Knowledge base processing method and apparatus, and computer device, medium and program product

Abstract

17. Abstract of description Provided in the embodiments of the present application are a knowledge base processing method and apparatus, and a computer device, a medium and a program product. The knowledge base processing method comprises: receiving a sharing credential of a shared knowledge base; in response to an access operation on the sharing credential, displaying the shared knowledge base, which is shared, wherein the shared knowledge base comprises at least one knowledge file; receiving question information for the shared knowledge base; and displaying answer information for the question information, wherein the answer information is obtained by means of performing answer generation on the question information on the basis of the knowledge file in the shared knowledge base. By means of the embodiments of the present application, the knowledge utilization rate of a knowledge base can be increased.

Description

Knowledge base processing methods, devices and computer equipment, media, and program products

[0001] This application claims priority to Chinese Patent Application No. 202411982705.8, filed on December 27, 2024, entitled "Method, Apparatus and Computer Equipment, Media and Program Product for Processing Knowledge Base", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of computer technology, specifically to a method for processing a knowledge base, a device for processing a knowledge base, a computer device, a computer-readable storage medium, and a computer program product. Background Technology

[0003] A knowledge base is a collection of knowledge files. In current knowledge base usage scenarios, knowledge bases are often private, meaning they are only accessible to their creator. This private nature results in low usage rates, making improving knowledge base utilization a current research hotspot. Summary of the Invention

[0004] This application provides a method, apparatus, computer equipment, medium, and program product for processing a knowledge base, which can improve the utilization rate of knowledge in the knowledge base.

[0005] On the one hand, embodiments of this application provide a method for processing a knowledge base, which includes:

[0006] Receive sharing credentials for the shared knowledge base;

[0007] In response to an access operation to the sharing credentials, the shared knowledge base being shared is displayed, which includes at least one knowledge file;

[0008] Receive questions about the shared knowledge base;

[0009] The system displays the answer to the question, which is generated by processing the question based on knowledge files in a shared knowledge base.

[0010] Accordingly, embodiments of this application provide a knowledge base processing apparatus, which includes:

[0011] The processing unit is used to receive the sharing credentials for the shared knowledge base;

[0012] The display unit is used to display the shared knowledge base being shared in response to an access operation on the sharing credential. The shared knowledge base includes at least one knowledge file.

[0013] The processing unit is used to receive query information for the shared knowledge base;

[0014] The display unit is also used to display the answer information to the question information, which is generated by processing the question information based on knowledge files in the shared knowledge base.

[0015] On the other hand, embodiments of this application also provide a method for processing a knowledge base, the method comprising:

[0016] In response to a login action from the sharing object, the shared knowledge base is displayed, which includes at least one knowledge file;

[0017] In response to a sharing operation on a shared knowledge base, the shared knowledge base is shared, enabling the recipient to answer questions based on the knowledge files in the shared knowledge base.

[0018] Accordingly, embodiments of this application provide a knowledge base processing apparatus, which includes:

[0019] The display unit is used to display the shared knowledge base of the sharing object in response to the login operation of the sharing object. The shared knowledge base includes at least one knowledge file.

[0020] The processing unit is used to respond to the sharing operation of the shared knowledge base, share the shared knowledge base, and enable the recipient to generate and process answers to the questions based on the knowledge files in the shared knowledge base.

[0021] Accordingly, embodiments of this application provide a computer device, which includes:

[0022] A processor is a tool for implementing computer programs.

[0023] A computer-readable storage medium storing a computer program adapted to be loaded by a processor and executed by the aforementioned method for processing the knowledge base.

[0024] Accordingly, embodiments of this application provide a computer-readable storage medium storing a computer program. When the computer program is read and executed by the processor of a computer device, the computer device performs the aforementioned knowledge base processing method.

[0025] Accordingly, this application provides a computer program product comprising 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 executes the computer program, causing the computer device to perform the aforementioned knowledge base processing method.

[0026] In this embodiment, a sharing credential for a shared knowledge base can be received. In response to an access operation to the sharing credential, the shared knowledge base can be displayed. Therefore, the shared knowledge base possesses sharing attributes and can be shared, thus improving its shareability. Furthermore, the recipient can conduct question-and-answer sessions based on the knowledge files in the shared knowledge base. Thus, this embodiment provides two methods for using a shared knowledge base: sharing and question-and-answer sessions, thereby increasing its utilization rate. Attached Figure Description

[0027] Figure 1 is a schematic diagram of a scenario for a knowledge base processing method provided in an embodiment of this application;

[0028] Figure 2 is a flowchart illustrating a knowledge base processing method provided in an embodiment of this application;

[0029] Figure 3 is a schematic diagram of the interface of a knowledge base list provided in an embodiment of this application;

[0030] Figure 4a is a schematic diagram of an interface for creating a blank shared knowledge base according to an embodiment of this application;

[0031] Figure 4b is a schematic diagram of an interface for adding local files in a shared knowledge base according to an embodiment of this application;

[0032] Figure 4c is a schematic diagram of an interface for adding knowledge files from a personal knowledge base to a shared knowledge base, according to an embodiment of this application.

