decision maker agent, f(x i ) is the node x i The self-dynamic behavior of ij is the weighted adjacency matrix, c is the coupling strength, d is i is the diversionary control path of the decision maker agent.

[0026] To achieve the above purpose, another aspect of the embodiment of the present application provides a knowledge question answering system, the system comprising:

[0027] The first module is used to obtain the question information of the object;

[0028] The second module is used to perform intention recognition processing on the question information through the decision maker intelligent agent to determine the intention information;

[0029] The third module is used to determine the target sub-agent to be called according to the intention information, and to establish a directed path between the target sub-agent and the decision-maker agent and a communication path between the target sub-agents to form a question-answering system;

[0030] The fourth module is used to answer the intention information based on the question-answering system, obtain the target answer corresponding to the question information, and feed back the target answer to the object.

[0031] To achieve the above objective, another aspect of an embodiment of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above method when executing the computer program.

[0032] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0033] The embodiments of the present application include at least the following beneficial effects: The method provided by the embodiments of the present application includes: obtaining the question information of the object; performing intent recognition processing on the question information by the decision maker agent to determine the intent information; determining the target sub-agent to be called according to the intent information, and establishing a directed path between the target sub-agent and the decision maker agent, and a communication path between the target sub-agents, to form a question-and-answer system; answering the intent information based on the question-and-answer system, obtaining the target answer corresponding to the question information, and feeding back the target answer to the object. The embodiments of the present application enable the agents to work together by establishing a directed path between the sub-agent and the decision maker agent, and a communication path between the sub-agents, thereby improving the answer accuracy of the question-and-answer system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a flow chart of an embodiment of the knowledge question answering method provided by the present application;

[0035] Figure 2 is a flow chart of another embodiment of the knowledge question answering method provided by the present application;

[0036] Figure 3 It is a structural diagram of an embodiment of the multi-agent architecture provided by the present application;

[0037] Figure 4 is a flow chart of an embodiment of establishing a communication path provided by the present application;

[0038] Figure 5 It is a structural diagram of an embodiment of the question-answering system provided by the present application;

[0039] Figure 6 is a flow chart of an embodiment of establishing a directed path provided by the present application;

[0040] Figure 7 This is a flowchart of an embodiment of the whole process of knowledge question answering provided by this application;

[0041] Figure 8 This is a display diagram of an embodiment of the knowledge question and answer interface provided by this application;

[0042] Fig. 9 is a schematic diagram of the structure of the knowledge question answering system provided in an embodiment of the present application;

[0043] Fig.10 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0045] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0046] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0048] Before describing the embodiments of the present application in detail, some nouns and terms involved in the embodiments of the present application are first described. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.

[0049] An agent is a component that has autonomous behavior, can perceive the environment, understand information, make decisions and perform corresponding actions. In the knowledge intelligent question answering system, agents play a core role. They are responsible for receiving user query requests, analyzing user intentions, and retrieving and integrating relevant information in the knowledge base or data resources, and finally generating accurate and useful answers to return to users.

[0050] Multi-agent: Multi-intelligence refers to a system or system composed of multiple agents. Each agent has its own perception, decision-making and action capabilities, and can achieve the control objectives of the system by communicating and cooperating with other agents.

[0051] With the rapid development of 5G technology, the complexity and professionalism of 5G customized networks, as key infrastructure supporting the digital transformation of vertical industries, are becoming increasingly prominent, and there is an urgent need for efficient and accurate knowledge question and answer. The knowledge system of 5G customized networks covers everything from basic 5G communication technology to application scenarios and atomic capabilities unique to each industry. This highly specialized knowledge demand makes efficient and accurate knowledge question and answer systems an indispensable tool.

[0052] Previously, large language models were mainly used in embodied intelligence to solve the task planning problem of a single agent. However, due to the misalignment between the knowledge of the large model and the specific embodied environment, the planning generated by the large model is often difficult to execute in the environment. At this time, the large model needs to adapt to its own environment through multiple rounds of interaction and feedback with the environment, which has a high interaction cost. In a multi-agent environment, most of the methods currently adopted are for the agent to use a large model for control. When multiple agents work together to complete tasks, in addition to encountering environmental incompatibility problems similar to single agents, there is also the problem of how multiple agents can communicate and collaborate efficiently. If the method of collaboration is directly carried out through multi-agent dialogue and negotiation, it is inefficient. On the one hand, it is difficult to derive an effective collaborative strategy completely through dialogue; on the other hand, it is impossible to measure the contribution of a single agent to the overall task goal (i.e., credit distribution), and it is difficult to drive each agent to improve its strategy to increase overall benefits.

