Knowledge question and answer method, device and equipment, medium and program product

Abstract

The embodiment of the invention discloses a knowledge question answering method and device, equipment, a medium and a program product. The method comprises the following steps: displaying at least one first tag in a knowledge question-answer interface; receiving a target label selected for the question information; and outputting answer information corresponding to the question information in the knowledge question and answer interface. According to the embodiment of the invention, the specific knowledge content in the knowledge base is indexed through the tag, so that the answer personalization is improved and the knowledge question-answering efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and specifically to a knowledge question answering method, a knowledge question answering device, a computer device, a computer-readable storage medium, and a computer program product. Background Art

[0002] Knowledge question answering refers to the process of answering questions raised by users in accurate and concise natural language. It is a direction in the field of natural language processing (NLP) that has attracted much attention and has broad development prospects.

[0003] At present, the existing mainstream knowledge question answering solutions use the entire knowledge base as the information source for the answer when answering questions raised by users. However, as the knowledge content in the knowledge base continues to increase, when generating answers to questions raised by users, it is necessary to find specific information from the massive knowledge content in the knowledge base to answer, which not only faces the problem of information overload, but also reduces the efficiency of knowledge question answering. Summary of the invention

[0004] The embodiments of the present application provide a knowledge question and answer method, apparatus, device, medium and program product, which improve the personalization of answers and the efficiency of knowledge question and answer by indexing specific knowledge content in a knowledge base through tags.

[0005] On the one hand, an embodiment of the present application provides a knowledge question answering method, the method comprising:

[0006] Displaying at least one first tag in the knowledge question and answer interface, each first tag is used to index at least one knowledge content from the knowledge base;

[0007] receiving a target tag selected for the question information, the target tag being any one of the at least one first tag;

[0008] The answer information corresponding to the question information is output in the knowledge question and answer interface, and the answer information is generated based on the question information and the knowledge content indexed by the target tag.

[0009] On the other hand, an embodiment of the present application provides a knowledge question answering device, the device comprising:

[0010] A display unit, used to display at least one first tag in the knowledge question and answer interface, each first tag is used to index at least one knowledge content from the knowledge base;

[0011] a processing unit, configured to receive a target tag selected for the question information, the target tag being any one of the at least one first tag;

[0012] The processing unit is also used to output answer information corresponding to the question information in the knowledge question and answer interface, and the answer information is generated based on the question information and the knowledge content indexed by the target tag.

[0013] In one implementation, the display unit, when used to display at least one first label in the knowledge question and answer interface, is specifically used to:

[0014] In response to a tag recall operation in the knowledge question and answer interface, a tag filter window is displayed; the tag recall operation includes any of the following: an input operation for a shortcut key, a trigger operation for a tag component, and a specified gesture operation;

[0015] At least one first tag is displayed in the tag filter window.

[0016] In one implementation, the processing unit is further configured to:

[0017] Receive input information edited in the knowledge question and answer interface;

[0018] In the tag filtering window, at least one first tag is updated and displayed as at least one second tag related to the input information; the second tag belongs to the at least one first tag.

[0019] In one implementation, the method is applied to a knowledge question and answer application, the knowledge question and answer application corresponds to a knowledge base interaction system, the knowledge base interaction system includes at least one index cluster and at least one label storage shard; the index cluster stores an address that can be routed to a sub-shard in the label storage shard, and one sub-shard is used to store a first label of knowledge content in the knowledge base corresponding to an object identifier for logging into the knowledge question and answer application;

[0020] The processing unit is also used to:

[0021] Routing to a target index cluster in the knowledge base interaction system according to the object identifier, the target index cluster storing an address that can be routed to at least one reference sub-shard, the at least one reference sub-shard being a sub-shard storing a first label of knowledge content in the knowledge base corresponding to the object identifier;

[0022] According to the input information and the address of each reference sub-slice, a first tag related to the input information is obtained; and the first tag related to the input information is used as the second tag.

[0023] In one implementation, the knowledge question and answer interface includes at least two information source components, each of which corresponds to an information source, and the information source includes: the Internet and at least one knowledge base; the processing unit is further used to:

[0024] In response to a selection operation on the first information source component, displaying the first information source component as a selected state;

[0025] When the information source corresponding to the first information source component is a knowledge base, the step of displaying at least one first tag in the knowledge question and answer interface is triggered.

[0026] In one implementation, a round of knowledge question and answer includes at least one question and answer dialogue, and one question and answer dialogue includes a question message and an answer message, the question message includes question information, and the answer message includes answer information; during the first question and answer dialogue, the first information source component is in a selected state, and the processing unit is further used to:

[0027] After the first question-and-answer dialogue ends, in response to an information source switching operation performed in the knowledge question-and-answer interface, the first information source component is displayed as being in an unselected state; and,

[0028] The selected second information source component is displayed as a selected state, and the question information in the second question-and-answer dialogue is edited in the knowledge question-and-answer interface; the second information source component is different from the first information source component.

[0029] In one implementation, the selected information source component is represented as a target information source component, and when the number of the selected target information source components is at least two, the first label is a label defined in the information source corresponding to the at least two target information source components; the processing unit is further configured to:

[0030] According to the priority setting operation for at least two target information source components, a priority identifier is displayed at an adjacent display position of each target information source component; the priority identifier is used to represent: the priority level of the knowledge content indexed by the target tag in the information source corresponding to the target information source component as a reference for generating answer information.

[0031] In one implementation, the knowledge question and answer interface includes a question and answer area, the question and answer area includes a message display sub-area, the message display sub-area is used to display question messages and answer messages generated by a round of knowledge question and answer, and the processing unit is further used to:

[0032] When a target tag included in the question message in the message display sub-area is triggered, displaying a content identifier of at least one knowledge content indexed by the target tag;

[0033] In response to a triggering operation on a target content identifier, the knowledge content corresponding to the target content identifier is displayed; the target content identifier is any one of the content identifiers of at least one knowledge content.

[0034] In one implementation, the knowledge question and answer interface further includes an information display area, which is used to display knowledge content in the knowledge base. The processing unit is also used to:

[0035] According to a hiding operation on the question and answer area, the question and answer area is hidden in the knowledge question and answer interface; the hiding operation includes any of the following: an operation in which a target tag included in the question message is triggered, and an operation in which a hidden component in the question and answer area is triggered;

[0036] Enlarge the information display area.

[0037] In one implementation, the question-and-answer area further includes a dialogue sub-area, which is used to receive question information and target tags; the knowledge question-and-answer interface includes a question-and-answer display component, a processing unit, and is further used to:

[0038] In response to the triggering operation of the question and answer display component, the question and answer area is displayed in the knowledge question and answer interface; the content displayed in the message display sub-area and the dialogue sub-area of ​​the question and answer area is the same as before the question and answer area is hidden.

[0039] In one implementation, the information display area includes a search box and a processing unit, and is further configured to:

[0040] According to the third tag input in the search box, the content identifier of the knowledge content defined with the third tag in the knowledge base is displayed in the information display area.

[0041] In one implementation, the answer information is composed of at least one sub-information; the processing unit is further configured to:

[0042] Displaying an information source entrance at a display position adjacent to a target sub-information, the target sub-information being any one of the at least one sub-information;

[0043] In response to a trigger operation on an information source entry, displaying target knowledge content as a reference for generating target sub-information;

[0044] In the target knowledge content, the fragmented content used as a reference for generating the target sub-information is annotated and displayed.

[0045] In one implementation, the method is applied to a knowledge question answering application; the processing unit is further used to:

[0046] In response to a login operation for an object identifier in a knowledge question and answer application, a knowledge service interface is displayed; the knowledge service interface includes a question and answer trigger component;

[0047] When the question-and-answer trigger component is triggered, the knowledge question-and-answer interface is displayed.

[0048] In one implementation, the processing unit is further configured to:

[0049] Filter one or more knowledge contents with defined target tags from the knowledge base, where each knowledge content includes multiple fragment contents;

[0050] The matching degree of the question information and each fragment content included in each knowledge content is calculated to obtain the matching degree result corresponding to each fragment content included in each knowledge content; the matching degree result is used to indicate: the reliability of the fragment content as a reference for generating answer information corresponding to the question information;

[0051] Based on the target segment content whose matching result is greater than the matching threshold, answer information corresponding to the question information is generated.

[0052] On the other hand, an embodiment of the present application provides a computer device, the computer device comprising:

[0053] a processor adapted to execute a computer program;

[0054] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the knowledge question and answer method as described above is implemented.

[0055] On the other hand, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and the computer program is suitable for being loaded by a processor and executing the knowledge question and answer method as described above.

[0056] On the other hand, an embodiment of the present application provides a computer program product, which includes a computer program stored in a computer-readable storage medium. A processor of a computer device reads the computer program from the computer-readable storage medium, and the processor executes the computer program, so that the computer device performs the above-mentioned knowledge question answering method.

[0057] In the embodiment of the present application, at least one first label is displayed in the knowledge question and answer interface, and each first label can index at least one knowledge content in the knowledge base; the first label is pre-defined by the user for the knowledge content in the knowledge base, such as defining the first label 1 and the first label 2 for the knowledge content 1, and defining the first label 1 for the knowledge content 2. In the process of editing the question information in the knowledge question and answer interface, it supports the user to select the target label for the edited question information from at least one first label. In this way, after receiving the target label and the question information, the computer device can quickly filter the knowledge content indexed by the target label from the knowledge base through the label index, and generate the answer information based on the knowledge content and the question information indexed by the target label. It can be seen that, compared with the generation of answer information for question information based on the massive knowledge content in the knowledge base, the embodiment of the present application supports the label index to quickly locate the specific knowledge content in the knowledge base as the information source of the answer information, avoiding problems such as equipment failure caused by information overload. On the other hand, the target label is selected by the user in a personalized manner, so that when different users select different target labels for the same question, the generated answer information is also different, which meets the user's requirements for personalized generation of answer information and improves the user experience. On the other hand, answer information only needs to be generated based on the knowledge content indexed by the target tag in the knowledge base, which greatly improves the speed and efficiency of answer generation, thereby improving the efficiency of knowledge question and answering and helping users improve work efficiency and personal learning effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1a It is a schematic diagram of the architecture of a knowledge question-answering scenario provided by an exemplary embodiment of the present application;

[0060] Figure 1b It is a schematic diagram of the architecture of another knowledge question-and-answer scenario provided by an exemplary embodiment of the present application;

[0061] Figure 2 It is a flowchart of a knowledge question answering method provided by an exemplary embodiment of the present application;

[0062] Figure 3a is a schematic diagram of knowledge content in a first tag index knowledge base provided by an exemplary embodiment of the present application;

[0063] Figure 3bIt is a schematic diagram of a login knowledge quiz application provided by an exemplary embodiment of the present application;

[0064] Figure 3c is a schematic diagram of another login knowledge quiz application provided by an exemplary embodiment of the present application;

[0065] Figure 3d is a schematic diagram of another login knowledge quiz application provided by an exemplary embodiment of the present application;

[0066] Figure 4a is a schematic diagram of a knowledge question and answer interface provided by an exemplary embodiment of the present application;

[0067] Figure 4b is a schematic diagram of another knowledge question and answer interface provided by an exemplary embodiment of the present application;

[0068] Figure 5a It is a schematic diagram of a first label directly displayed on the knowledge question and answer interface by default provided by an exemplary embodiment of the present application;

[0069] Figure 5b It is a schematic diagram of another first label directly displayed on the knowledge question and answer interface by default provided by an exemplary embodiment of the present application;

[0070] Figure 6a It is a schematic diagram of displaying a first label only when a label recall operation is performed, provided by an exemplary embodiment of the present application;

[0071] Figure 6b It is a schematic diagram of displaying a first label only when a label recall operation is performed, provided by an exemplary embodiment of the present application;

[0072] Figure 6c It is a schematic diagram of displaying a first label only when a label recall operation is performed, provided by an exemplary embodiment of the present application;

[0073] Figure 7 is a schematic diagram of selecting a first tag and deleting the selected first tag provided by an exemplary embodiment of the present application;

[0074] Figure 8a is a schematic diagram of a label screening provided by an exemplary embodiment of the present application;

[0075] Figure 8b is a schematic diagram of establishing an inverted index for a first tag provided by an exemplary embodiment of the present application;

[0076] Figure 8c It is a schematic diagram of the architecture of a knowledge base interaction system provided by an exemplary embodiment of the present application;

[0077] Fig. 9 It is a schematic diagram of generating answer information corresponding to question information based on a target tag provided by an exemplary embodiment of the present application;

[0078] Fig.10 is a flowchart of another knowledge question answering method provided by an exemplary embodiment of the present application;

[0079] Fig.11a It is a schematic diagram of setting a first label for a knowledge content in a knowledge base provided by an exemplary embodiment of the present application;

[0080] Fig.11b It is a schematic diagram of an object selecting multiple information sources for question information provided by an exemplary embodiment of the present application;

[0081] Fig.12 It is a schematic diagram of setting priorities for information sources by an object provided by an exemplary embodiment of the present application;

[0082] Fig.13 It is a schematic diagram of switching information sources during a round of knowledge question answering provided by an exemplary embodiment of the present application;

[0083] Fig.14 It is a schematic diagram of quickly locating knowledge content by triggering a target tag in a question message provided by an exemplary embodiment of the present application;

[0084] Fig.15 It is a schematic diagram of hiding or restoring the question and answer area in a knowledge question and answer interface provided by an exemplary embodiment of the present application;

[0085] Fig.16 It is a schematic diagram of a method of quickly searching for knowledge content indexed by a third tag through a search box in an information display area provided by an exemplary embodiment of the present application;

[0086] Fig.17 It is a schematic diagram of a method of quickly locating knowledge content and fragmented content based on an information source portal provided by an exemplary embodiment of the present application;

[0087] Fig.18 It is a schematic diagram of the overall process of a knowledge question answering method provided by an exemplary embodiment of the present application;

[0088] Fig.19 is a schematic diagram of the structure of a knowledge question-answering device provided by an exemplary embodiment of the present application;

[0089] Fig. 20 It is a structural diagram of a computer device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0090] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0091] The embodiment of the present application provides a knowledge question answering solution, specifically a solution for implementing knowledge question answering by indexing specified knowledge content in a knowledge base through tags; the following is a brief introduction to the technical terms and related concepts involved in the knowledge question answering solution provided in the embodiment of the present application, wherein:

[0092] 1. Knowledge quiz.