[0033] Figure 4d is a schematic diagram of an interface for adding knowledge files from a personal knowledge base to a shared knowledge base, provided in an embodiment of this application.

[0034] Figure 5a is a schematic diagram of an interface for retrieving knowledge files by keywords in a shared knowledge base according to an embodiment of this application;

[0035] Figure 5b is a schematic diagram of an interface for editing file tags in a shared knowledge base provided in an embodiment of this application;

[0036] Figure 5c is a schematic diagram of an interface for retrieving knowledge files in a shared knowledge base by file tags, provided in an embodiment of this application.

[0037] Figure 6 is a schematic diagram of an interface for a sharing method of a shared knowledge base provided in an embodiment of this application;

[0038] Figure 7 is a schematic diagram of the interface for setting sharing permissions for a shared knowledge base according to an embodiment of this application;

[0039] Figure 8 is a schematic diagram of an interface for a shared knowledge base member management method provided in an embodiment of this application;

[0040] Figure 9 is a schematic diagram of a sharing interface for some knowledge files in a shared knowledge base provided in an embodiment of this application;

[0041] Figure 10 is a schematic diagram of an interface for sharing multiple shared knowledge bases according to an embodiment of this application;

[0042] Figure 11 is a flowchart illustrating another knowledge base processing method provided in an embodiment of this application;

[0043] Figure 12 is a schematic diagram of an access interface for a shared knowledge base provided in an embodiment of this application;

[0044] Figure 13 is a schematic diagram of a knowledge base interface provided in an embodiment of this application;

[0045] Figure 14a is a schematic diagram of a questioning method for all knowledge files in a shared knowledge base provided in an embodiment of this application;

[0046] Figure 14b is a schematic diagram of a questioning method for some knowledge files in a shared knowledge base provided in an embodiment of this application;

[0047] Figure 14c is a schematic diagram of another way of asking questions about some knowledge files in a shared knowledge base provided in an embodiment of this application;

[0048] Figure 14d is a schematic diagram of another way of asking questions about some knowledge files in a shared knowledge base provided in an embodiment of this application;

[0049] Figure 15a is a schematic diagram of an interface for answering information provided in an embodiment of this application;

[0050] Figure 15b is a schematic diagram of an interface for switching knowledge base answers provided in an embodiment of this application;

[0051] Figure 16 is a schematic diagram of an interface for viewing the source of answer information provided in an embodiment of this application;

[0052] Figure 17a is a schematic diagram of an interface where the number of questions can be limited before joining a shared knowledge base, according to an embodiment of this application.

[0053] Figure 17b is a schematic diagram of an interface where the answer display is restricted before joining a shared knowledge base, provided by an embodiment of this application;

[0054] Figure 17c is a schematic diagram of an interface where the source of answers is restricted before joining a shared knowledge base, according to an embodiment of this application.

[0055] Figure 18 is a schematic diagram of an interface for viewing knowledge files in a shared knowledge base according to an embodiment of this application;

[0056] Figure 19 is a schematic diagram of an interface for selecting a shared knowledge base for question answering provided in an embodiment of this application;

[0057] Figure 20 is a schematic diagram of the technical architecture of a knowledge base processing method provided in an embodiment of this application;

[0058] Figure 21 is a schematic diagram of a knowledge document parsing process provided in an embodiment of this application;

[0059] Figure 22 is a schematic diagram of the knowledge file update process in a knowledge base provided in an embodiment of this application;

[0060] Figure 23 is a schematic diagram of a knowledge base index reconstruction process provided in an embodiment of this application;

[0061] Figure 24 is a schematic diagram of a knowledge base-based question-and-answer process provided in an embodiment of this application;

[0062] Figure 25 is a schematic diagram of a vector index creation process provided in an embodiment of this application;

[0063] Figure 26 is a schematic diagram of an index organization provided in an embodiment of this application;

[0064] Figure 27 is a schematic diagram of a file fragment retrieval process provided in an embodiment of this application;

[0065] Figure 28 is a schematic diagram of the structure of a knowledge base processing device provided in an embodiment of this application;

[0066] Figure 29 is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0067] To better understand the technical solutions provided in the embodiments of this application, the technical terms used in the embodiments of this application will be introduced first:

[0068] I. Intelligent Knowledge Platform:

[0069] This application relates to an intelligent knowledge platform. An intelligent knowledge platform refers to an intelligent work platform that utilizes artificial intelligence technology to provide knowledge capabilities, including reading, searching, and writing knowledge. Specifically, reading knowledge refers to browsing knowledge within the intelligent knowledge platform, searching for knowledge refers to retrieving knowledge from the platform, and writing knowledge refers to using the platform to assist in knowledge writing. The intelligent knowledge platform aims to leverage artificial intelligence technology to improve the efficiency of information retrieval, browsing, and information accumulation.

