Updating method and device of intelligent question-answering system, equipment and storage medium

Abstract

The invention discloses an updating method, device and equipment of an intelligent question-answering system and a computer readable storage medium, and the method comprises the steps that the intelligent question-answering system receives input information of a user, and obtains corresponding question information; according to the question information, determining a type to which the question information belongs, so as to obtain a corresponding answer and provide the answer to the user; obtaining feedback information sent by the user for the answer; obtaining feedback information sent by the user; if the feedback information is unsolved, performing manual answering according to a manual mechanism to obtain a corresponding target answer; the target answer and the question information are added to a knowledge base, so that the intelligent question answering system is updated, the intelligent question answering system comprises the knowledge base, the technical problem that in the related technology, the answering efficiency and accuracy of an intelligent question answering system are low is solved, and the user experience is improved. And the answering efficiency, the accuracy, the user experience and the overall performance of the intelligent question-answering system are remarkably improved.

Description

Technical Field

[0001] This application relates to the technical field of intelligent question answering, and particularly to an update method, device, equipment and computer-readable storage medium for an intelligent question answering system. Background Art

[0002] Current intelligent question answering systems have been widely used in various application scenarios, such as enterprise internal knowledge bases, customer service, and technical support. However, there are mainly some problems in the existing systems: they usually only support text input and cannot process the picture information uploaded by users. After receiving the text question input by the user, the system parses it through natural language processing (NLP) technology and searches for answers in the knowledge base. However, the current system does not have an effective user feedback mechanism, so it cannot know the satisfaction of users with the answers and self-learn and optimize the system according to the user feedback, that is, it lacks a closed-loop mechanism for self-improvement. Secondly, the update of the knowledge base for traditional knowledge question answering depends on manual operations and cannot be updated in real time. Finally, for complex or unsolved problems, these systems also lack an effective manual intervention mechanism, resulting in low answering efficiency and accuracy. Summary of the Invention

[0003] This application provides an update method, device, equipment and computer-readable storage medium for an intelligent question answering system, which can solve the technical problem of low answering efficiency and accuracy of the intelligent question answering system in the prior art.

[0004] In a first aspect, an embodiment of this application provides an update method for an intelligent question answering system. The update method for the intelligent question answering system includes:

[0005] Receiving the input information of the user through the intelligent question answering system and obtaining the corresponding question information;

[0006] Determining the type to which the question information belongs according to the question information, so as to obtain the corresponding answer and provide it to the user;

[0007] Obtaining the feedback information sent by the user for the answer;

[0008] If the feedback information is unresolved, then perform manual answering according to the manual mechanism to obtain the corresponding target answer;

[0009] Adding the target answer and the question information to the knowledge base to update the intelligent question answering system, where the intelligent question answering system includes a knowledge base.

[0010] Combined with the first aspect, in an implementation manner, the determining the type to which the question information belongs according to the question information, so as to obtain the corresponding answer and provide it to the user includes:

[0011] Based on the preset semantic analysis model and the problem information, determine the type to which the problem information belongs;

[0012] According to the type to which the problem information belongs, match the corresponding knowledge base;

[0013] According to the knowledge base, obtain the corresponding answer and provide it to the user.

[0014] Combined with the first aspect, in an implementation manner, the artificial answering according to the artificial mechanism to obtain the corresponding target answer includes:

[0015] Create a work group for the artificial mechanism to analyze the problem information;

[0016] Obtain the analysis result of the artificial person, and use the analysis result as the target answer corresponding to the problem information.

[0017] Combined with the first aspect, in an implementation manner, before adding the target answer and the problem information to the knowledge base to update the intelligent question-answering system, it further includes:

[0018] Obtain the feedback information sent by the user based on the target answer;

[0019] If the feedback information is that the problem has been solved, add the target answer and the problem information to the knowledge base to update the intelligent question-answering system.

[0020] Combined with the first aspect, in an implementation manner, the preset semantic analysis model is obtained by pre-training a preset neural network with a training data set to be trained, the training data set to be trained includes multiple groups of training data to be trained, and the training data to be trained includes problem information and corresponding types.

[0021] Combined with the first aspect, in an implementation manner, after obtaining the feedback information sent by the user for the answer, it further includes:

[0022] If the feedback information is to retry, send the answer to the user again.