[0093] Knowledge question answering is also known as intelligent question answering (QA), open-ended question answering or interactive dialogue, etc., and belongs to the field of human-computer interaction. Knowledge question answering is an advanced form of information retrieval system, which mainly receives questions raised by users through computer devices, generates answers to the questions corresponding to the questions, and uses accurate and concise natural language to answer the questions raised by users. With the rapid development and application of artificial intelligence (AI), knowledge question answering technology has become a direction in the field of NLP that has attracted much attention and has broad development prospects.

[0094] In the knowledge question and answer scenario, a knowledge question and answer application is usually involved. The knowledge question and answer application is an application with an interactive dialogue function; in the embodiment of the present application, the knowledge question and answer application is taken as an example of an intelligent assistant. Among them, the intelligent assistant, or intelligent assistant application, refers to an intelligent software deployed in the terminal that can interact with the user through voice, text, and images, and realize rich functions such as information query, life server, and learning guidance. The intelligent assistant can be deployed in the terminal as an application; the application refers to a computer program for completing one or more specific tasks. According to the classification of the operation mode of the application, the application may include but is not limited to: ① The client that needs to download the installation package and deploy the installation package in the terminal to realize intelligent question and answer by running the installation package. ② There is no need to download the installation package, but a small program that runs on the client as a subroutine. And, ③ a we b (World Wide Web, Global Wide Area Network) application opened and run by the browser in the terminal; and so on. The embodiments of the present application do not limit the operation mode of the intelligent assistant in the terminal; in the implementation of the present application, the intelligent assistant is taken as an example of a client deployed and running on the PC side, that is, the intelligent assistant can be understood as a PC (Personal Computer) side intelligent workbench (such as a client with knowledge question and answer capabilities installed and run on the PC side - a browser), which aims to use artificial intelligence technology to improve the efficiency of retrieving information, browsing information and precipitating information in the knowledge question and answer process.

[0095] The process of implementing human-computer dialogue / knowledge question and answer in the knowledge question and answer scenario may include: the user inputs a question message in the intelligent assistant. After receiving the question message, the intelligent assistant sends the question message to the background system, and the background system processes the question indicated by the question message to obtain an answer message corresponding to the question message, which is a message that can answer the question indicated by the question message. The background system returns the answer message to the intelligent assistant, so that the intelligent assistant uses accurate and concise natural language to answer and output the answer message to solve the question raised by the user. Among them, the process of processing the question message mainly includes: question understanding, knowledge retrieval and answer generation; question understanding mainly includes the use of question classification and keyword extraction and other technologies, aiming to make the computer device understand the semantics of the question to be answered input by the user; knowledge retrieval mainly retrieves the knowledge content matching the question to be answered input by the user from the knowledge base management system corresponding to the intelligent assistant; answer generation mainly includes answer extraction and answer verification, aiming to generate corresponding answers for the question to be answered from the knowledge content retrieved from the knowledge.

[0096] 2. Knowledge base.

[0097] The knowledge base refers to the personal knowledge base in the smart assistant. In other words, the knowledge base can be understood as a collection of knowledge content created by users on the smart assistant platform; users add knowledge content to the knowledge content collection through operations such as collection, upload or download to build a dedicated personal knowledge base. Among them, the knowledge content in the knowledge base may include but is not limited to: multimedia resources in various formats such as documents, files, pictures, web pages, audio, video and animation. The embodiment of this application does not limit the number and format of the knowledge content in the knowledge base.

[0098] In the embodiment of the present application, the knowledge base management system supports the management of knowledge bases created by multiple users who register on the intelligent assistant platform; the management here may include but is not limited to: adding / deleting / modifying the knowledge content in the knowledge base, as well as querying or retrieving the knowledge base or the knowledge content in the knowledge base. For example, as mentioned above, in the process of conducting knowledge documents in the knowledge question and answer scenario, knowledge retrieval is involved, and the knowledge base management system can use the user's knowledge base as an information source for knowledge retrieval (or called an information source, that is, as a reference for answer generation); that is, when retrieving knowledge content that matches the question to be answered, the knowledge base management system can use the knowledge content in the user's knowledge base as the source of the answer corresponding to the question.

[0099] In practical applications, knowledge question-answering systems have problems such as slow question-answering efficiency, serious templateization, and poor question-answering quality. For example, when performing knowledge retrieval, the entire knowledge base is used as the information source. When retrieving knowledge points related to the question from the massive knowledge content in the knowledge base, on the one hand, the correlation between the searched knowledge points and the question will decrease, thereby reducing the quality of question-answering; on the other hand, as the knowledge content in the knowledge base increases, knowledge retrieval may have problems such as information overload, resulting in slow question-answering efficiency. For another example, when different users ask the same question in the same knowledge base, the knowledge points retrieved from the knowledge base by the computer device according to the same retrieval logic are also the same or similar, so that the answers to different users' questions are the same or similar, resulting in serious answer templateization.

[0100] In order to improve the question-answering quality and efficiency of the knowledge question-answering system, the knowledge question-answering solution provided in the embodiment of the present application mainly introduces tags for each knowledge content in the knowledge base included in the knowledge base management system, which helps users to manage knowledge content in a personalized way while facilitating users to find and organize knowledge content more conveniently. For example, users can directly conduct knowledge questions and answers based on specific knowledge content under a certain tag, thereby improving the personalization of knowledge questions and answers. The knowledge question-answering solution provided in the embodiment of the present application mainly includes: tag definition and tag index. The following is a brief introduction to the specific implementation process of the tag definition and tag index of the knowledge question-answering solution:

[0101] (1) Tag definition means: users define tags for different knowledge contents in the knowledge base in a personalized way; that is, the embodiments of the present application support users to customize the first tag for knowledge contents according to their own understanding of the knowledge contents; the tags defined for different knowledge contents can be the same or different. Among them, tags can be understood as summaries of the main contents of the knowledge contents, such as tags such as "new energy", "quantum theory", "autonomous driving" and "relativity".

[0102] (2) Tag indexing refers to the process of automatically indexing the knowledge content defined by the first tag selected by the user for the question information during the knowledge question and answer process. Specifically, at least one first tag is displayed in the knowledge question and answer interface provided by the intelligent assistant, and each first tag is used to index at least one knowledge content from the user's personalized knowledge base; when the user edits the question information in the knowledge question and answer interface, the user is supported to select a target tag for the question information in a personalized manner, and the target tag is any one of the at least one first tag; in this way, the intelligent assistant can index the specific knowledge content defined by the target tag in the user's knowledge base in the knowledge base management system based on the target tag selected by the user, thereby generating answer information corresponding to the question information based on the specific knowledge content and the question information, and outputting the answer information, so that the user can view the answer information in a timely manner.

[0103] It has been found in practice that the knowledge question-and-answer solution provided in the embodiment of the present application has obvious advantages in realizing knowledge retrieval. The following is an example of comparing the solution of the present application with the solution for realizing knowledge retrieval in the existing mainstream question-and-answer scenario to illustrate the advantages of the embodiment of the present application, wherein:

[0104] The solutions for implementing knowledge retrieval in the existing mainstream question-and-answer scenarios have at least the following defects: ① Single information source: Some products only support the collection of knowledge content, but only rely on information sources on the public network when conducting knowledge questions and answers, and cannot fully utilize the knowledge content organized / collected by users themselves, and cannot match information sources with user needs. ② Limitations of knowledge retrieval (or information retrieval): Most products cannot implement flexible questions based on the knowledge base, and users are restricted when retrieving knowledge content, and cannot efficiently use the content of their personal knowledge base. ③ Single knowledge base management function: Some products only support the entire knowledge base as an information source for knowledge retrieval, which means that when conducting knowledge retrieval, they only stay at the simple keyword matching level and lack the ability to conduct certain in-depth analysis and intelligent response.

[0105] However, the knowledge question and answer solution provided by this application has at least the following advantages: ① Customized tags: Supporting users to customize the first tag for knowledge content, not only ensures that users can rely on the memory of tag definition to select the first tag corresponding to the knowledge content they want to refer to in the subsequent knowledge question and answer process, thereby helping users to achieve personalized answer generation; and, considering that different users have different understandings of the same knowledge content, allowing users to customize the first tag of the knowledge content makes it convenient for the knowledge content retrieved through the first tag knowledge to be the content that users really want to refer to in the subsequent knowledge question and answer process, thereby ensuring that the answers generated based on the knowledge content meet user expectations and improve user experience. ② Improve information retrieval efficiency: In the process of knowledge question and answer, users customize the target tag for the question information to help users quickly and accurately find the specific knowledge content indexed by the target tag in the knowledge base. Computer equipment (or intelligent assistant, knowledge question and answer system, knowledge base management system) does not need to screen through a large amount of irrelevant knowledge content, which greatly saves time and energy and improves the efficiency and accuracy of information retrieval. ③ Improve user experience: The embodiment of the present application introduces the function of asking questions based on knowledge base tags and interactive optimization, which not only enhances the interactivity between users and the knowledge base, improves the pertinence and efficiency of information query, but also simplifies the user's operation process and promotes the effective management and application of personal knowledge, thereby improving work efficiency and personal learning effects.

[0106] The knowledge question and answer system provided in the embodiment of the present application is an automatic question and answer solution for human-computer interaction, which can understand, analyze and answer questions raised by users; this makes the knowledge question and answer solution provided in the embodiment of the present application applicable to a variety of interactive dialogue scenarios that require the use of the knowledge question and answer system to realize inquiries. Interactive dialogue scenarios may include but are not limited to: ① Customer support: The knowledge question and answer system can be used as a customer support tool to answer common questions of users, reduce the workload of customer service personnel, and improve customer satisfaction. ② Internal knowledge base of the enterprise: The enterprise can use the knowledge question and answer system to build an internal knowledge base to help enterprise employees quickly find the required information and improve work efficiency. ③ Virtual assistant: The knowledge question and answer system can be used as a virtual assistant for individuals or enterprises, providing functions such as daily task management, scheduling, and reminder services. ④ Online education: The knowledge question and answer system can be applied to the field of online education to provide students with personalized learning resources and real-time question and answer services. ⑤ E-commerce: The knowledge question and answer system can help users answer questions during the shopping process, provide shopping suggestions, and improve the shopping experience. ⑥ Financial services: The knowledge question and answer system can provide real-time consulting services to customers of financial institutions such as banks and insurance companies, and answer questions about accounts, transactions, products, etc. ⑦ Medical consultation: The knowledge question-and-answer system can provide basic medical consultation services for patients, answering questions about diseases, treatments, drugs, etc. ⑧ Tourism consultation: The knowledge question-and-answer system can provide tourists with real-time tourism information, answering questions about attractions, hotels, transportation, etc. ⑨ News and information retrieval: The knowledge question-and-answer system can help users quickly find the news and information they need, improving the efficiency of information retrieval.

[0107] It should be understood that the above description is only an exemplary product performance and interactive dialogue scenario given in the embodiment of the present application, and does not limit the product performance and interactive dialogue scenario of the message processing solution provided in the embodiment of the present application. The knowledge question and answer system provided in the embodiment of the present application can provide efficient, accurate and convenient question and answer services in various interactive dialogue scenarios, showing high value and practicality in various interactive dialogue scenarios, and helping to improve user experience and satisfaction.

[0108] For the convenience of explanation, the following Figure 1a A schematic diagram of the architecture of a knowledge question-answering scenario is given; Figure 1a As shown, the knowledge question and answer scenario includes object 101, terminal 102, server 103 and server 104. The embodiment of the present application does not limit the number and naming of object 101, terminal 102, server 103 and server 104.

[0109] Among them, object 101 refers to a user holding terminal 102. Terminal 102 may refer to a terminal device with an interactive dialogue function; the terminal device runs the aforementioned intelligent assistant (such as the intelligent assistant is a PC-side intelligent question-and-answer platform, in which case the terminal device is a PC device), and object 101 may conduct an interactive dialogue with the intelligent assistant through the terminal 102 to achieve problem consultation. Terminal devices may include, but are not limited to: smartphones (such as smartphones deploying an Android system, or smartphones deploying an Internetworking Operating System (IOS)), tablet computers, portable personal computers, mobile Internet devices (Mobile Internet Devices, MID), vehicle-mounted devices, head-mounted devices, intelligent chat robots, aircraft and other terminal devices. The embodiments of this application do not limit the types of terminal devices, which are explained here.