[0053] In view of this, it is particularly important and necessary to build a 5G customized network knowledge intelligent question-answering system based on distributed collaborative control of multi-agent collaboration. The system builds multiple agents, each of which focuses on knowledge processing in different fields to form specialized knowledge processing capabilities. At the same time, by building a distributed control mechanism for regulation, the agents can work together, share information, and complement each other's advantages, so that they can understand user queries more comprehensively, locate relevant knowledge more accurately, and generate question-answering results that meet industry characteristics and user needs.

[0054] The knowledge question and answer method provided in the embodiment of the present application relates to the field of large model technology. The knowledge question and answer method provided in the embodiment of the present application can be applied to a terminal, can be applied to a server, or can be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or it can be configured as a server cluster or distributed system composed of multiple physical servers, and can also be configured to provide 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 and other basic cloud computing services. The cloud server, the server can also be a node server in the blockchain network; the software can be an application that implements the knowledge question and answer method, etc., but is not limited to the above forms.

[0055] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0056] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0057] Figure 1 is an optional flow chart of the knowledge question answering method provided in the embodiment of the present application; Figure 1The method may include but is not limited to steps S100 to S400.

[0058] Step S100, obtaining the question information of the object;

[0059] Step S200, performing intention recognition processing on the question information by the decision maker agent to determine the intention information;

[0060] Step S300, determining the target sub-agent to be called according to the intention information, and establishing a directed path between the target sub-agent and the decision-maker agent, and a communication path between the target sub-agents, to form a question-answering system;

[0061] Step S400, answer the intention information based on the question-answering system, obtain the target answer corresponding to the question information, and feed back the target answer to the object.

[0062] The multi-agent in the embodiment of the present application may include a decision maker agent and multiple sub-agents. Figure 2 An embodiment shown, Figure 2 The 5G customized network application scenario question-and-answer agent, the 5G customized network atomic question-and-answer agent, the 5G terminal question-and-answer agent, and the 5G customized network standard product question-and-answer agent are the sub-agents in the embodiments of this application. Of course, those skilled in the art can also set the specific applicable fields of the sub-agents according to actual needs, and this application does not make specific limitations. The embodiments of this application construct a distributed collaborative control protocol for multiple agents by establishing directed paths between sub-agents and decision-maker agents and communication paths between sub-agents, so that each agent can work in coordination. Specifically, refer to Figure 3 As shown, for multiple agents in the embodiment of the present application:

[0063] A. 5G Customized Network Question and Answer Decision Maker Agent: The decision maker agent, as the "brain" of the system, is responsible for receiving and analyzing the queries input by users and understanding the user's intentions through natural language processing technology. Based on the understanding results, the agent decision maker will intelligently dispatch other agents to work together. At the same time, the answers output by each sub-agent will be fed back to the decision maker agent, and the decision maker agent will eliminate the unfavorable ones and store them to assist the next decision. For example, if the user queries about a certain application scenario of the 5G customized network, the agent decision maker will start the 5G customized network application scenario question and answer agent to complete the question and answer of the user's question.

[0064] B. 5G customized network application scenario question-answering agent: This agent can quickly provide accurate and comprehensive answers to users' questions about 5G customized network application scenarios. This agent can call the application scenario search tool to complete the application scenario knowledge search and transmit the results to the application scenario question answering tool to realize the application scenario question-answering service.

[0065] C. 5G Customized Network Atomic Capability Q&A Agent: This agent focuses on answering questions about 5G customized network atomic capabilities (such as network slicing, edge computing, etc.). It can use atomic capability search tools and atomic capability question answering tools to implement atomic capability information search services and atomic capability Q&A services, providing users with professional and in-depth answering services.