[0023] Combined with the first aspect, in an implementation manner, the knowledge base includes: FAQ knowledge base, local knowledge base and Internet knowledge base.

[0024] Combined with the second aspect, in an implementation manner, the update device of the intelligent question-answering system includes:

[0025] The first acquisition module is used to receive the input information of the user through the intelligent question-answering system and acquire the corresponding problem information;

[0026] A second acquisition module, configured to determine the type to which the question information belongs according to the question information, so as to obtain a corresponding answer and provide it to the user;

[0027] A third acquisition module, configured to acquire feedback information sent by the user for the answer;

[0028] A fourth acquisition module, configured to, if the feedback information indicates that the problem is not solved, perform manual answering according to a manual mechanism to obtain a corresponding target answer;

[0029] An update module, configured to add the target answer and the question information to a knowledge base to update the intelligent question-answering system, where the intelligent question-answering system includes the knowledge base.

[0030] In a third aspect, an embodiment of the present application provides an update device for an intelligent question-answering system. The update device for the intelligent question-answering system includes a processor, a memory, and an intelligent question-answering system update program stored on the memory and executable by the processor. When the intelligent question-answering system update program is executed by the processor, the steps of the intelligent question-answering system update method as described above are implemented.

[0031] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which an intelligent question-answering system update program is stored. When the intelligent question-answering system update program is executed by a processor, the steps of the intelligent question-answering system update method as described above are implemented.

[0032] The beneficial effects brought by the technical solutions provided by the embodiments of the present application include:

[0033] By receiving the input information of the user through the intelligent question-answering system, obtaining the corresponding question information; determining the type to which the question information belongs according to the question information, so as to obtain a corresponding answer and provide it to the user; obtaining the feedback information sent by the user for the answer; if the feedback information indicates that the problem is not solved, performing manual answering according to a manual mechanism to obtain a corresponding target answer; adding the target answer and the question information to the knowledge base to update the intelligent question-answering system, where the intelligent question-answering system includes the knowledge base, the technical problem of low answering efficiency and accuracy of the intelligent question-answering system in the related art is solved, and the answering efficiency, accuracy, user experience, and overall performance of the intelligent question-answering system are significantly improved. Description of the Drawings

[0034] Figure 1 It is a schematic flowchart of the first embodiment of the intelligent question-answering system update method of the present application;

[0035] Figure 2 It is a schematic flowchart of the second embodiment of the intelligent question-answering system update method of the present application;

[0036] Figure 3 Schematic diagram of functional modules of an embodiment of the update device for the intelligent question - answering system of the present application;

[0037] Figure 4 Schematic diagram of the hardware structure of the update device of the intelligent question - answering system involved in the solution of the embodiment of the present application. Detailed implementation manners

[0038] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0039] First, some technical terms in the present application are explained to facilitate the understanding of the present application by those skilled in the art.

[0040] To make the purpose, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0041] In a first aspect, an embodiment of the present application provides an update method for an intelligent question - answering system.

[0042] In one embodiment, referring to Figure 1 , Figure 1 is the flowchart of the first embodiment of the update method for the intelligent question - answering system of the present application. As Figure 1 shown, the update method for the intelligent question - answering system includes:

[0043] Step S10: Receive the input information of the user through the intelligent question - answering system, and obtain the corresponding question information;

[0044] Exemplarily, the input information sent by the user is received through the intelligent question - answering system, and the output information includes text and / or pictures. The input information received by the user is recognized. If the input information is recognized as a picture, the OCR technology is used to recognize the text information in the picture, and then the recognized text information is processed; if the input information is recognized as text information, the natural language processing (NLP) module is called to parse and understand the user's text question, and the user's question is converted into processable text information; if the input information is recognized as text and pictures, the text and picture information are processed respectively, and the text recognized by OCR is integrated with the text input by the user to form a complete text information. This text information is the question information.

[0045] Step S20: Determine the type to which the problem information belongs according to the problem information, so as to obtain a corresponding answer and provide it to the user;

[0046] Exemplarily, after obtaining the problem information, analyze the problem information to determine the type to which the problem belongs. The system will first determine the category to which the problem belongs through a problem classification module according to the content of the user's question, and then match the corresponding knowledge base for answering. This step ensures that the system can select the best answer from a specific knowledge base, improving the pertinence of the answer.