[0110] Server 103 and server 104 are servers corresponding to terminal 102, which are used to interact with terminal 102 to provide computing and application service support for terminal 102, specifically to provide application services and technical support for the intelligent assistant running in terminal 102. Exemplarily, server 103 can be a knowledge question and answer system, which is mainly used to interact with terminal 102 to realize some interactive processes in the knowledge question and answer process; server 104 can be a knowledge base management system, which is used to store, manage and retrieve the user's personalized knowledge base and the first label of the knowledge content. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms. For example, server 103 is an independent physical server, wherein the terminal 102, server 103 and server 104 can be directly or indirectly connected by wired or wireless communication, and this application is not limited here.

[0111] The knowledge question-answering solution provided in the embodiment of the present application can be executed by a computer device in which an intelligent assistant is deployed or run. The computer device can be Figure 1aThe terminal or server in the system shown, or the computer device includes a terminal and a server. That is, the knowledge question answering scheme provided in the embodiment of the present application can be executed by the terminal or the server, or by the terminal and the server together. Figure 1a The knowledge question answering scenario shown introduces the process of the knowledge question answering solution, where:

[0112] 1) Object 101 is a customized first label for knowledge content in the knowledge base.

[0113] Specifically, the object 101 can set a first label for the knowledge content. For example, if the knowledge content is a web page about new energy vehicles, and the object 101 wants to set a first label of "new energy" for the web page, then the object 101 can upload the first label "new energy" through the label setting interface displayed by the terminal 102. The terminal 102 sends the first label "new energy" and the relevant information of the web page (such as the unique identifier of the web page) to the server 104 corresponding to the knowledge base management system, and the server 104 establishes an index relationship between the first label "new energy" and the web page, and stores the index relationship between the first label "new energy" and the web page in the database; if the object 101 sets the first label 1 for both the knowledge content 1 and the knowledge content 2, and sets the first label 2 for the knowledge content 3, then the knowledge content and the first label can be stored in the database in the form of kv (key-value) data pairs.

[0114] 2) Object 101 interacts with the intelligent assistant to implement knowledge question and answer.

[0115] Specifically, when the object 101 has an interactive dialogue requirement, the terminal 102 can open the knowledge question and answer interface provided by the intelligent assistant, and the knowledge question and answer interface is used for knowledge question and answer. At least one first tag can be displayed in the knowledge question and answer interface, and the object 101 can select a target tag for the question and answer information to be answered from the at least one first tag. The terminal 102 sends the question information edited by the object 101 and the selected target tag to the server 103 corresponding to the knowledge question and answer system. The server 103 analyzes the question information and sends the target tag to the server 104 corresponding to the knowledge base management system, so that the server 104 can index at least one knowledge content in the knowledge base according to the target tag. The server 104 returns the at least one knowledge content indexed by the target tag to the server 103; the server 103 determines at least one content segment for generating an answer based on the matching degree between the content segments included in the question information and the knowledge content, and generates answer information based on the content segment and the question information. The server 103 returns the answer information to the terminal 102, so that the terminal 102 can display the answer information in the knowledge question and answer interface through the intelligent assistant, so that the object 101 can intuitively perceive the answer information corresponding to the question information.

[0116] Based on the above brief introduction to the knowledge question answering scheme and knowledge question answering scenario provided in the embodiment of the present application, the following points need to be explained:

[0117] ① The above-mentioned embodiments of the present application Figure 1a The system architecture shown is for the purpose of more clearly illustrating the technical solution of the embodiment of the present application, and does not constitute a limitation on the technical solution provided by the embodiment of the present application. A person skilled in the art will know that with the evolution of the system architecture and the emergence of new business scenarios, the technical solution provided by the embodiment of the present application is also applicable to similar technical problems. In other words, Figure 1a What is shown is only an exemplary architecture diagram of the knowledge question and answer scenario; in actual applications, the number and distribution of computer devices included in the knowledge question and answer scenario may change, and the embodiments of the present application do not limit the architecture diagram of the knowledge question and answer scenario.

[0118] For example, Figure 1aThe introduction is based on the example that both the knowledge question answering system and the knowledge base management system are centralized systems. A centralized system means that: the system (such as the knowledge question answering system or the knowledge base management system) is controlled by a central processor (the central processor can be deployed in one or more computer devices); all tasks in the knowledge question answering scenario are uniformly processed and executed by the central processor, without the need for network connection and node configuration. However, in actual applications, the knowledge question answering system and the knowledge base management system can also be distributed systems; a distributed system means that: the system is formed by multiple dispersed computer devices interconnected through a network; in a distributed system, computer devices can exchange information, and processing and control functions are distributed on each computer device, and there is no master-slave relationship between the computer devices; such as Figure 1b As shown, the knowledge question answering system and the knowledge base management system are composed of multiple servers deployed on different servers, and the knowledge question answering system and the knowledge base management system are distributed networks. Alternatively, in actual applications, the knowledge question answering system and the knowledge base management system can be deployed on the same server or in the same distributed network.

[0119] ② The collection and processing of relevant data in the embodiments of this application should be strictly in accordance with the requirements of relevant laws and regulations. The acquisition of personal information requires the knowledge or consent of the individual subject (or the legal basis for obtaining the information), and the subsequent use and processing of data shall be carried out within the scope of authorization of laws and regulations and the subject of personal information. For example, when the embodiments of this application are applied to specific products or technologies, such as submitting question information, the user's permission or consent is required, and the collection, use and processing of relevant data (such as problem analysis, knowledge retrieval and answer generation of question information, etc.) shall comply with relevant laws, regulations and standards of the relevant countries and regions.

[0120] Based on the knowledge question and answer scheme described above, the embodiment of the present application proposes a more detailed knowledge question and answer method. The knowledge question and answer method proposed in the embodiment of the present application will be introduced in detail below in conjunction with the accompanying drawings.

[0121] Figure 2 A flow chart of a knowledge question answering method provided by an exemplary embodiment of the present application is shown; the knowledge question answering method can be executed by a computer device, which can be Figure 1a The terminal 101 in the system / scenario shown. The knowledge question answering method may include but is not limited to steps S201-S203:

[0122] S201: Display at least one first tag in the knowledge question and answer interface.

[0123] The knowledge question and answer interface is a user interface (UI) that can perform knowledge question and answer; the user interface is a medium for interaction and information exchange between the operating system and the object, and can realize the conversion between the internal form of information and the form acceptable to humans. In other words, the knowledge question and answer interface is a service interface provided by the intelligent assistant running in the computer device; the object can perform human-computer interaction in the knowledge question and answer interface to realize knowledge question and answer.

[0124] In a specific implementation, when an object has a query demand, it can open a knowledge question and answer application in a computer device (such as the aforementioned browser deployed on a PC device and having knowledge question and answer capabilities), and use its own object identifier (i.e., an identifier that can uniquely identify the identity of the object, such as a user ID, user nickname, etc., such as an object identifier that is assigned to the object by the knowledge question and answer application when the object registers the knowledge question and answer application) to perform a login operation in the knowledge question and answer application; at this time, the computer device responds to the login operation for the object identifier in the knowledge question and answer application, and displays the knowledge service interface of the knowledge question and answer application. Among them, the knowledge service interface includes a question and answer trigger component (or question and answer trigger option, question and answer trigger button, question and answer trigger control, etc.) for entering the knowledge question and answer interface; in this way, when the object performs a trigger operation on the question and answer trigger component in the knowledge service interface, the knowledge question and answer interface is displayed, and the knowledge question and answer interface can display at least one first tag defined for knowledge content in the personalized knowledge base of the object, and each tag is used to index at least one knowledge content from the knowledge base. Exemplarily, a schematic diagram of a first tag indexing knowledge content in a knowledge base can be seen in Figure 3a ;like Figure 3a As shown: the knowledge base includes knowledge content 1, knowledge content 2 and knowledge content 3, knowledge content 1 is defined with first label 1, first label 2 and first label 3, knowledge content 2 is defined with first label 2 and knowledge content 3 is defined with first label 4. Then, the first labels displayed in the knowledge question and answer interface include: first label 1, first label 2, first label 3 and first label 4; wherein, the knowledge content indexed by first label 1 includes knowledge content 1, the knowledge content indexed by first label 2 includes knowledge content 1 and knowledge content 2, the knowledge content indexed by first label 3 includes knowledge content 1, and the knowledge content indexed by first label 4 includes knowledge content 3.

[0125] It should be noted that the login operation of the subject using the object identifier to log in to the knowledge quiz application may include but is not limited to: scanning login, password login, link login, etc. Among them: ① Scanning login refers to: a graphic code (such as a QR code, etc.) is displayed in the login interface provided by the knowledge quiz application, and the subject automatically logs in after scanning the graphic code through a terminal device (such as a smart phone or tablet, etc.); an exemplary scanning code login login diagram can be seen in Figure 3b, a graphic code 301 is displayed in the login interface. When the terminal scans the graphic code 301 successfully and confirms the login, the object ID is automatically used to log in to the knowledge quiz application. ② Password login means: the object enters a password in the login interface provided by the knowledge quiz application to log in using the object ID; the password can be a static password or a dynamic password. A static password refers to a fixed password pre-set by the object, while a dynamic password is a password (such as a verification code) dynamically issued by the knowledge quiz application each time the object logs in; an exemplary schematic diagram of password login can be seen in Figure 3c ③ Link login refers to the process of automatically logging into the knowledge quiz application using the object identifier logged into the computer device (such as the object identifier logged into another application independent of the knowledge quiz application, or the object identifier logged into the computer device directly); an exemplary schematic diagram of link login can be found in Figure 3d A login link 302 is displayed in the login interface of the knowledge quiz application. When the login link 302 is triggered, the object identifier corresponding to the login link 302 is used to automatically log in to the knowledge quiz application.

[0126] It should also be noted that the embodiment of the present application sets one or more question trigger components in the knowledge service interface, each question trigger component corresponds to a different knowledge question and answer interface, and each knowledge question and answer interface supports knowledge question and answer. The following two exemplary processes of displaying the knowledge question and answer interface through the question and answer trigger component are given in conjunction with the accompanying drawings; among which:

[0127] (1) Figure 4aAs shown, after logging into the knowledge question and answer application using the object identifier, a knowledge service interface 401 is displayed, and the knowledge service interface 401 includes a question trigger component 402 corresponding to the knowledge base; when the object performs a trigger operation on the question trigger component 402, the computer device displays a knowledge question and answer interface 403 corresponding to the knowledge base. Considering that the knowledge question and answer interface 403 is displayed by triggering the question trigger component 402 corresponding to the knowledge base, the knowledge question and answer interface 403 may include a question and answer area 404 and an information display area 405. Among them: ① The question and answer area 404 is an area that supports interactive dialogue, and the question and answer area 404 includes a message display sub-area 4041 and a dialogue sub-area 4042; the message display sub-area 4041 is used to display or present the question message and answer message generated by a round of knowledge question and answer, and the question message can be understood as a message bubble (or dialogue bubble) that at least includes question information, such as question message 40411 includes question information "Is the weather suitable for playing football today?"; the question message provided in the embodiment of the present application can also include a target tag selected by the object for the question information, such as question message 40412 includes a target tag "#new energy" and question information "tram quality...". Similarly, the answer message is a message bubble that includes answer information, such as the answer message 40413 for question message 40411 includes answer information "The weather is sunny, suitable for playing football". ② The information display area 405 is used to display the relevant content of the information source. For example, when the information source is a knowledge base, the information display area 405 is used to display the knowledge content in the knowledge base. For example, when the information source is the Internet, the information display area 405 may display relevant content on the Internet.

[0128] (2) Figure 4b As shown, after the object identifier is used to log in to the knowledge question and answer application, a knowledge service interface 401 is displayed, and the knowledge service interface 401 includes a question trigger component 406 corresponding to the knowledge question and answer application; when the object performs a trigger operation on the question trigger component 406, the computer device displays a knowledge question and answer interface 407 corresponding to the knowledge question and answer application. Since the knowledge question and answer interface 407 is displayed by triggering the question trigger component 406 corresponding to the knowledge question and answer application, the knowledge question and answer interface 407 only includes a question and answer area 408; wherein the question and answer area 408 also includes a message display sub-area and a dialogue sub-area, which can be seen in FIG. Figure 4a The relevant content will not be elaborated here.

[0129] In summary, Figure 4a and Figure 4bThese are the triggering methods and interface styles of the two knowledge question and answer interfaces given in the embodiments of this application; the knowledge question and answer method provided in the embodiments of this application is applicable to each knowledge question and answer interface, and functions such as label indexing and label filtering can be implemented on both knowledge question and answer interfaces. For the sake of ease of explanation, the knowledge question and answer interface displayed by triggering the question and answer trigger component corresponding to the knowledge base is used as an example to continue to introduce the embodiments of this application, which is specially explained here.

[0130] Furthermore, when the subject has a query need, the subject can conduct a knowledge question and answer in the question and answer area of ​​the knowledge question and answer interface. Specifically, when the subject is editing question information in the dialogue sub-area of ​​the question and answer area in the knowledge question and answer interface, the subject can select a target tag for the question of this knowledge question and answer; in this way, at least one knowledge content in the knowledge base indexed by the target tag can be used as an information source for the answer information corresponding to the question information. In other words, the scope of the information source is narrowed from the entire knowledge base to at least one knowledge content indexed by the target tag, which effectively reduces the workload of knowledge retrieval and improves the matching between the information source and the question information, thereby improving the accuracy of the answer information to a certain extent.

[0131] Specifically, the specific implementation process of the object selecting a target tag for question information in the knowledge question and answer interface may include: displaying at least one first tag in the knowledge question and answer interface; in the process of the object editing question information in the dialogue sub-area in the question and answer area, selecting a target tag for the question information from at least one first tag, and the number of target tags can be one or more; the selected target tag and question information are displayed together in the dialogue sub-area as the question message to be sent.