[0066] D. 5G customized network terminal question and answer agent: This agent can provide 5G terminal question and answer services. It calls the terminal search tool to complete the terminal manufacturer, characteristics, price and other data, and calls the terminal question answering tool to realize the question and answer of 5G terminals, ensuring the timeliness and accuracy of the answer;

[0067] E. 5G Customized Network Standard Product Q&A Agent: In response to user inquiries about 5G Customized Network Standard products, this agent can provide detailed information on product functions, specifications, application scenarios, etc. It provides users with clear and accurate answers based on standardized product data and descriptions, helping users better understand and choose products that suit them.

[0068] This application aims to achieve comprehensive coverage and efficient question-answering of 5G customized network knowledge by integrating multiple specialized agents and building a distributed multi-agent collaborative control mechanism. The system consists of decision-maker agents, 5G customized network application scenario question-answering agents, 5G customized network atomic question-answering agents, 5G terminal question-answering agents, and 5G customized network standard product question-answering agents. Each agent is controlled by a distributed collaborative control algorithm to ensure close coordination between the agents and jointly complete complex question-answering tasks, providing accurate and reliable intelligent question-answering services for users in the field of 5G customized networks.

[0069] In some embodiments, reference Figure 4 As shown, the method provided in the embodiment of the present application determines the target sub-agent to be called according to the intention information, and establishes a communication path between the target sub-agents, including:

[0070] Step S310: Determine the weighted adjacency matrix corresponding to the target sub-agent and the coupling strength of each sub-agent according to all sub-agents;

[0071] Step S320: Determine the communication path between target sub-agents based on the weighted adjacency matrix and the coupling strength.

[0072] In some possible implementations, refer to Figure 5 As shown, a communication path between sub-agents is established, the communication between sub-agents is strengthened, and the prediction accuracy is improved.

[0073] In some embodiments, reference Figure 6 As shown, the method provided in the embodiment of the present application determines the target sub-agent to be called according to the intention information, and establishes a directed path between the target sub-agent and the decision maker agent, including:

[0074] Step S330: establish a directed path from the decision maker agent to the target sub-agent, and set the target sub-agent as a pinning node to synchronize the target sub-agent with the decision maker agent.

[0075] In some embodiments, the method provided in the embodiments of the present application determines the target sub-agent according to the intention information, including:

[0076] Disassemble the intent information and determine the segmented information;

[0077] Based on the domain information of the segmented information, a comparative analysis is performed with the applicable domain information of each sub-agent to determine the target sub-agent.

[0078] In some possible implementations, refer to Figure 7 As shown, the embodiment of the present application disassembles the intention information to determine the segment information; then the sub-agent can be matched with the domain information reflected in the segment information, and then the target sub-agent can be determined. Of course, those skilled in the art can also match the segment information with the sub-agent through other information, and this application does not make specific limitations.

[0079] In some embodiments, the method provided in the embodiments of the present application answers the intention information based on the question-answering system to obtain the target answer corresponding to the question information, including:

[0080] Input the intention information into each target sub-agent and obtain some answer information;

[0081] The target answer is obtained by screening and integrating several answer information through the decision-maker intelligent agent.

[0082] In some possible implementations, the question-and-answer system in the embodiments of the present application answers the intention information through multiple sub-agents to obtain answer information, and then the decision-maker agent screens and integrates the answer information to obtain the target answer.

[0083] In some embodiments, the method provided by the embodiments of the present application screens and integrates a plurality of answer information through a decision maker agent to obtain a target answer, including:

[0084] Obtaining trajectory information of the target sub-agent during interaction;

[0085] Determine the report information of each track in the track information, and determine the state-action pair corresponding to each report information;

[0086] Averaging the reward information with the same state-action pair to obtain a first function;

[0087] The second function is obtained by averaging the first function values ​​of all actions in each state.

[0088] According to the first function and the second function, construct the advantage function;

[0089] Based on the advantage function, several answer information are screened and integrated to obtain the target answer.

[0090] In some possible implementations, the process of screening and integrating a plurality of answer information is not limited to the above-mentioned method.