[0047] Specifically, the step of determining the type to which the problem information belongs according to the problem information, so as to obtain a corresponding answer and provide it to the user includes: determining the type to which the problem information belongs based on a preset semantic analysis model and the problem information; matching a corresponding knowledge base according to the type to which the problem information belongs; and obtaining a corresponding answer according to the knowledge base and providing it to the user.

[0048] Exemplarily, input the problem information into a preset semantic analysis model to obtain the type output by the preset semantic analysis model based on the problem information. The preset semantic analysis model is obtained by pre-training a preset neural network with a training dataset to be trained. The training dataset to be trained includes multiple groups of training data, and each group of training data includes problem information and a corresponding type.

[0049] For example, train a preset neural network with a training dataset to be trained, and determine whether the trained preset neural network is in a converged state. If the trained preset neural network is in a converged state, then generate a preset semantic analysis model with the trained preset neural network; if the trained preset neural network is not in a converged state, then continue to train the trained preset neural network until the trained preset neural network is in a converged state. Determining whether the trained preset neural network is in a converged state includes obtaining the loss value of the trained preset neural network and / or the number of training times of the trained preset neural network, comparing the loss value of the trained preset neural network with a preset loss value. If the loss value of the trained preset neural network is less than or equal to the preset loss value, then determine that the trained preset neural network is in a converged state; if the loss value of the trained preset neural network is greater than the preset loss value, then determine that the trained preset neural network is not in a converged state.

[0050] Or, compare the number of training times of the trained preset neural network with a preset number of training times. If the number of training times of the trained preset neural network is greater than or equal to the preset number of training times, then determine that the trained preset neural network is in a converged state; if the number of training times of the trained preset neural network is less than the preset number of training times, then determine that the trained preset neural network is not in a converged state.

[0051] Alternatively, compare the loss value of the trained preset neural network with the preset loss value, and compare the number of training times of the trained preset neural network with the preset number of training times. If the loss value of the trained preset neural network is less than or equal to the preset loss value, and the number of training times of the trained preset neural network is greater than or equal to the preset number of training times, it is determined that the trained preset neural network is in a converged state; if the loss value of the trained preset neural network is less than or equal to the preset loss value, and the number of training times of the trained preset neural network is less than the preset number of training times, it is determined that the trained preset neural network is not in a converged state; or, if the loss value of the trained preset neural network is greater than the preset loss value, and the number of training times of the trained preset neural network is greater than or equal to the preset number of training times, it is determined that the trained preset neural network is not in a converged state.

[0052] By determining the type to which the question information belongs, match the corresponding knowledge base, where the knowledge base includes: FAQ knowledge base, local knowledge base, and Internet knowledge base. The FAQ knowledge base is a list of Q&A pairs consisting of one or more similar questions and one answer. The source of the FAQ knowledge base is currently mainly manual entry, and the administrator enters and updates the confirmed accurate answers. Prioritizing the search of the FAQ knowledge base can improve the efficiency of the system and the accuracy of the answers. There is no need to deeply search the vast knowledge base or the Internet. This method can return the matching answer more quickly and improve the system response speed. The answers in the FAQ knowledge base are pre-entered standardized answers, ensuring the consistency and accuracy of the answers, and can also avoid the uncertainty that may occur when the model organizes language and integrates answers by itself. Additionally, if a matching answer is obtained from the search of the FAQ knowledge base, the system can even answer exactly according to the entered answer. These entered answers are in markdown text format and can include pictures. However, when the model searches the knowledge base, it cannot learn and process the pictures in the document, so the answers given by the model will not include pictures. The answers provided by the FAQ knowledge base are more rich and intuitive, further improving the user experience.

[0053] Step S30: Obtain the feedback information sent by the user for the answer;

[0054] Exemplarily, the feedback information includes solved, unsolved, and retry.