[0132] Among them, the display timing of the first label defined by the knowledge content in the knowledge base in the knowledge question and answer interface includes: the first label is displayed in the knowledge question and answer interface by default, or the first label is displayed in the knowledge question and answer interface according to the label call operation. That is to say, the first label can be displayed in the knowledge question and answer interface by default when the knowledge question and answer interface is displayed, or it can be displayed in the knowledge question and answer interface after the object performs the label call operation; the following is an introduction to these two display timings in conjunction with the accompanying drawings, wherein:

[0133] 1) The first label is displayed by default in the knowledge question and answer interface.

[0134] Specifically, after the knowledge question and answer interface is triggered to display, the first label defined by the knowledge content in the knowledge base is directly displayed on the knowledge question and answer interface by default. Figure 5a As shown, the first label can be displayed in the designated position 501 in the dialogue sub-area. Of course, the first label can also be displayed in other display positions in the knowledge question and answer interface by default, and there is no limitation on this. Figure 5bAs shown, the first label can also be displayed in a floating window 502, which is displayed on the knowledge question and answer interface. On the one hand, the display position of the floating window 502 can be flexibly controlled by the object, such as the object can adjust the display position of the floating window 502 by dragging the floating window 502 to meet the personalized display requirements of the object. On the other hand, the floating window 502 can also be hidden or displayed by the object's autonomous control, such as the floating window 502 includes a hidden option 5021, and the computer device responds to the object's triggering operation on the hidden option 5021, hides the floating window 502 on the knowledge question and answer interface, and displays a prompt message 503; when the prompt message 503 is triggered, the display of the floating window 502 can be restored, which greatly facilitates the object's control of the display state of the floating window 502 and enhances the object experience.

[0135] 2) Displaying at least one first tag in the knowledge question and answer interface according to the tag invoking operation performed by the object.

[0136] Specifically, during the process of editing question information in the dialogue sub-area of ​​the question-and-answer area (such as before the question information begins to be edited, the question information is being edited, or after the question information is edited), if the object wants to select a target tag for the question information, then the object can perform a tag recall operation in the knowledge question-and-answer interface. At this time, the computer device displays a tag filtering window corresponding to the tag recall operation in the knowledge question-and-answer interface, and displays at least one first tag defined for the knowledge content in the knowledge base in the tag recall window. Among them, the tag recall operation performed in the knowledge question-and-answer interface may include any of the following:

[0137] ① Input operation for shortcut keys in the dialogue sub-area. Shortcut keys may include but are not limited to: specific characters (such as "#", "*" or "label", etc.) input through a virtual keyboard or a physical keyboard, or a key on a computer device (such as two consecutive clicks on a volume key), etc. Take the example of inputting a specific character "#" as an example. Figure 6a As shown, when the object enters the "#" character in the dialog sub-area, the tag filtering window 601 is automatically called up, and at least one first tag is displayed in the tag filtering window 601. ② Triggering operation of the tag component. Figure 6bAs shown, it is assumed that a label component 602 is displayed in the dialogue sub-area of ​​the question and answer area in the knowledge question and answer interface; when the subject wants to add a target label, the subject can actively click on the label component 602. At this time, the computer device responds to the click of the label component 602, triggers the display of the label filter window 601, and displays at least one first label in the label filter window 601. ③ Specified gesture operations. Gesture operations may include but are not limited to: double-click operations (such as two consecutive clicks on the same position on the display plane through a finger, mouse, or electronic pen, etc.), sliding operations with specified trajectories (such as sliding operations with a specified trajectory shape as an S shape), long press operations or knuckle operations (such as clicking operations through finger joints). Figure 6c As shown, a sliding operation with a specified trajectory as a gesture operation is taken as an example; if the object performs an S-shaped sliding operation with a trajectory on the dialogue sub-area (or the knowledge question and answer interface), indicating that the object wants to call up the first tag, the tag filtering window 601 is displayed, and at least one first tag is displayed in the tag filtering window 601.

[0138] It should be noted that the above are only three exemplary tag call-up operations given in the embodiments of the present application; in actual product design, developers can also set other feasible tag call-up operations according to object requirements. The embodiments of the present application do not limit the specific implementation process of the tag call-up operation.

[0139] S202: Receive a target tag selected for question information.

[0140] Based on step S201, after at least one first tag is displayed in the knowledge question and answer interface (such as calling up a tag filter window to display or displaying it by default in the knowledge question and answer interface), the subject can select a tag from the at least one first tag as the target tag of the question and answer information to be answered this time; when the computer device detects that the subject performs a selection operation on one or more first tags, it is determined that the computer device receives the target tag selected by the subject for the question information. Figure 7 As shown, the first tag 701, the first tag 702 and the first tag 703 are displayed in the tag filtering window; the object can select one or more first tags as target tags according to its own needs, such as selecting the first tag 701 as the target tag for the question information, or selecting the first tag 701 and the first tag 702 as the target tags for the question information; at this time, the computer device receives the target tag selected for the question information, and displays the selected target tag in the dialogue sub-area, so that the object can intuitively perceive the selected target tag.

[0141] The embodiment of the present application also supports the object to delete the target tag in the dialogue sub-area to meet the need of the object to change the target tag selected by the question information; continue to see Figure 7If the subject wants to cancel the first tag 701 as the target tag, when the subject triggers the first tag 701 in the dialog sub-area (such as controlling the mouse to place the cursor on the first tag 701, or performing a click operation on the first tag), a cancel option 7011 corresponding to the first tag 701 appears. If the subject triggers the cancel option 7011, it indicates that the subject wants to cancel the first tag 701 as the target tag of the question information, and the first tag 701 is deleted in the dialog sub-area.

[0142] Furthermore, considering that the number of first tags defined for knowledge content in the knowledge base may be large, in order to facilitate the object to quickly select the desired target tag; the embodiment of the present application also provides a tag filtering function (or tag retrieval function), and the so-called tag filtering refers to real-time filtering and displaying at least one second tag related to the input information entered by the object in the dialogue sub-area, and the second tag belongs to at least one first tag. Through this tag filtering function, the object can flexibly and quickly locate the second tag that it wants to use from a large number of first tags, shorten the tag selection time, and improve the object's question-and-answer experience.

[0143] Specifically, after the object performs a tag invoking operation in the knowledge question and answer interface, such as after the object enters #invokes the tag filtering window in the dialogue sub-area in the knowledge question and answer interface, the object can continue to input information, such as entering information in the dialogue sub-area. At this time, the computer device receives the input information entered by the object in the knowledge question and answer interface, specifically the input information edited in the dialogue sub-area in the knowledge question and answer interface; and, in the tag filtering window, at least one first tag is updated and displayed as at least one second tag related to the input information. The second tag being related to the input information may mean that: the input information is included in the second tag, such as if the input information is "new energy", and the second tag is "new energy"; or, the semantic features of the second tag are the same or similar to the semantic features of the input information, such as if the input information is "electric car", and the second label is "new energy vehicle". Figure 8aAs shown, when the object enters # in the dialogue sub-area, a tag filtering window is output, and the tag filtering window includes all the first tags defined for all the knowledge content in the knowledge base. Then, the object can continue to edit the input information in the dialogue sub-area. For example, if the input information is "new", after the computer device receives the input information, it performs tag filtering processing based on the input information to obtain at least one second tag related to the input information, such as the second tags include "news", "new energy" and "new energy car...". The computer device displays at least one second tag in the tag filtering window in real time; this helps the object to quickly define and find the specified second tag from the massive first tags, and improves the flexibility of tag selection. The object can continue to edit the input information. If the input information is "new energy", the computer device continues to perform tag filtering processing according to the received new input information to obtain at least one second tag related to the new input. At this time, the second tags include "new energy" and "new energy car..." until the object filters the second tag it wants to use.

[0144] In some cases, at least one of the first labels may not contain a second label related to the input information; in this case, it is supported to use the input information edited by the object as part of the question information. That is to say, if the edited input information does not search for the relevant second label or the object does not select the second label after the object performs the label recall operation, the input information is used as part of the question information, and the object continues to edit the question information. In this way, according to the correlation between the input information edited by the object and the first label, the input information edited by the object can be flexibly processed (such as searching for the second label or using it as part of the question information), reducing the sense of disconnection between the two processes of inputting the question information and selecting the target label when the object is edited, thereby improving the question-answering experience of the object.

[0145] Above Figure 8a The label filtering function is introduced from the interface dimension. The following is an introduction to the background logic of label filtering based on the input information after the computer device receives the input information. Specifically, in the process of setting / defining the first label for the knowledge content, the computer device will perform word segmentation on the set first label, and establish a multi-field and multi-dimensional inverted index of the first label in the knowledge base interaction system corresponding to the knowledge question application, in order to quickly retrieve or index the second label related to the input information based on the input information edited by the object. The various concepts mentioned above are explained below, including:

[0146] ① The word segmentation process for the first label is to segment the first label, and divide the first label into multiple independent fields, such as vocabulary or words, according to the word segmentation rules, so as to facilitate subsequent label screening. For example, the word segmentation rule is to segment the first label by character string; if the first label is "new energy", then the keywords that may be obtained by segmenting the first label include "new", "energy", "source", "new energy" and "energy". For another example, the word segmentation rule is to segment the first label by semantic division; if the first label is "new energy", the keyword obtained by the word segmentation process is "energy".

[0147] ② Inverted index is an indexing method used to store the mapping of a keyword's storage location in a set of first tags; it allows objects to quickly find the first tag containing the keyword (the first tag is called the second tag) by editing the keyword (i.e. inputting information), which greatly improves the efficiency of tag retrieval. Multi-field and multi-dimensional inverted index means that multiple inverted indexes can be established for the same first tag based on different fields or dimensions. Figure 8b As shown, the inverted index is established according to the field, specifically, the first label is divided according to the string to establish the inverted index. Assuming that the first label "new energy" is segmented to obtain a keyword "new" and "energy", the first label "news" is segmented to obtain a keyword "new", and the first label "traditional energy" is segmented to obtain a keyword "energy"; then when the inverted index of the keyword "new" is established, the first labels that can be retrieved according to the keyword "new" include "new energy" and "news". Similarly, when the inverted index of the keyword "energy" is established, the first labels that can be retrieved according to the keyword "energy" include "new energy" and "traditional energy". The inverted index is established according to semantics, specifically, the first label is divided according to semantics. Assuming that the keyword corresponding to the first label "new energy" is "energy", the keyword corresponding to the first label "news" is "information", and the keyword corresponding to the first label "traditional energy" is "energy"; then, when the inverted index of the keyword "energy" is established, the first labels that can be retrieved include "new energy" and "traditional energy". Similarly, when the inverted index of the keyword "information" is established, the first labels that can be retrieved include "new energy" and "traditional energy". In summary, the above-mentioned establishment of inverted indexes according to fields and establishment of inverted indexes according to semantics are two dimensions of establishing inverted indexes. Of course, in actual applications, inverted indexes established according to other dimensions may also be included, and there is no limitation on this.

[0148] ③The knowledge base interaction system is an organizational structure based on multiple clusters and multiple indexes, which can cope with the storage and retrieval of large-scale TB (Terabyte) level user data; for example, the knowledge base interaction system can be an EleasticSearch system. In other words, the tag retrieval solution (i.e., RAG (Retrieval-Augmented Generation) solution) for the knowledge base tags provided in the embodiment of the present application can be implemented based on the EleasticSearch system. The EleasticSearch system is a distributed full-text search engine that can process massive amounts of data quickly and in real time. The inverted index mentioned above is established in the knowledge base interaction system; among them, the system architecture diagram of the knowledge base interaction system is shown in the figure. Figure 8c As shown. The knowledge base interaction system includes at least one index cluster and at least one label storage shard; wherein, the index cluster stores addresses that can be routed to sub-shards in the label storage shard, specifically, the index cluster includes one or more index modules, each index module stores addresses that can be routed to sub-shards in the label storage shard. The sub-shard is used to store the first label (the first label may be defined by the object, or automatically generated by a computer device) of the knowledge content in the knowledge base (i.e., the knowledge base created or received by the object corresponding to the object identifier) ​​of the login knowledge question and answer application. For example, Figure 8c As shown, the sub-shard 801 in the tag storage shard 1 stores the first tag "new energy" defined as the knowledge content, the sub-shard 802 stores the first tag "news" defined as the knowledge content, and so on.

[0149] Based on this, the computer device can route to the index module (or Index) in different index clusters according to the object identifier of different objects, and then perform shard routing through the input information to retrieve the second label related to the input information; it is worth noting that if the object does not edit the input information, then the computer device directly routes to the sub-shard in the label storage shard storing all the first labels defined by the object according to the object identifier when performing shard routing to retrieve all the first labels. Taking the object edited input information as an example, the above routing process may include: after the computer device receives the input information edited by the object in the knowledge question and answer interface, the computer device routes to the target index cluster in the knowledge base interaction system according to the object identifier, and the target index cluster is an index cluster in at least one index cluster that stores an address that can be routed to at least one reference sub-shard, and the reference sub-shard is a sub-shard that stores the first label of the knowledge content in the knowledge base corresponding to the object identifier. Then, the computer device obtains the first label related to the input information according to the input information edited by the object, the address of each reference sub-shard and the object identifier, and the retrieved first label related to the input information is used as the second label to be displayed.