[0091] In some embodiments, in the method provided by the embodiments of the present application, the pinning node is established by the following formula:

[0092]

[0093] Where i = 1, 2, 3...N, N is the number of target sub-agents, x i is the target sub-agent node, t is the time, s(t) is the decision maker agent, f(x i ) is the node x i The self-dynamic behavior of ij is the weighted adjacency matrix, c is the coupling strength, d is i A diversionary control path for the decision-maker agent.

[0094] The establishment of the question-and-answer system and the question-and-answer process in the embodiment of the present application are as follows:

[0095] Step S21: Establish a controlled node:

[0096] Make each intelligent agent a controlled node under the distributed control protocol, and the formula is:

[0097] x i =f(x i ), s(t)=f(s(t));

[0098] Where i = 1, 2, 3, ..., N, where x iis the node, t is the time, s(t) is the decision maker agent, f(x i ) is the dynamic behavior of node i itself, which represents the set of adjacent units of node i.

[0099] Step S22: Establish a distributed control protocol:

[0100] Distributed control of multiple sub-agents is used to reduce the communication between agents as much as possible and maintain system performance. The formula is:

[0101]

[0102] where a ij is the weighted adjacency matrix, c is the coupling strength, which represents the interaction between agents.

[0103] Step S23: containment control:

[0104] When users ask questions, the big model can analyze the questions through the decision-maker agent through intent recognition. The decision-maker agent can wake up the sub-agent through restraint control, so that the sub-agent can synchronize with the decision-maker agent and conduct knowledge search and response.

[0105] The control goal is to make each sub-agent and the decision-maker agent consistent, so that the information of the decision-maker agent is fed back to the system dynamics. If and only if there is a directed path from the decision-maker to the i-th node, the node is called a pinning node or a controlled node. The establishment of the directed path is shown in the agent node display Figure 5 As shown in , it is the path established from the leader node to the controlled node. Its purpose is to drive the sub-agents to synchronize with the decision-maker agent by controlling a small number of sub-agents, that is, pinning control. The pinning control node is formulated as:

[0106]

[0107] where d i That is, the control path for the decision maker agent.

[0108] Step S24, introduce advantage function to analyze the results:

[0109] Each sub-agent feeds back the response results to the decision-maker agent. After the decision-maker agent analyzes the results separately, it uses the Monte Carlo estimation method to learn the advantage function of the group answer and extract the essence of the answer and remove the dross.

[0110] Step S25: After the answers are merged and the analysis results are stored, the decision maker agent submits the answers to the big model, and the big model organizes the language to deliver the answers to the user.

[0111] The following is a detailed description and explanation of the solution of the embodiment of the present invention in conjunction with a specific application example:

[0112] Based on the indicative question-and-answer method provided above, a 5G capability Rubik's Cube intelligent question-and-answer system is established. The system is an intelligent assistant specially created by a certain communication operator for users in the 5G customized network industry. The system integrates 5G customized network application scenarios, atomic capabilities, terminals, and standard product question-and-answer capabilities, and can quickly respond to customer needs and provide accurate and reliable answers. The system includes application scenario question-and-answer agents, atomic capability question-and-answer agents, terminal question-and-answer agents, and standard product question-and-answer agents, which can decompose professional issues of 5G customized networks and obtain answers to sub-questions through different agents. It is a powerful tool for the communication operator's 5G customized network professionals. Reference Figure 8 As shown in the system application interface, the system can accurately determine the user's intention and give the user satisfactory answer information.

[0113] The present invention aims to propose a customized network knowledge intelligent question and answer system based on distributed collaborative control of multi-agent collaboration. Its core technical point is the distributed collaborative control protocol. The system innovatively designs a distributed collaborative control protocol for multiple agents and introduces an advantage function, so that the decision-maker agent can efficiently control agents including application scenario question and answer agents, atomic capability question and answer agents, terminal question and answer agents, and standard product question and answer agents, so that each agent can focus on question and answer processing in a specific field, and jointly achieve comprehensive analysis and question and answer of complex 5G customized network problems.