[0055] Step S40: If the feedback information is unsolved, perform manual answering according to the manual mechanism to obtain the corresponding target answer;

[0056] Exemplarily, when encountering complex or unsolved problems, the system automatically notifies the administrator to intervene. Through the collaboration of human and intelligence, it ensures that all problems can be effectively solved: Administrator intervention: When the problem is marked as unsolved, the system automatically notifies the relevant administrator for manual intervention. Manual processing: The administrator processes the problem and enters the standard answer into the knowledge base to improve the accuracy of the system's answers. When the problem is marked as "unsolved" and requires manual intervention, the system not only notifies the administrator but also automatically creates a discussion group including the relevant responsible persons to speed up the problem-solving process and avoid repeated communication.

[0057] Step S50: Add the target answer and the question information to the knowledge base to update the intelligent question-answering system, where the intelligent question-answering system includes a knowledge base.

[0058] Exemplarily, add the target answer and the question information to the knowledge base to update the intelligent question-answering system, where the intelligent question-answering system includes: FAQ knowledge base, local knowledge base, and Internet knowledge base.

[0059] In this embodiment, through the intelligent question-answering system, the input information of the user is received, and the corresponding question information is obtained, where the input information includes text and / or pictures; according to the question information, the type to which the question information belongs is determined to obtain the corresponding answer; the feedback information sent by the user is obtained; if the feedback information is unsolved, then according to the manual mechanism, a manual answer is given to obtain the corresponding target answer; the target answer and the question information are added to the knowledge base to update the intelligent question-answering system, where the intelligent question-answering system includes a knowledge base, which solves the technical problems of low answer efficiency and accuracy in the related intelligent question-answering system, and significantly improves the answer efficiency, accuracy, user experience, and overall performance of the intelligent question-answering system.

[0060] In one embodiment, refer to Figure 2 , Figure 2 is a schematic flowchart of the second embodiment of the update method of the intelligent question-answering system of this application. As Figure 2 shown, the update method of the intelligent question-answering system includes:

[0061] Step S11: Receive the input information of the user through the intelligent question-answering system and obtain the corresponding question information;

[0062] Exemplarily, input information sent by the user is received through an intelligent question-and-answer system, and the output information includes text and / or pictures. The input information received by the user is recognized. When it is recognized that the input information is a picture, OCR technology is used to recognize the text information in the picture, and then it is processed in combination with the recognized text information; when it is recognized that the input information is text information, the natural language processing (NLP) module is called to parse and understand the user's text question, and the user's question is converted into processable text information; when it is recognized that the input information is text and pictures, the text and picture information are processed respectively, and the text recognized by OCR is integrated with the text input by the user to form a complete text information. This text information is question information.

[0063] Step S12: According to the question information, determine the type to which the question information belongs, so as to obtain a corresponding answer and provide it to the user;

[0064] Exemplarily, after obtaining the question information, the question information is analyzed to determine the type to which the question belongs. The system will first determine the category to which the question belongs through a question classification module according to the content of the user's question, and then match the corresponding knowledge base for answering. This step ensures that the system can select the best answer from a specific knowledge base and improves the pertinence of the answer.

[0065] Specifically, the determining the type to which the question information belongs according to the question information to obtain a corresponding answer includes: determining the type to which the question information belongs based on a preset semantic analysis model and the question information; matching a corresponding knowledge base according to the type to which the question information belongs; and obtaining a corresponding answer according to the knowledge base.

[0066] Exemplarily, the question information is input into a preset semantic analysis model to obtain the type output by the preset semantic analysis model based on the question information. The preset semantic analysis model is obtained by pre-training a preset neural network through a training dataset to be trained. The training dataset to be trained includes multiple groups of training data, and each group of training data includes question information and a corresponding type.

[0067] For example, a preset neural network is trained using a dataset to be trained, and it is determined whether the trained preset neural network is in a convergent state. If the trained preset neural network is in a convergent state, the trained preset neural network is used to generate a preset semantic analysis model; if the trained preset neural network is not in a convergent state, the trained preset neural network is continuously trained until the trained preset neural network is in a convergent state. Determining whether the trained preset neural network is in a convergent state includes obtaining the loss value of the trained preset neural network and / or the number of training times of the trained preset neural network, and comparing the loss value of the trained preset neural network with a preset loss value. If the loss value of the trained preset neural network is less than or equal to the preset loss value, it is determined that the trained preset neural network is in a convergent state; if the loss value of the trained preset neural network is greater than the preset loss value, it is determined that the trained preset neural network is not in a convergent state.