[0150] It can be seen that, on the one hand, through the above-mentioned label screening method, it is possible to quickly locate or retrieve the second label related to the input information according to the keywords edited by the object (i.e., the input information), meet the needs of fast label screening of the object, and improve the flexibility of label selection of the object. On the other hand, the knowledge base interaction system is composed of index clusters and label storage shards to ensure that clusters can be dynamically allocated according to the growth of business data; specifically, as the user data of the knowledge question and answer application grows, it supports the rapid expansion of the knowledge base interaction system through scalable index clusters and label storage shards, ensuring that the performance of data writing and retrieval for the knowledge base interaction system is within a controllable range to cope with the rapid growth of the business in the later stage.

[0151] S203: Outputting answer information corresponding to the question information in the knowledge question and answer interface.

[0152] After receiving the target tag selected by the object and the edited question information, the computer device can perform knowledge questions and answers based on the target tag and question information; specifically, the terminal sends the target tag and question information to the knowledge question answering system, and the server in the knowledge question answering system performs knowledge questions and answers based on the target tag and question information. Among them, the knowledge question answering process specifically includes: performing tag indexing based on the target tag, and generating answers based on the segmented content indexed by the tag to obtain answer information corresponding to the question information. In this way, the computer device then outputs the answer information corresponding to the question information in the knowledge question answering interface so that the object can understand the answer information; specifically, the server in the knowledge question answering system returns the answer information to the terminal so that the terminal can output an answer message in the knowledge question answering interface, and the answer message includes the answer information corresponding to the question information.

[0153] A computer device performs tag indexing and outputs answer information corresponding to question information in a knowledge question and answer interface. Fig. 9 .like Fig. 9 As shown, a question message 901 edited and sent by an object is displayed in the message display sub-area in the question-and-answer area, and the question message 901 contains question information and a target tag. The computer device performs tag indexing based on the target tag, specifically, filtering one or more knowledge contents with defined target tags from the knowledge base, such as knowledge content 1 is a document, knowledge content 2 is a picture, etc. Usually, to facilitate knowledge indexing, it is supported to segment the knowledge content, such as when the format of the knowledge content is a document, the document is segmented to obtain multiple document segments, and when the format of the knowledge content is audio, the audio is segmented to obtain multiple audio segments; in an embodiment of the present application, each knowledge content includes multiple segmented contents, such as document segments for documents, audio segments for audio, etc.

[0154] Taking into account that the source of the answer information corresponding to the question information is the fragmented content in the knowledge content, in order to reduce the amount of calculation and improve the accuracy of the answer information; the embodiment of the present application also supports, after indexing one or more knowledge contents based on the target tag, based on the content identifier of the knowledge content (such as used to uniquely identify a knowledge content identifier, such as a document ID), calculating the matching degree of the question information and the fragmented content included in each knowledge content, and obtaining the matching degree results corresponding to the fragmented content included in each knowledge content. The matching degree results are used to characterize: the reliability of the fragmented content as a reference for generating the answer information corresponding to the question information. The higher the reliability, the more the fragmented content should be recalled for reference in answer generation.

[0155] In this way, the computer device can recall the target fragment content whose matching result is greater than the matching threshold (such as 80% or other reasonable values) from each fragment content included in each knowledge content indexed based on the target tag based on the matching result. The computer device then generates answer information corresponding to the question information based on the target fragment content whose matching result is greater than the matching threshold; wherein, the embodiment of the present application supports the introduction of an AI model to generate answer information, specifically assembling the recalled fragment content as input data of the IA model, and inputting the input data into the AI ​​model, so that the AI ​​model generates answer information based on the question information and the input data, and automatically generates answers by introducing an AI model in the knowledge question and answer process, which not only greatly improves the accuracy and reliability of the answers by relying on the powerful knowledge reserve of the AI ​​model, but also relies on the powerful language understanding and organization ability of the AI ​​model to ensure the fluency and readability of the answers. Among them, the AI ​​model is a machine learning model (or a large model) with information generation capability, and the embodiment of the present application does not limit the specific model type of the AI ​​model; such as the AI ​​model may be a mixed model, a large language model, a linear regression model, and a random forest model, etc. Finally, the answer information is output to the knowledge question and answer interface in the form of an answer message.

[0156] In summary, on the one hand, the embodiments of the present application support label indexing to quickly locate specific knowledge content in the knowledge base as the information source of answer information, avoiding problems such as equipment failure caused by information overload. On the other hand, it supports label retrieval / label screening to help objects quickly locate the desired target label, thereby improving the flexibility of selecting target labels; and for the same question information, when the target labels selected by the object are different, even if the knowledge content is the same, the answer information finally generated is different, meeting the user's requirements for personalized generation of answer information and improving the user experience. On the other hand, answer information only needs to be generated based on the knowledge content indexed by the target label in the knowledge base, which greatly improves the speed and efficiency of answer generation, thereby improving the efficiency of knowledge question and answering, and helping users improve work efficiency and personal learning effects.

[0157] Fig.10 A flow chart of a knowledge question answering method provided by an exemplary embodiment of the present application is shown; the knowledge question answering method can be executed by a computer device, which can be Figure 1a The terminal 101 in the system / scenario shown. The knowledge question answering method may include but is not limited to steps S1001-S1004:

[0158] S1001: Receive a first tag set for knowledge content in a knowledge base.

[0159] Considering that different subjects have different understandings of the same knowledge content, the first labels that different subjects give to the same knowledge content will also be different; by allowing subjects to set the first label for the knowledge content in a personalized way, it can be ensured that the index relationship between the first label and the knowledge content is more in line with the requirements of the subjects; and the first labels of the knowledge content are customized by the subjects, which also avoids blind selection in the label selection process when the subjects are not familiar with the first label.

[0160] The embodiment of the present application supports the object to set the label of the knowledge content at any time. Optionally, when the knowledge content is imported into the knowledge base, the first label can be set for the knowledge content. Optionally, after the knowledge content has been imported into the knowledge base, the first label setting is initiated for the knowledge content; Fig.11a As shown, the computer device outputs a label editing window 1102 in response to a trigger operation (such as a right-click operation or a double-click operation, etc.) of the object on the knowledge content 1101 in the knowledge base. The object can set a new first label in the label editing window 1102, or re-edit the original first label, etc., to set the first label for the knowledge content 1101.

[0161] It should be noted that the first label defined for knowledge content can be set actively by the object, or it can be automatically generated by a computer device after performing content analysis on the knowledge content; wherein, content analysis includes extracting content data (such as text and pictures, etc.) from knowledge content (such as web pages, PDF documents, word documents and other media), and performing data cleaning, sorting and irrelevant information removal on the extracted content data, so as to extract keywords that can summarize the main content of the knowledge content, and use the keywords as the first label of the knowledge content. Based on this, the first label that the object sets for the knowledge content can be obtained by directly re-editing the first label generated by the computer device, such as in Fig.11a The label editing window 1102 shown modifies the first label automatically generated by the computer device to overcome inaccurate or other problems of the first label generated by the computer device.

[0162] S1002: Display at least one first label in the knowledge question and answer interface.

[0163] S1003: Receive a target tag selected for question information.

[0164] It should be noted that the specific implementation process shown in steps S1002-S1003 is the same as the above Figure 2 The specific implementation process shown in steps S201-S203 in the illustrated embodiment is the same, and reference may be made to the relevant description of the specific implementation process shown in steps S201-S202, which will not be repeated here.

[0165] It should also be noted that the embodiments of the present application support objects to customize information sources during the knowledge question and answer process; the so-called customized information source means that the object independently selects the information source according to its own answer source needs. Among them, the information source may include the Internet and at least one knowledge base; the knowledge base may be created by the object or shared by other objects. For example, the object creates knowledge base 1 in the knowledge question and answer application, and receives knowledge base 2 shared by other objects, then it is determined that the information sources that the object can independently select include: knowledge base 1, knowledge base 2 and the Internet. By allowing the object to choose to obtain answers from a wide range of information on the Internet according to its own needs, or to focus on specific content in the personal knowledge base to obtain answers, the object's requirements for generating multiple answers are met, thereby improving the object experience.

[0166] In a specific implementation, it is assumed that at least one information source component is included in the knowledge question and answer interface, and each information source component corresponds to an information source. Before the object selects the target tag in the dialogue sub-area, the object will first perform the selection of the information source component. For example, the object performs a selection operation on the first information source component in at least one information source component, and the computer device responds to the selection operation on the first information source component and displays the first information source component as a selected state; wherein, the first information source component is displayed as a selected state to prompt the object that the information source corresponding to the first information source component has been selected as the source of the answer. The embodiment of the present application does not limit the specific form of the display state, such as the transparency of the first information source component is lower than that of other information source components. In this way, when the information source corresponding to the first information source component is a knowledge base, at least one first label defined in the knowledge base corresponding to the selected first information source component can be displayed in the knowledge question and answer interface; for example, the computer device automatically displays at least one first label according to the operation of selecting the first information source component, or the computer device displays at least one first label according to the label call-up operation of the object.

[0167] It is worth noting that when the subject selects an information source component, the number of selected information source components can be one or more; for example, if there are multiple knowledge bases, multiple knowledge bases can be selected as the source of answers for this knowledge question and answer, or one or more knowledge bases and the Internet can be selected as the source of answers for this knowledge question and answer. This method of allowing answers to be generated based on multiple information sources not only helps the subject integrate knowledge content across information sources, but also improves the breadth and accuracy of the answer sources. For example, a schematic diagram of selecting multiple information sources can be seen in Fig.11b ;like Fig.11bAs shown, information source component 1103 and information source component 1104 are displayed in the knowledge question and answer interface. The information source corresponding to information source component 1103 is the Internet, and the information source corresponding to information source component 1104 is knowledge base 1. Optionally, when information source component 1104 is triggered, component candidate window 1105 is displayed, in which information source components corresponding to each knowledge base in all knowledge bases of the object are displayed, such as information source component 1104 corresponding to knowledge base 1, information source component 11051 corresponding to knowledge base 2, and information source component 11052 corresponding to knowledge base 3 in all knowledge bases; the object can select one or more information source components in the component candidate window 1105 to use the knowledge base corresponding to each information source component as the information source.

[0168] Furthermore, when the selected multiple information sources include a knowledge base, the at least one first tag displayed in the knowledge question and answer interface includes: a first tag in which the knowledge content in each selected knowledge base is defined. Fig.11b As shown, assuming that information source component 1104 and information source component 11051 are selected, and the first labels defined in the knowledge base 1 corresponding to information source component 1104 include first label 1 and first label 2, and the first labels defined in the knowledge base 2 corresponding to information source component 11051 include first label 1 and first label 3, then the first labels displayed in the knowledge question and answer interface include: first label 1, first label 2 and first label 3.

[0169] Furthermore, in the case where multiple knowledge bases are selected as information sources, the computer device performs label indexing based on the selected target label (which may come from different knowledge bases) and recalls multiple knowledge contents defined by the target label, and supports the generation of answer information based on the fragmented content in the multiple knowledge contents according to the content selection strategy. Among them, the content selection strategy may include but is not limited to: ① Indifferent selection strategy, the indifferent selection strategy refers to generating answer information based on the recalled fragmented content in each of the multiple knowledge contents recalled. For example, assuming that the selected knowledge bases are knowledge base 1 and knowledge base 2, and knowledge content 1 in knowledge base 1 is indexed by the target label, and knowledge content 2 in knowledge base 2 is indexed by the target label, when the indifferent recall strategy is used, the recalled knowledge content includes knowledge content 1 and knowledge content 2. In this way, the computer device generates answer information according to the fragmented content in the recalled knowledge content 1 that matches the question information, and the fragmented content in the knowledge content 2 that matches the question information.

[0170] ② Priority selection strategy. Priority selection strategy refers to generating answer information based on the priority order of the information source to which each of the multiple knowledge contents to be recalled belongs; wherein the information source can be a knowledge base and the Internet. For example, assuming that the selected knowledge bases are knowledge base 1 and knowledge base 2, and knowledge content 1 in knowledge base 1 is indexed by the target label, knowledge content 2 in knowledge base 2 is indexed by the target label, and the priority of knowledge base 1 is greater than the priority of knowledge base 2, then when the priority selection strategy is used, the recalled knowledge contents include knowledge content 1 and knowledge content 2; since the priority of knowledge base 1 to which knowledge content 1 belongs is higher than the priority of knowledge base 2 to which knowledge content 2 belongs, when generating answer information, the fragmented content in knowledge content 1 that matches the question information is used first to generate answer information; when the answer information generation fails or the generation quality is poor, the answer is generated based on the fragmented content in knowledge content 1 that matches the question information and the fragmented content in knowledge content 2 that matches the question information. It can be seen from this that this method of generating answer information based on information sources according to priority, such as when the answer information generated based on the knowledge content in the knowledge base is poor or fails, other information sources can be used as a backup information source for generating answer information, which can improve the success rate and accuracy of answer information generation to a certain extent, avoid the failure of computer equipment to answer in knowledge questions and answers, and improve the question-and-answer experience of the subject.