[0114] See also Fig. 9 The present application also provides a knowledge question answering system, which can implement the above-mentioned knowledge question answering method. The system includes:

[0115] The first module 810 is used to obtain the question information of the object;

[0116] The second module 820 is used to perform intention recognition processing on the question information through the decision maker agent to determine the intention information;

[0117] The third module 830 is used to determine the target sub-agent to be called according to the intention information, and establish a directed path between the target sub-agent and the decision maker agent and a communication path between the target sub-agents to form a question-answering system;

[0118] The fourth module 840 is used to answer the intention information based on the question-answering system, obtain the target answer corresponding to the question information, and feed back the target answer to the object.

[0119] It can be understood that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0120] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned knowledge question answering method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0121] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0122] See also Fig.10 , Fig.10 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0123] The processor 901 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0124] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other applications. When the technical solution provided in the embodiments of this specification is implemented by software or firmware, the relevant program code is stored in the memory 902, and the processor 901 calls and executes the knowledge question and answer method of the embodiments of this application;

[0125] Input / output interface 903, used to implement information input and output;

[0126] Communication interface 904, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0127] A bus 905 that transmits information between various components of the device (e.g., the processor 901, the memory 902, the input / output interface 903, and the communication interface 904);

[0128] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0129] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the above-mentioned knowledge question and answer method is implemented.

[0130] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0131] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0132] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0133] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0134] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0135] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0136] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0137] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0138] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, 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 through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0139] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0140] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0141] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0142] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A knowledge question answering method, characterized in that: The method comprises: Get the object's question information; Performing intention recognition processing on the question information by a decision maker intelligent agent to determine the intention information; According to the intention information, determine the target sub-agent to be called, and establish a directed path between the target sub-agent and the decision-maker agent, and a communication path between the target sub-agents to form a question-answering system; The intention information is answered based on the question-answering system to obtain a target answer corresponding to the question information, and the target answer is fed back to the object.

2. The method according to claim 1, characterized in that: According to the intention information, determining the target sub-agent to be called, and establishing a communication path between the target sub-agents; including: According to all sub-agents, determining the weighted adjacency matrix corresponding to the target sub-agent and the coupling strength of each sub-agent; The communication path between the target sub-agents is determined according to the weighted adjacency matrix and the coupling strength.

3. The method according to claim 1, characterized in that According to the intention information, a target sub-agent to be called is determined, and a directed path between the target sub-agent and the decision maker agent is established, including: A directed path is established from the decision maker agent to the target sub-agent, and the target sub-agent is set as a pinning node to synchronize the target sub-agent with the decision maker agent.

4. The method according to claim 1, characterized in that: Determining the target sub-agent according to the intention information includes: Decomposing the intention information to determine segment information; Based on the domain information of the segmented information, a comparative analysis is performed with the applicable domain information of each of the sub-agents to determine the target sub-agent.

5. The method according to claim 1, characterized in that The answering the intention information based on the question-answering system to obtain a target answer corresponding to the question information includes: Input the intention information into each of the target sub-agents to obtain a number of answer information; The target answer is obtained by screening and integrating the plurality of answer information through the decision maker intelligent agent.

6. The method according to claim 5, characterized in that The step of screening and integrating the plurality of answer information by the decision maker agent to obtain the target answer includes: Obtaining trajectory information of the target sub-agent during interaction; Determine the report information of each track in the track information, and determine the state-action pair corresponding to each report information; Averaging the reward information with the same state-action pair to obtain a first function; The second function is obtained by averaging the first function values ​​of all actions in each state. constructing an advantage function according to the first function and the second function; Based on the advantage function, the plurality of answer information are screened and integrated to obtain the target answer.

7. The method according to claim 3, characterized in that The pinning node is established by the following formula: Among them, x i is the target sub-agent node, t is the time, s(t) is the decision maker agent, f(x i ) is the node x i The self-dynamic behavior of ij is the weighted adjacency matrix, c is the coupling strength, d i is the diversionary control path of the decision maker agent.

8. A knowledge question answering system, characterized in that: The system comprises: The first module is used to obtain the question information of the object; The second module is used to perform intention recognition processing on the question information through the decision maker intelligent agent to determine the intention information; The third module is used to determine the target sub-agent to be called according to the intention information, and to establish a directed path between the target sub-agent and the decision-maker agent and a communication path between the target sub-agents to form a question-answering system; The fourth module is used to answer the intention information based on the question-answering system, obtain the target answer corresponding to the question information, and feed back the target answer to the object.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images