[0068] Alternatively, the number of training times of the trained preset neural network is compared with a preset number of training times. If the number of training times of the trained preset neural network is greater than or equal to the preset number of training times, it is determined that the trained preset neural network is in a convergent state; if the number of training times of the trained preset neural network is less than the preset number of training times, it is determined that the trained preset neural network is not in a convergent state.

[0069] Alternatively, the loss value of the trained preset neural network is compared with a preset loss value, and the number of training times of the trained preset neural network is compared with a preset number of training times. If the loss value of the trained preset neural network is less than or equal to the preset loss value, and the number of training times of the trained preset neural network is greater than or equal to the preset number of training times, it is determined that the trained preset neural network is in a convergent state; if the loss value of the trained preset neural network is less than or equal to the preset loss value, and the number of training times of the trained preset neural network is less than the preset number of training times, it is determined that the trained preset neural network is not in a convergent state; or, if the loss value of the trained preset neural network is greater than the preset loss value, and the number of training times of the trained preset neural network is greater than or equal to the preset number of training times, it is determined that the trained preset neural network is not in a convergent state.

[0070] By determining the type to which the question information belongs, the corresponding knowledge base is matched. The knowledge base includes: FAQ knowledge base, local knowledge base, and Internet knowledge base. The FAQ knowledge base is a list of Q&A pairs composed of one or more similar questions and one answer. The source of the FAQ knowledge base is currently mainly manual input. The administrator inputs and updates the confirmed accurate answers. Prioritizing the search of the FAQ knowledge base can improve the efficiency of the system and the accuracy of the answers. There is no need to deeply search the huge knowledge base or the Internet. This way can return the matching answer more quickly and improve the system response speed. The answers in the FAQ knowledge base are pre-entered standardized answers, ensuring the consistency and accuracy of the answers, and can also avoid the uncertainty that may occur when the model organizes language and integrates answers by itself. In addition, if a matching answer is obtained by searching the FAQ knowledge base, the system can even answer exactly according to the entered answer. These entered answers are in markdown text format and can include pictures. When the model searches the knowledge base, it cannot learn and process the pictures in the document, so the answers given by the model will not include pictures. The answers provided by the FAQ knowledge base are richer and more intuitive, further improving the user experience.

[0071] Step S13: Obtain the feedback information sent by the user for the answer;

[0072] Exemplarily, the feedback information includes solved, unsolved, and retry.

[0073] Step S14: If the feedback information is unsolved, perform manual answering according to the manual mechanism to obtain the corresponding target answer;

[0074] Exemplarily, when encountering complex or unsolved problems, the system automatically notifies the administrator to intervene, and through the cooperation of manual and intelligent, ensures that all problems can be effectively solved: Administrator intervention: When the problem is marked as unsolved, the system automatically notifies the relevant administrator for manual intervention. Manual processing: The administrator processes the problem and enters the standard answer into the knowledge base to improve the accuracy of the system answer. When the problem is marked as "unsolved" and requires manual intervention, the system not only notifies the administrator, but also automatically creates a discussion group including the relevant person in charge to speed up the problem processing speed and avoid repeated communication.

[0075] Step S15: Obtain the feedback information sent by the user based on the target answer;

[0076] Step S16: If the feedback information is solved, add the target answer and the question information to the knowledge base to update the intelligent Q&A system

[0077] Exemplarily, the target answer and the question information are added to the knowledge base to update the intelligent question-answering system, where the intelligent question-answering system includes: a FAQ knowledge base, a local knowledge base, and an Internet knowledge base.

[0078] In this embodiment, the input information of the user is received through the intelligent question-answering system, and the corresponding question information is obtained, where the input information includes text and / or pictures; according to the question information, the type to which the question information belongs is determined to obtain the corresponding answer; the feedback information sent by the user is obtained; if the feedback information is unresolved, manual answering is performed according to the manual mechanism to obtain the corresponding target answer; the target answer and the question information are added to the knowledge base to update the intelligent question-answering system, where the intelligent question-answering system includes a knowledge base, solving the technical problem of low answering efficiency and accuracy in the related art of the intelligent question-answering system, and significantly improving the answering efficiency, accuracy, user experience, and overall performance of the intelligent question-answering system.

[0079] In a second aspect, an embodiment of the present application further provides an updating device for an intelligent question-answering system.