[0171] Among them, the priority of the information source can be pre-set by the developer of the knowledge question and answer application, or customized by the object when selecting the information source. Specifically, the exemplary process of customizing the priority of the information source when the object selects the information source may include: the object performs a priority setting operation on at least two selected target information source components, the target information source component is the information source component selected by the object as a reference for answer generation, and when the number of selected target information source components is at least two, the first label is a label defined in the information source corresponding to the at least two target information source components. The computer device displays a priority identifier at an adjacent display position of each target information source component according to the priority setting operation of the object for at least two target information source components, and the priority identifier is used to represent: the knowledge content indexed by the target label in the information source corresponding to the target information source component, as a reference priority when generating answer information. The embodiment of the present application does not limit the expression form of the priority identifier, such as the priority identifier is expressed in digital form, the smaller the value of the number, the higher the priority, such as the priority of number 1 is higher than the priority of number 2. For example: assuming that the priority identifier (such as number 1) displayed at the adjacent display position of the target information source component corresponding to knowledge base 1 is greater than the priority identifier (number 2) displayed at the adjacent display position of the target information source component corresponding to knowledge base 2, then the priority level of knowledge content 1 indexed by the target tag in knowledge base 1 as a reference when generating answer information is greater than the priority level of knowledge content 2 indexed by the target tag in knowledge base 2 as a reference when generating answer information; in this way, the computer device gives priority to referring to the fragmented content in knowledge content 1 when generating answer information, and when the answer information generation fails or is poor, it simultaneously refers to the fragmented content in knowledge content 1 and the fragmented content in knowledge content 2 to generate answer information.

[0172] like Fig.12 As shown, assuming that the information source component 1201 corresponding to the knowledge base 1 and the information source component 1202 corresponding to the knowledge base 2 are selected, it indicates that the object wants to select both the knowledge base 1 and the knowledge base 2 as reference information sources for generating answer information. In this case, if the object wants to set priorities for the knowledge base 1 and the knowledge base 2, the object can enter a priority value 1 in the input area 1203 in the display area occupied by the information source component 1201, and enter a priority value 2 in the input area 1204 in the display area occupied by the information source component 1202, indicating that the priority of the knowledge base 1 is higher than the priority of the knowledge base 2. Among them, the priority value 1 can be used as the priority identifier of the information source component 1201, and similarly, the priority value 2 can be used as the priority identifier of the information source component 1202. It should be understood that the priority setting operation performed by the object on at least two target information source components, except as follows Fig.12The specific implementation process shown may also be other operations, and the embodiment of the present application does not limit the specific operation process of the priority setting operation.

[0173] It should be noted that a round of knowledge questions and answers may include at least one question-and-answer dialogue, and a question-and-answer dialogue includes a question message and an answer message corresponding to the question message. In order to improve the flexibility of the source of answer information in different question-and-answer dialogues during a round of knowledge questions and answers, the embodiment of the present application supports flexible switching of information sources for different question-and-answer dialogues during a round of knowledge questions and answers. In other words, during a round of knowledge questions and answers, the information sources selected by different question-and-answer dialogues can be changed, which effectively avoids the need to restart a new round of knowledge questions and answers when the object wants to switch to another information source to continue the discussion, and improves the convenience of continuous question-and-answer dialogues under frequent topic switching scenarios (such as different information sources corresponding to different topic scenarios). Among them, each question-and-answer dialogue in a round of knowledge questions and answers supports the aforementioned functions of selecting multiple information sources and setting priority levels for multiple information sources.

[0174] Specifically, it is assumed that during the first question-and-answer dialogue in a round of knowledge question-and-answer, the information source component selected by the object is the first information source component, that is, the first information source component is in a selected state during the first question-and-answer dialogue. After the first question-and-answer dialogue ends, if the object wants to switch the information source, the object can perform an information switching operation, at which time the computer device responds to the source switching operation performed by the object in the knowledge question-and-answer interface, and displays the first information source component from a selected state to an unselected state. Also, the second information source component newly selected by the object is displayed from an unselected state to a selected state, and the object can continue to edit the question information in the second question-and-answer dialogue in the knowledge question-and-answer interface; wherein the second information source component is different from the first information source component, such as the first information source component is an information source component corresponding to the Internet, and the second information source component is an information source component corresponding to the knowledge base. The schematic diagram of the above-mentioned source switching can be seen in Fig.13 ;like Fig.13 As shown, assuming that the first information source component selected by the subject for question information 1301 in the first question-and-answer dialogue is information source component 1302, after the answer information 1303 corresponding to question information 1301 is displayed in the message display sub-area; if the subject wants to switch the information source and continue to ask questions, the subject can select a new second information source component, such as the second information source component is information source component 1304. At this time, information source component 1302 is displayed as unselected, and information source component 1304 is displayed as selected, and the subject can continue to input question information in the second question-and-answer dialogue in the dialogue sub-area, so as to realize flexible switching of information sources for question-and-answer in one round of knowledge question-and-answer.

[0175] S1004: Outputting answer information corresponding to the question information in the knowledge question and answer interface.

[0176] Among them, the answer information corresponding to the question information is displayed in the knowledge question and answer interface in the form of an answer message (such as the answer message is in the form of a message bubble), specifically in the message display sub-area in the knowledge question and answer interface.

[0177] In order to improve the object experience, the embodiment of the present application supports the introduction of a highlight tag display function in the message bubble (i.e., question and answer message) in the question and answer area. The highlight tag means that the target tag in the question message can be displayed in a highlighted state, so that the object can intuitively perceive the target tag; the highlight tag display function means that the object can directly filter and display all the indication contents defined by the target tag in the knowledge base by clicking on these highlight tags, so as to quickly locate and deeply explore the knowledge content related to the answer information generated by this question and answer dialogue.

[0178] As mentioned above, the question and answer area in the knowledge question and answer interface includes a message display sub-area, which can be used to display the question message and answer message generated by a round of knowledge question and answer; when the computer device detects that the target tag included in the question message in the message display sub-area is triggered, it indicates that the subject wants to quickly locate the relevant knowledge content, and then displays the content identifier of at least one knowledge content indexed by the target tag, and the content identifier is used to uniquely identify the knowledge content, such as the content identifier is in the form of a card or ID, etc. If the subject wants to view the knowledge content in detail, then the subject can perform a trigger operation on the target content identifier, which is any one of the content identifiers of at least one knowledge content, and at this time the computer device responds to the trigger operation on the target content identifier and displays the knowledge content corresponding to the target content identifier.

[0179] A schematic diagram of highlight label display can be seen in Fig.14 ;like Fig.14 As shown, a question message 1401 is displayed in the question and answer area, and the question message 1401 includes a highlighted target tag 1402 and question information 1403. When the target tag 1402 is triggered, the content identifier of at least one knowledge content indexed by the target tag is displayed; for example, the content identifier of at least one knowledge content indexed by the target tag is displayed in the information display area 1403 in the knowledge question and answer interface; for another example, when the question and answer area is closed, the content identifier of at least one knowledge content indexed by the target tag is displayed in the enlarged information display area 1403, such as the content identifier 14031 of the knowledge content 1, the content identifier 14032 of the knowledge content 2, etc. When the object performs a trigger operation on the target content identifier, if the target content identifier is the content identifier 14032, the knowledge content 2 is displayed.

[0180] As mentioned above, the knowledge question and answer interface includes an information display area and a question and answer area; the question and answer area can be understood as a sidebar in the knowledge question and answer interface, and the sidebar is a companion question and answer style that supports expanding and asking questions at any time. In other words, the embodiment of the present application supports hiding or restoring the display of the question and answer area in the knowledge question and answer interface to meet the browsing needs of the object for the content of different areas in the knowledge question and answer interface and enhance the object experience. Specifically, if the computer device detects a hiding operation on the question and answer area in the knowledge question and answer interface, the question and answer area is hidden in the knowledge question and answer interface, and the information display area is enlarged and displayed, such as full-screen enlargement display. After the information display area is enlarged and displayed in full screen, the display area of ​​the information display area is: the sum of the display area of ​​the information display area and the display area of ​​the question and answer area before the full-screen enlargement; in this case, the knowledge question and answer interface only includes the information display area, but not the question and answer area.

[0181] The hiding operation for the Q&A area may include any of the following: the operation in which the target tag included in the question message is triggered, and the operation in which the hidden component in the Q&A area is triggered. The interface schematic process of hiding the Q&A area when the target tag in the question message is triggered can be seen in Fig.14 For the hidden operation of the question and answer area, the interface schematic process of the operation of the hidden component in the question and answer area being triggered can be seen in Fig.15 ;like Fig.15 As shown, the question and answer area includes a hidden component 1501; when the hidden component 1501 is triggered, the question and answer area is closed or hidden in the knowledge question and answer interface, and the information display area is enlarged. Further, the closed or hidden question and answer area can also be restored to display; specifically, a question and answer display component 1502 is displayed in the knowledge question and answer interface, such as the question and answer display component 1502 can be specifically displayed in the information display area; when the object performs a trigger operation on the question and answer display component 1502, it indicates that the object may have the need to continue asking questions or view historical conversations. The computer device responds to the trigger operation of the question and answer display component, and displays (i.e., restores the display) the question and answer area in the knowledge question and answer interface, and correspondingly reduces the display area of ​​the information display area. Among them, the content displayed in the message display sub-area and the dialogue sub-area in the question and answer area that is restored to display is the same as the content displayed in the message display sub-area and the dialogue sub-area before the question and answer area is hidden, so as to facilitate the object to continue this round of knowledge questions and answers.

[0182] In addition, in order to facilitate the management and viewing of the knowledge content indexed by the first tag in the knowledge base by the subject, the embodiment of the present application provides the subject with a convenient content viewing function. For example, before selecting a target tag for the question information during the knowledge question answering process, the knowledge content indexed by each first tag is previewed using the content viewing function to help the subject better select the target tag. Specifically, the information display area in the knowledge question answering interface includes a search box, such as Fig.16 The search box 1601 shown in the figure; the object can enter a third tag in the search box 1601, and the computer device performs a tag index based on the third tag according to the third tag entered in the search box 1601, and outputs the content identifier of the knowledge content indexed by the third tag, specifically, the content identifier of the knowledge content of the third tag defined in the knowledge base is displayed in the information display area. The object can perform a trigger operation on the content identifier of the knowledge content, and the details of the knowledge content are displayed in the information display area, so that the object can preview the knowledge content. Among them, the third tag entered in the search box belongs to one of the at least one first tag.

[0183] In addition, the embodiment of the present application also supports the rapid positioning of the fragmented content used as a reference in the knowledge content based on the answer information, helping the object to deeply understand the relevant knowledge points of the answer information based on the fragmented content used as a reference, thereby improving the quality of the question and answer of the knowledge question and answer. Specifically, the answer information can be understood as being composed of at least one sub-information, and the reference sources of different sub-information may be different fragmented contents in the same knowledge content, or different fragmented contents in different knowledge contents. Based on this, the embodiment of the present application supports displaying the information source entrance of the information sub-information at the adjacent display position of the target sub-information in the answer information, and the information source entrance can be expressed as an icon or a serial number, etc. The target sub-information is any one of the at least one sub-information that constitutes the answer information. If the object wants to view the knowledge content and fragmented content referenced by the target sub-information, the object can click on the information source entrance, and at this time the computer device responds to the trigger operation of the object for the information source entrance, and displays the target knowledge content as a reference for generating the target sub-information, such as displaying the target knowledge content in the information display area. Among them, it supports marking and displaying the fragmented content that serves as a reference for generating target sub-information in the target knowledge content, helping the object to intuitively perceive the source location of the target sub-information in the knowledge content and avoid information comparison and search in the knowledge content.

[0184] A schematic diagram of quickly locating knowledge content and fragmented content based on information source entry can be seen in Fig.17 ;like Fig.17As shown, an answer message is displayed in the message display sub-area, and the answer information included in the answer message is composed of sub-information 1701, sub-information 1702 and sub-information 1703. An information source entry is displayed at the adjacent display position of each sub-message, such as an information source entry 1704 is displayed at the adjacent display position of sub-information 1701. When the object performs a trigger operation on the information source entry 1704, the computer device displays the target knowledge content used as a reference for generating the target sub-information in the information display area. For example, if the target knowledge content is knowledge content 1, the content identifier of the knowledge content 1 is displayed in the information display area, and the detailed content of the knowledge content 1 is displayed when the content identifier of the knowledge content 1 is triggered, or the detailed content of the knowledge content 1 is directly displayed in the information display area. Among them, when the detailed content of the knowledge content 1 is displayed in the information display area, the segmented content in the knowledge content 1 that is used as a reference for generating the target sub-information is annotated and displayed, such as the way of annotating and displaying is to deepen the background color of the segmented content, so as to facilitate the object to intuitively perceive and quickly locate the display position of the segmented content used as a reference in the knowledge content. It should be understood that the way of displaying the annotations for the slice content may also be other ways. Fig.17 The marking display method shown is only an example and does not limit the embodiments of the present application.

[0185] The above mainly introduces the knowledge question answering method proposed in the embodiment of the present application from the perspective of the interface. Fig.18 This paper introduces the overall background process of the knowledge question answering method. Fig.18 As shown, the knowledge question answering method is mainly implemented based on the knowledge base interactive system (such as the EleasticSearch system), which generally includes label screening and the establishment of label indexes, as well as label index knowledge content. Among them: ① Label screening and label index establishment include: the object is a knowledge content setting a first label, and after the computer device receives the first label; on the one hand, it will perform word segmentation processing on the first label (the specific implementation process of word segmentation processing can be referred to the aforementioned Figure 2The relevant content in the illustrated embodiment will not be elaborated here), so as to establish a multi-field and multi-dimensional inverted index of the first label in the knowledge base interactive system, aiming to be able to quickly implement label screening or label retrieval according to the input information edited by the object during the knowledge question and answer process, and accurately locate the second label related to the input information. On the other hand, an index relationship between the first label and the knowledge content will be established, so that the corresponding knowledge content can be indexed based on the first label to generate answer information. ② Label indexing includes: in the actual knowledge question and answer process, after the computer device receives the target label set by the object for the question information, it performs label indexing processing based on the target label and indexes at least one knowledge content; then, based on the content identifier of the knowledge content and the question information, recalls the segmented content that matches the question information, so that the computer device can generate answer information for the question information based on the segmented content. Among them, the label index and the calculation of the matching between the question information and the segmented content, etc., can be found in the aforementioned Figure 2 The relevant contents in the illustrated embodiment will not be described in detail here.

[0186] In summary, on the one hand, the embodiment of the present application supports the object to customize the first label for the knowledge content, and to label the knowledge content with a personalized label, so as to facilitate the object's understanding and management of the knowledge content in the knowledge base. On the other hand, when the object selects the target label for the question information, it supports the accurate and rapid positioning of the second label that the object may select through label screening, thereby improving the efficiency and flexibility of label selection. On the other hand, a label index is established between the first label and the knowledge content, so that when the answer information is generated, personalized answer information can be generated based on the knowledge content indexed by the target label, thereby improving the user's satisfaction with the answer information, and also reducing the search scope compared to content retrieval based on the entire knowledge base, thereby improving the efficiency of knowledge question and answer.

[0187] The method of the embodiment of the present application is described in detail above. In order to facilitate the above-mentioned scheme of the embodiment of the present application to be better implemented, the device of the embodiment of the present application is provided below accordingly. In the embodiment of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of an overall module or unit that includes the module or unit function.

[0188] Fig.19 A schematic diagram of the structure of a knowledge question answering device provided by an exemplary embodiment of the present application is shown; the knowledge question answering device can be used to perform Figure 2 and Fig.10Some or all of the steps in the method embodiment shown. Fig.19 , the device comprises the following units:

[0189] The display unit 1901 is used to display at least one first tag in the knowledge question and answer interface, each first tag is used to index at least one knowledge content from the knowledge base;

[0190] The processing unit 1902 is configured to receive a target tag selected for the question information, where the target tag is any one of the at least one first tag;

[0191] The processing unit 1902 is further used to output answer information corresponding to the question information in the knowledge question and answer interface, where the answer information is generated based on the question information and the knowledge content indexed by the target tag.

[0192] In one implementation, the display unit 1901 is used to display at least one first label in the knowledge question and answer interface, specifically to:

[0193] In response to a tag recall operation in the knowledge question and answer interface, a tag filter window is displayed; the tag recall operation includes any of the following: an input operation for a shortcut key, a trigger operation for a tag component, and a specified gesture operation;

[0194] At least one first tag is displayed in the tag filter window.

[0195] In one implementation, the processing unit 1902 is further configured to:

[0196] Receive input information edited in the knowledge question and answer interface;

[0197] In the tag filtering window, at least one first tag is updated and displayed as at least one second tag related to the input information; the second tag belongs to the at least one first tag.

[0198] In one implementation, the method is applied to a knowledge question and answer application, the knowledge question and answer application corresponds to a knowledge base interaction system, the knowledge base interaction system includes at least one index cluster and at least one label storage shard; the index cluster stores an address that can be routed to a sub-shard in the label storage shard, and one sub-shard is used to store a first label of knowledge content in the knowledge base corresponding to an object identifier for logging into the knowledge question and answer application;

[0199] The processing unit 1902 is further configured to:

[0200] Routing to a target index cluster in the knowledge base interaction system according to the object identifier, the target index cluster storing an address that can be routed to at least one reference sub-shard, the at least one reference sub-shard being a sub-shard storing a first label of knowledge content in the knowledge base corresponding to the object identifier;

[0201] According to the input information and the address of each reference sub-slice, a first tag related to the input information is obtained; and the first tag related to the input information is used as the second tag.

[0202] In one implementation, the knowledge question and answer interface includes at least two information source components, each of which corresponds to an information source, and the information source includes: the Internet and at least one knowledge base; the processing unit 1902 is further used to:

[0203] In response to a selection operation on the first information source component, displaying the first information source component as a selected state;

[0204] When the information source corresponding to the first information source component is a knowledge base, the step of displaying at least one first tag in the knowledge question and answer interface is triggered.

[0205] In one implementation, a round of knowledge question and answer includes at least one question and answer dialogue, and one question and answer dialogue includes a question message and an answer message, the question message includes question information, and the answer message includes answer information; during the first question and answer dialogue, the first information source component is in a selected state, and the processing unit 1902 is further used to:

[0206] After the first question-and-answer dialogue ends, in response to an information source switching operation performed in the knowledge question-and-answer interface, the first information source component is displayed as being in an unselected state; and,

[0207] The selected second information source component is displayed as a selected state, and the question information in the second question-and-answer dialogue is edited in the knowledge question-and-answer interface; the second information source component is different from the first information source component.

[0208] In one implementation, the selected information source component is represented as a target information source component. When the number of the selected target information source components is at least two, the first tag is a tag defined in the information source corresponding to the at least two target information source components. The processing unit 1902 is further configured to:

[0209] According to the priority setting operation for at least two target information source components, a priority identifier is displayed at an adjacent display position of each target information source component; the priority identifier is used to represent: the priority level of the knowledge content indexed by the target tag in the information source corresponding to the target information source component as a reference for generating answer information.

[0210] In one implementation, the knowledge question and answer interface includes a question and answer area, the question and answer area includes a message display sub-area, the message display sub-area is used to display question messages and answer messages generated by a round of knowledge question and answer, and the processing unit 1902 is further used to:

[0211] When a target tag included in the question message in the message display sub-area is triggered, displaying a content identifier of at least one knowledge content indexed by the target tag;

[0212] In response to a triggering operation on a target content identifier, the knowledge content corresponding to the target content identifier is displayed; the target content identifier is any one of the content identifiers of at least one knowledge content.

[0213] In one implementation, the knowledge question and answer interface further includes an information display area, which is used to display knowledge content in the knowledge base. The processing unit 1902 is further used to:

[0214] According to a hiding operation on the question and answer area, the question and answer area is hidden in the knowledge question and answer interface; the hiding operation includes any of the following: an operation in which a target tag included in the question message is triggered, and an operation in which a hidden component in the question and answer area is triggered;

[0215] Enlarge the information display area.

[0216] In one implementation, the question-and-answer area further includes a dialogue sub-area, which is used to receive question information and target tags; the knowledge question-and-answer interface includes a question-and-answer display component, and the processing unit 1902 is further used to:

[0217] In response to the triggering operation of the question and answer display component, the question and answer area is displayed in the knowledge question and answer interface; the content displayed in the message display sub-area and the dialogue sub-area of ​​the question and answer area is the same as before the question and answer area is hidden.

[0218] In one implementation, the information display area includes a search box, and the processing unit 1902 is further configured to:

[0219] According to the third tag input in the search box, the content identifier of the knowledge content defined with the third tag in the knowledge base is displayed in the information display area.

[0220] In one implementation, the answer information is composed of at least one sub-information; the processing unit 1902 is further configured to:

[0221] Displaying an information source entrance at a display position adjacent to a target sub-information, the target sub-information being any one of the at least one sub-information;

[0222] In response to a trigger operation on an information source entry, displaying target knowledge content as a reference for generating target sub-information;

[0223] In the target knowledge content, the fragmented content used as a reference for generating the target sub-information is annotated and displayed.

[0224] In one implementation, the method is applied to a knowledge question answering application; the processing unit 1902 is further configured to:

[0225] In response to a login operation for an object identifier in a knowledge question and answer application, a knowledge service interface is displayed; the knowledge service interface includes a question and answer trigger component;

[0226] When the question-and-answer trigger component is triggered, the knowledge question-and-answer interface is displayed.

[0227] In one implementation, the processing unit 1902 is further configured to:

[0228] Filter one or more knowledge contents with defined target tags from the knowledge base, where each knowledge content includes multiple fragment contents;

[0229] The matching degree of the question information and each fragment content included in each knowledge content is calculated to obtain the matching degree result corresponding to each fragment content included in each knowledge content; the matching degree result is used to indicate: the reliability of the fragment content as a reference for generating answer information corresponding to the question information;

[0230] Based on the target segment content whose matching result is greater than the matching threshold, answer information corresponding to the question information is generated.

[0231] According to one embodiment of the present application, Fig.19 The various units in the knowledge question and answer device shown can be individually or completely combined into one or several other units to form, or one (some) of the units can be further divided into multiple functionally smaller units to form, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application. The above-mentioned units are divided based on logical functions. In practical applications, the function of one unit can also be implemented by multiple units, or the functions of multiple units can be implemented by one unit. In other embodiments of the present application, the knowledge question and answer device may also include other units. In practical applications, these functions can also be implemented with the assistance of other units, and can be implemented by the collaboration of multiple units. According to another embodiment of the present application, it can be executed by running on a general computing device such as a computer that includes processing elements and storage elements such as a central processing unit (CPU), a random access storage medium (RAM), and a read-only storage medium (ROM). Figure 2 and Fig.10 A computer program (including program code) for each step involved in the corresponding method shown in FIG. Fig.19 The knowledge question answering device shown in and the knowledge question answering method of the embodiment of the present application are implemented. The computer program can be recorded on, for example, a computer readable recording medium, and loaded into the above-mentioned computing device through the computer readable recording medium and run therein.

[0232] In the embodiment of the present application, at least one first label is displayed in the knowledge question and answer interface, and each first label can index at least one knowledge content in the knowledge base; the first label is pre-defined by the user for the knowledge content in the knowledge base, such as defining the first label 1 and the first label 2 for the knowledge content 1, and defining the first label 1 for the knowledge content 2. In the process of editing the question information in the knowledge question and answer interface, it supports the user to select the target label for the edited question information from at least one first label. In this way, after receiving the target label and the question information, the computer device can quickly filter the knowledge content indexed by the target label from the knowledge base through the label index, and generate the answer information based on the knowledge content and the question information indexed by the target label. It can be seen that, compared with the generation of answer information for question information based on the massive knowledge content in the knowledge base, the embodiment of the present application supports the label index to quickly locate the specific knowledge content in the knowledge base as the information source of the answer information, avoiding problems such as equipment failure caused by information overload. On the other hand, the target label is selected by the user in a personalized manner, so that when different users select different target labels for the same question, the generated answer information is also different, which meets the user's requirements for personalized generation of answer information and improves the user experience. On the other hand, answer information only needs to be generated based on the knowledge content indexed by the target tag in the knowledge base, which greatly improves the speed and efficiency of answer generation, thereby improving the efficiency of knowledge question and answering and helping users improve work efficiency and personal learning effects.

[0233] Fig. 20 FIG. 1 shows a schematic diagram of a computer device provided by an exemplary embodiment of the present application. Fig. 20 , the computer device includes a processor 2001, a communication interface 2002 and a computer-readable storage medium 2003. The processor 2001, the communication interface 2002 and the computer-readable storage medium 2003 can be connected via a bus or other means. The communication interface 2002 is used to receive and send data. The computer-readable storage medium 2003 can be stored in the memory of the computer device, the computer-readable storage medium 2003 is used to store computer programs, and the processor 2001 is used to execute the computer programs stored in the computer-readable storage medium 2003. The processor 2001 (or CPU (Central Processing Unit)) is the computing core and control core of the computer device, which is suitable for implementing one or more computer programs, and is specifically suitable for loading and executing one or more computer programs to implement corresponding method flows or corresponding functions.

[0234] The embodiment of the present application also provides a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space that stores the processing system of the computer device. In addition, one or more computer programs suitable for being loaded and executed by the processor 2001 are also stored in the storage space. It should be noted that the computer-readable storage medium here can be a high-speed RAM memory, or a non-volatile memory, such as at least one disk storage; optionally, it can also be at least one computer-readable storage medium located away from the aforementioned processor.

[0235] In one embodiment, one or more computer programs are stored in a computer-readable storage medium; the processor 2001 loads and executes the one or more computer programs stored in the computer-readable storage medium to implement the corresponding steps in the above-mentioned knowledge question answering method embodiment; in a specific implementation, the one or more computer programs in the computer-readable storage medium are loaded by the processor 2001 and execute the steps of each embodiment of the present application:

[0236] Displaying at least one first tag in the knowledge question and answer interface, each first tag is used to index at least one knowledge content from the knowledge base;

[0237] receiving a target tag selected for the question information, the target tag being any one of the at least one first tag;

[0238] The answer information corresponding to the question information is output in the knowledge question and answer interface, and the answer information is generated based on the question information and the knowledge content indexed by the target tag.

[0239] In one implementation, one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and when executed to display at least one first label in the knowledge question and answer interface, the following steps are specifically performed:

[0240] In response to a tag recall operation in the knowledge question and answer interface, a tag filter window is displayed; the tag recall operation includes any of the following: an input operation for a shortcut key, a trigger operation for a tag component, and a specified gesture operation;

[0241] At least one first tag is displayed in the tag filter window.

[0242] In one implementation, one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and further perform the following steps:

[0243] Receive input information edited in the knowledge question and answer interface;

[0244] In the tag filtering window, at least one first tag is updated and displayed as at least one second tag related to the input information; the second tag belongs to the at least one first tag.