[0080] In one embodiment, referring to Figure 3 , Figure 3 is a schematic diagram of the functional modules of an embodiment of the updating device for the intelligent question-answering system of the present application. As Figure 3 shown, the updating device for the intelligent question-answering system includes:

[0081] The first acquisition module 10 is configured to receive the input information of the user through the intelligent question-answering system and obtain the corresponding question information;

[0082] The second acquisition module 20 is configured to determine the type to which the question information belongs according to the question information to obtain the corresponding answer and provide it to the user;

[0083] The third acquisition module 30 is configured to obtain the feedback information sent by the user for the answer;

[0084] The fourth acquisition module 40 is configured to, if the feedback information is unresolved, perform manual answering according to the manual mechanism to obtain the corresponding target answer;

[0085] The update module 50 is configured to add the target answer and the question information to the knowledge base to update the intelligent question-answering system, where the intelligent question-answering system includes a knowledge base.

[0086] Further, in one embodiment, the second acquisition module 20 is configured to:

[0087] Based on a preset semantic analysis model and the question information, determine the type to which the question information belongs;

[0088] Match the corresponding knowledge base according to the type to which the problem information belongs;

[0089] According to the knowledge base, obtain the corresponding answer and provide it to the user.

[0090] Further, in one embodiment, the fourth acquisition module 40 is configured to:

[0091] Create a work group for manually analyzing the problem information;

[0092] Obtain the analysis result of the manual, and use the analysis result as the target answer corresponding to the problem information.

[0093] Further, in one embodiment, the updating device of the intelligent question-answering system further includes a new module for:

[0094] Obtain the feedback information sent by the user based on the target answer;

[0095] If the feedback information is that the problem has been solved, add the target answer and the problem information to the knowledge base to update the intelligent question-answering system.

[0096] Wherein, the functions of each module in the updating device of the intelligent question-answering system correspond to the steps in the embodiment of the updating method of the intelligent question-answering system, and their functions and implementation processes will not be described in detail here.

[0097] In a third aspect, an embodiment of the present application provides an updating device for an intelligent question-answering system. The updating device for the intelligent question-answering system may be a device with data processing functions such as a personal computer (PC), a laptop computer, a server, etc.

[0098] Refer to Figure 4 , Figure 4 This is a schematic diagram of the hardware structure of the updating device for the intelligent question-answering system involved in the embodiment of the present application. In the embodiment of the present application, the updating device for the intelligent question-answering system may include a processor, a memory, a communication interface, and a communication bus.

[0099] Among them, the communication bus can be of any type and is used to interconnect the processor, the memory, and the communication interface.

[0100] The communication interface includes interfaces such as input / output (I / O) interfaces, physical interfaces, and logical interfaces, which are used to implement the interconnection of components inside the device for updating the intelligent question-answering system, as well as interfaces for implementing the interconnection between the device for updating the intelligent question-answering system and other devices (such as other computing devices or user devices). The physical interface can be an Ethernet interface, a fiber optic interface, an ATM interface, etc.; the user device can be a display, a keyboard, etc.

[0101] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical memory, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0102] The processor can be a general-purpose processor, which can call the update program of the intelligent question-answering system stored in the memory and execute the update method of the intelligent question-answering system provided in the embodiments of the present application. For example, the general-purpose processor can be a central processing unit (CPU). Among them, the method executed when the update program of the intelligent question-answering system is called can refer to the various embodiments of the update method of the intelligent question-answering system of the present application, which will not be elaborated here.

[0103] Those skilled in the art can understand that Figure 4 the hardware structure shown in

[0104] does not constitute a limitation to the present application, and may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0105] The computer-readable storage medium of the present application stores an update program of the intelligent question-answering system. When the update program of the intelligent question-answering system is executed by a processor, the steps of the update method of the intelligent question-answering system as described above are implemented.

[0106] Among them, the method implemented when the update program of the intelligent question-answering system is executed can refer to the various embodiments of the update method of the intelligent question-answering system of the present application, which will not be elaborated here.

[0107] It should be noted that the serial numbers of the embodiments of the present application above are only for description and do not represent the superiority or inferiority of the embodiments.