[0245] In one implementation, the method is applied to a knowledge question and answer application, the knowledge question and answer application corresponds to a knowledge base interaction system, the knowledge base interaction system includes at least one index cluster and at least one label storage shard; the index cluster stores an address that can be routed to a sub-shard in the label storage shard, and one sub-shard is used to store a first label of knowledge content in the knowledge base corresponding to an object identifier for logging into the knowledge question and answer application;

[0246] The one or more instructions in the computer readable storage medium are loaded by the processor 2001 and further perform the following steps:

[0247] Routing to a target index cluster in the knowledge base interaction system according to the object identifier, the target index cluster storing an address that can be routed to at least one reference sub-shard, the at least one reference sub-shard being a sub-shard storing a first label of knowledge content in the knowledge base corresponding to the object identifier;

[0248] According to the input information and the address of each reference sub-slice, a first tag related to the input information is obtained; and the first tag related to the input information is used as the second tag.

[0249] In one implementation, the knowledge question and answer interface includes at least two information source components, each of which corresponds to an information source, and the information source includes: the Internet and at least one knowledge base; one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and the following steps are also executed:

[0250] In response to a selection operation on the first information source component, displaying the first information source component as a selected state;

[0251] When the information source corresponding to the first information source component is a knowledge base, the step of displaying at least one first tag in the knowledge question and answer interface is triggered.

[0252] In one implementation, a round of knowledge question and answer includes at least one question and answer dialogue, and one question and answer dialogue includes a question message and an answer message, the question message includes question information, and the answer message includes answer information; during the first question and answer dialogue, the first information source component is in a selected state, and one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and the following steps are further executed:

[0253] After the first question-and-answer dialogue ends, in response to an information source switching operation performed in the knowledge question-and-answer interface, the first information source component is displayed as being in an unselected state; and,

[0254] The selected second information source component is displayed as a selected state, and the question information in the second question-and-answer dialogue is edited in the knowledge question-and-answer interface; the second information source component is different from the first information source component.

[0255] In one implementation, the selected information source component is represented as a target information source component. When the number of the selected target information source components is at least two, the first tag is a tag defined in the information source corresponding to the at least two target information source components. The one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and the following steps are further executed:

[0256] According to the priority setting operation for at least two target information source components, a priority identifier is displayed at an adjacent display position of each target information source component; the priority identifier is used to represent: the priority level of the knowledge content indexed by the target tag in the information source corresponding to the target information source component as a reference for generating answer information.

[0257] In one implementation, the knowledge question and answer interface includes a question and answer area, the question and answer area includes a message display sub-area, the message display sub-area is used to display question messages and answer messages generated by a round of knowledge question and answer, and one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and further perform the following steps:

[0258] When a target tag included in the question message in the message display sub-area is triggered, displaying a content identifier of at least one knowledge content indexed by the target tag;

[0259] In response to a triggering operation on a target content identifier, the knowledge content corresponding to the target content identifier is displayed; the target content identifier is any one of the content identifiers of at least one knowledge content.

[0260] In one implementation, the knowledge question and answer interface further includes an information display area, which is used to display knowledge content in the knowledge base. One or more instructions in the computer-readable storage medium are loaded by the processor 2001 and the following steps are also executed:

[0261] According to a hiding operation on the question and answer area, the question and answer area is hidden in the knowledge question and answer interface; the hiding operation includes any of the following: an operation in which a target tag included in the question message is triggered, and an operation in which a hidden component in the question and answer area is triggered;

[0262] Enlarge the information display area.

[0263] In one implementation, the question-and-answer area further includes a dialogue sub-area, which is used to receive question information and target tags; the knowledge question-and-answer interface includes a question-and-answer display component, and one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and further perform the following steps:

[0264] In response to the triggering operation of the question and answer display component, the question and answer area is displayed in the knowledge question and answer interface; the content displayed in the message display sub-area and the dialogue sub-area of ​​the question and answer area is the same as before the question and answer area is hidden.

[0265] In one implementation, the information display area includes a search box, and one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and further perform the following steps:

[0266] According to the third tag input in the search box, the content identifier of the knowledge content defined with the third tag in the knowledge base is displayed in the information display area.

[0267] In one implementation, the answer information is composed of at least one sub-information; one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and further perform the following steps:

[0268] Displaying an information source entrance at a display position adjacent to a target sub-information, the target sub-information being any one of the at least one sub-information;

[0269] In response to a trigger operation on an information source entry, displaying target knowledge content as a reference for generating target sub-information;

[0270] In the target knowledge content, the fragmented content used as a reference for generating the target sub-information is annotated and displayed.

[0271] In one implementation, the method is applied to a knowledge question answering application; one or more instructions in a computer-readable storage medium are loaded by a processor 2001 and further perform the following steps:

[0272] In response to a login operation for an object identifier in a knowledge question and answer application, a knowledge service interface is displayed; the knowledge service interface includes a question and answer trigger component;

[0273] When the question-and-answer trigger component is triggered, the knowledge question-and-answer interface is displayed.

[0274] In one implementation, one or more instructions in the computer-readable storage medium are loaded by the processor 2001 and further perform the following steps:

[0275] Filter one or more knowledge contents with defined target tags from the knowledge base, where each knowledge content includes multiple fragment contents;

[0276] The matching degree of the question information and each fragment content included in each knowledge content is calculated to obtain the matching degree result corresponding to each fragment content included in each knowledge content; the matching degree result is used to indicate: the reliability of the fragment content as a reference for generating answer information corresponding to the question information;

[0277] Based on the target segment content whose matching result is greater than the matching threshold, answer information corresponding to the question information is generated.

[0278] Based on the same inventive concept, the principles and beneficial effects of solving problems by the computer device provided in the embodiment of the present application are similar to the principles and beneficial effects of solving problems by the knowledge question and answer method in the method embodiment of the present application. Please refer to the principles and beneficial effects of the implementation of the method. For the sake of concise description, they will not be repeated here.

[0279] An embodiment of the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned knowledge question and answer method is implemented.

[0280] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0281] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes a computer program (one or more). When a computer program is loaded and executed on a computer device, the computer program executes the above-mentioned process or function of the embodiment of the present application. The computer device can be a general-purpose computer, a special-purpose computer, a computer network or other programmable devices. The computer program can be stored in a computer-readable storage medium or transmitted by a computer-readable storage medium. The computer program can be transmitted from a website site, a computer device, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer device, server or data center. The computer-readable storage medium can be any available medium that a computer device can access or a data storage device such as a server or a data center that contains one or more available media integrations. Available media can be magnetic media (e.g., floppy disk, hard disk, tape), optical media (e.g., DVD), or semiconductor media (e.g., solid-state drive (Solid State Disk, SSD)) and the like.

[0282] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A knowledge question answering method, characterized in that: include: Displaying at least one first tag in the knowledge question and answer interface, each of the first tags is used to index at least one knowledge content from the knowledge base; receiving a target tag selected for question information, wherein the target tag is any one of at least one of the first tags; Answer information corresponding to the question information is output in the knowledge question and answer interface, and the answer information is generated based on the question information and the knowledge content indexed by the target tag.

2. The method according to claim 1, characterized in that The displaying of at least one first label in the knowledge question and answer interface includes: In response to a tag recall operation in the knowledge question and answer interface, a tag filtering window is displayed; the tag recall operation includes any one of the following: an input operation for a shortcut key, a trigger operation for a tag component, and a specified gesture operation; At least one first tag is displayed in the tag filtering window.

3. The method according to claim 2, characterized in that The method further comprises: Receive input information edited in the knowledge question and answer interface; In the tag filtering window, at least one of the first tags is updated and displayed as at least one second tag related to the input information; the second tag belongs to at least one of the first tags.

4. The method according to claim 3, characterized in that The method is applied to a knowledge question and answer application, wherein the knowledge question and answer application corresponds to a knowledge base interaction system, and the knowledge base interaction system includes at least one index cluster and at least one label storage shard; the index cluster stores an address that can be routed to a sub-shard in the label storage shard, and one of the sub-shards is used to store a first label of the knowledge content in the knowledge base corresponding to the object identifier for logging into the knowledge question and answer application; After receiving the input information edited in the knowledge question and answer interface, the method further includes: Routing to a target index cluster in the knowledge base interaction system according to the object identifier, wherein the target index cluster stores an address that can be routed to at least one reference sub-shard, wherein the at least one reference sub-shard refers to a sub-shard storing a first label of the knowledge content in the knowledge base corresponding to the object identifier; According to the input information and the address of each reference sub-slice, a first tag related to the input information is acquired; and the first tag related to the input information is used as the second tag.

5. The method according to claim 1, characterized in that The knowledge question and answer interface includes at least two information source components, each of which corresponds to an information source, and the information source includes: the Internet and at least one of the knowledge bases; before displaying at least one first label in the knowledge question and answer interface, it also includes: In response to a selection operation on a first information source component, displaying the first information source component as a selected state; When the information source corresponding to the first information source component is the knowledge base, the step of displaying at least one first tag in the knowledge question and answer interface is triggered.

6. The method according to claim 5, characterized in that A round of knowledge question and answer includes at least one question and answer dialogue, and one question and answer dialogue includes a question message and an answer message, wherein the question message includes question information, and the answer message includes answer information; During the first question-and-answer dialogue, the first information source component is in a selected state, and the method further includes: After the first question-and-answer dialogue ends, in response to an information source switching operation performed in the knowledge question-and-answer interface, displaying the first information source component as an unselected state; as well as, The selected second information source component is displayed as a selected state, and the question information in the second question-and-answer dialogue is edited in the knowledge question-and-answer interface; the second information source component is different from the first information source component.

7. The method according to claim 5 or 6, characterized in that The selected information source component is represented as a target information source component, and when the number of the selected target information source components is at least two, the first label is a label defined in the information sources corresponding to at least two of the target information source components; The method further comprises: According to the priority setting operation for at least two of the target information source components, displaying a priority mark at an adjacent display position of each of the target information source components; The priority identifier is used to represent: the priority level of the knowledge content indexed by the target tag in the information source corresponding to the target information source component, which is used as a reference for generating answer information.

8. The method according to claim 1, characterized in that The knowledge question and answer interface includes a question and answer area, and the question and answer area includes a message display sub-area, and the message display sub-area is used to display question messages and answer messages generated by a round of knowledge question and answer. The method also includes: When the target tag included in the question message in the message display sub-area is triggered, displaying a content identifier of at least one of the knowledge contents indexed by the target tag; In response to a triggering operation on a target content identifier, the knowledge content corresponding to the target content identifier is displayed; the target content identifier is any one of the content identifiers of at least one of the knowledge contents.

9. The method according to claim 8, characterized in that The knowledge question and answer interface also includes an information display area, and the information display area is used to display the knowledge content in the knowledge base. The method also includes: According to a hiding operation for the question and answer area, hiding the question and answer area in the knowledge question and answer interface; the hiding operation includes any one of the following: an operation in which the target tag included in the question message is triggered, and an operation in which a hidden component in the question and answer area is triggered; The information display area is enlarged and displayed.

10. The method according to claim 9, characterized in that The question-and-answer area also includes a dialogue sub-area, and the dialogue sub-area is used to receive question information and target tags; The knowledge question and answer interface includes a question and answer display component, and the method further includes: In response to a triggering operation on the question and answer display component, displaying the question and answer area in the knowledge question and answer interface; The contents displayed in the message display sub-area and the conversation sub-area of ​​the question and answer area are the same as before the question and answer area is hidden.

11. The method according to claim 9, characterized in that The information display area includes a search box, and the method further includes: According to the third tag input in the search box, the content identifier of the knowledge content defined by the third tag in the knowledge base is displayed in the information display area.

12. The method according to any one of claims 1, 2-6 or 8-11, characterized in that: The answer information is composed of at least one sub-information; the method further comprises: Displaying an information source entrance at an adjacent display position of target sub-information, wherein the target sub-information is any one of at least one of the sub-information; In response to a trigger operation on the information source portal, displaying target knowledge content used as a reference for generating the target sub-information; In the target knowledge content, the fragment content used as a reference for generating the target sub-information is marked and displayed.

13. The method according to claim 1, characterized in that The method is applied to a knowledge question and answer application; before displaying at least one first label in the knowledge question and answer interface, the method further includes: In response to a login operation for an object identifier in the knowledge question and answer application, a knowledge service interface is displayed; the knowledge service interface includes a question and answer trigger component; When the question-and-answer trigger component is triggered, a knowledge question-and-answer interface is displayed.

14. The method according to claim 1, wherein: Before outputting the answer information corresponding to the question information in the knowledge question and answer interface, the method further includes: Filtering one or more knowledge contents defined with the target tag from the knowledge base, each of the knowledge contents including a plurality of fragment contents; Calculating the matching degree between the question information and each of the fragment contents included in each of the knowledge contents, and obtaining the matching degree result corresponding to each of the fragment contents included in each of the knowledge contents; the matching degree result is used to characterize: the reliability of the fragment contents as a reference for generating answer information corresponding to the question information; Based on the target segment content whose matching result is greater than a matching threshold, answer information corresponding to the question information is generated.

15. A knowledge question-answering device, characterized in that: include: A display unit, used to display at least one first tag in the knowledge question and answer interface, each of which is used to index at least one knowledge content from the knowledge base; a processing unit, configured to receive a target tag selected for question information, wherein the target tag is any one of at least one of the first tags; The processing unit is further used to output answer information corresponding to the question information in the knowledge question and answer interface, and the answer information is generated based on the question information and the knowledge content indexed by the target tag.

16. A computer device, characterized in that: include: a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, the knowledge question and answer method according to any one of claims 1 to 14 is implemented.

17. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer application, and when the computer application is executed, it implements the knowledge question and answer method according to any one of claims 1-14.

18. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the knowledge question and answer method as described in any one of claims 1 to 14 is implemented.

Images