[0108] The terms "including" and "having" and any variations thereof in the specification, claims and drawings of the present application are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products or devices. The descriptions of "first", "second", "third", etc. are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit that "first", "second" and "third" are of different types.

[0109] In the description of the embodiments of the present application, "exemplary", "for example" or "for instance" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary", "for example" or "for instance" in the embodiments of the present application should not be construed as more preferred or more advantageous than other embodiments or designs. Rather, the use of the words "exemplary", "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0110] In the description of the embodiments of the present application, unless otherwise specified, " / " means "or". For example, A / B may mean A or B; "and / or" in the text is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B may mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present application, "a plurality of" means two or more than two.

[0111] In some processes described in the embodiments of the present application, a plurality of operations or steps appear in a specific order. However, it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or may be executed in parallel. The serial numbers of the operations are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in order or in parallel, and these operations or steps may be combined.

[0112] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium as described above (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal device to execute the methods described in various embodiments of the present application.

[0113] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A method for updating an intelligent question-answering system, characterized in that: The updating method of the intelligent question answering system comprises: Receive user input information through the intelligent question-answering system and obtain corresponding question information; According to the question information, determine the type of the question information to which it belongs, so as to obtain a corresponding answer and provide it to the user; Obtaining feedback information sent by the user in response to the answer; If the feedback information is unresolved, a manual solution is performed according to a manual mechanism to obtain a corresponding target answer; The target answer and the question information are added to a knowledge base to update the intelligent question-answering system, wherein the intelligent question-answering system includes a knowledge base.

2. The updating method of the intelligent question answering system according to claim 1, characterized in that: The step of determining the type of the question information according to the question information to obtain a corresponding answer and provide it to the user includes: Determine the type of the question information based on a preset semantic analysis model and the question information; According to the type of the question information, matching the corresponding knowledge base; According to the knowledge base, corresponding answers are obtained and provided to the user.

3. The updating method of the intelligent question answering system according to claim 1, characterized in that: The manual answering according to the manual mechanism to obtain the corresponding target answer includes: Creating a work group for manual mechanism to analyze the problem information; The manual analysis result is obtained, and the analysis result is used as the target answer corresponding to the question information.

4. The updating method of the intelligent question-answering system according to claim 1, characterized in that: Before adding the target answer and the question information to the knowledge base to update the intelligent question answering system, the method further includes: Acquiring feedback information sent by the user based on the target answer; If the feedback information is resolved, the target answer and the question information are added to the knowledge base to update the intelligent question-answering system.

5. The updating method of the intelligent question answering system according to claim 2, characterized in that: The preset semantic analysis model is obtained by pre-training a preset neural network with a data set to be trained, wherein the data set to be trained includes multiple groups of data to be trained, and the data to be trained includes question information and corresponding types.

6. The updating method of the intelligent question answering system according to claim 2, characterized in that: After obtaining the feedback information sent by the user in response to the answer, the method further includes: If the feedback information is to retry, the answer is sent to the user again.

7. The updating method of the intelligent question answering system according to claim 2, characterized in that: The knowledge base includes: FAQ knowledge base, local knowledge base and Internet knowledge base.

8. An updating device for an intelligent question-answering system, characterized in that: The updating device of the intelligent question-answering system comprises: The first acquisition module is used to receive the user's input information through the intelligent question-answering system and obtain the corresponding question information; A second acquisition module is used to determine the type of the question information according to the question information, so as to obtain a corresponding answer and provide it to the user; A third acquisition module is used to acquire feedback information sent by the user in response to the answer; A fourth acquisition module is used to perform manual answering according to a manual mechanism to obtain a corresponding target answer if the feedback information is unresolved; An updating module is used to add the target answer and the question information to a knowledge base to update the intelligent question-answering system, wherein the intelligent question-answering system includes a knowledge base.

9. An updating device for an intelligent question-answering system, characterized in that: The update device of the intelligent question and answer system includes a processor, a memory, and an update program of the intelligent question and answer system stored in the memory and executable by the processor, wherein when the update program of the intelligent question and answer system is executed by the processor, the steps of the update method of the intelligent question and answer system as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an update program for the intelligent question and answer system, wherein when the update program for the intelligent question and answer system is executed by the processor, the steps of the update method for the intelligent question and answer system according to any one of claims 1 to 7 are implemented.

Images