[0070] Users of the intelligent knowledge platform can access the knowledge capabilities provided by the intelligent knowledge platform through the intelligent knowledge client. This application does not limit the type of intelligent knowledge client. For example, the intelligent knowledge client can be a browser, a web (World Wide Web) application, an application, or a mini-program.

[0071] II. Knowledge Base:

[0072] A knowledge base is a collection of knowledge files. Intelligent knowledge platforms provide the capability for intelligent question answering based on knowledge bases. This involves receiving questions from the knowledge base, utilizing the language processing capabilities of LLM (Large Language Model), and employing the knowledge content within the knowledge files to generate answers. LLM possesses powerful content creation capabilities, logical reasoning abilities in complex contexts, and reliable task execution capabilities. Intelligent question answering based on knowledge bases falls under the category of AI (Artificial Intelligence) search. AI search results differ from those returned by traditional search engines. Traditional search engines return search result lists, while AI search uses artificial intelligence technologies (such as LLM technology) to aggregate and organize results, directly providing answers.

[0073] Knowledge bases can include personal knowledge bases and shared knowledge bases. Personal knowledge bases are those that do not have sharing capabilities and are only accessible to their creator. Shared knowledge bases, on the other hand, are those that can be shared with others, allowing users to perform AI question-and-answer sessions based on the shared knowledge files.

[0074] III. Knowledge Documents:

[0075] A knowledge file refers to the basic unit of knowledge content in an intelligent knowledge platform. This application does not limit the file type of the knowledge file; the file type can include any of the following: text files (e.g., Text files, Word files, and PDF files), image files, video files, web page files, and internet articles, etc.

[0076] Based on the aforementioned technical terminology, this application provides a knowledge base processing method, as shown in Figure 1. This method offers innovative ways for sharing clients to create and share shared knowledge bases, and for clients receiving the knowledge base, it provides an innovative way to conduct question-and-answer sessions based on the content within the shared knowledge base. Shared knowledge bases can be created by uploading content assets from local or personal knowledge bases, and can be shared with others. In detail:

[0077] This application provides multiple paths for creating shared knowledge bases: for example, local content assets can be directly uploaded; alternatively, content can be saved to a personal knowledge base by browsing and saving web pages or searching and saving answers using AI on an intelligent knowledge platform, and then the personal knowledge base can be used to create shared knowledge bases in different fields or themes. Furthermore, the shared knowledge base also provides methods such as tag editing and filtering, and keyword search matching for quick management and retrieval of content assets within the shared knowledge base.

[0078] Furthermore, this application embodiment also allows for the sharing of created shared knowledge bases, enabling the sharing and dissemination of knowledge and experience. To protect the rights of the creators, when sharing, the sharing recipients can choose to set the visibility of knowledge files in the shared knowledge base to members of the shared knowledge base. If set to visible, members can open each knowledge file in the shared knowledge base to view its content and can also conduct question-and-answer sessions based on the shared knowledge base; if set to invisible, members can view the AI ​​summary of each knowledge file in the shared knowledge base, but cannot view the actual file content, yet can still conduct question-and-answer sessions based on the shared knowledge base. This protects the creators' ownership of the content assets in the shared knowledge base and, through large language model technology, provides the ability to share knowledge under this premise.

[0079] Furthermore, this application provides the ability to perform question-and-answer based on content in a shared knowledge base. Whether it's the creator, member, or recipient of the shared knowledge base, everyone can engage in question-and-answer based on it. This differs from AI search products that only provide general answers; this application can obtain more specialized and domain-specific answers based on the shared knowledge base, maximizing the value of its content.

[0080] Furthermore, this application embodiment utilizes a large language model based on a shared knowledge base to generate answers to questions, which differs from the question-and-answer function of traditional search engines. The difference lies in the fact that the answer information returned by this application embodiment is different from the search results returned by traditional search engines. Specifically, the search results returned by traditional search engines are search list pages (e.g., a list of web pages obtained through searching), while the answer information returned by this application embodiment is the result of aggregating and organizing knowledge content related to the question information in the shared knowledge base using a large language model. This application embodiment directly provides the answer, improving the accuracy of the answer information.

[0081] This application provides a knowledge base processing method, which can be executed by a computer device. The computer device can be a sharing terminal, which refers to the terminal device used by the sharing object (the terminal device can include, but is not limited to, any of the following: smartphone, tablet, laptop, desktop computer, smartwatch, smart home appliance, smart vehicle terminal, and aircraft, but is not limited to these, as described herein). The content of the knowledge base processing method includes: the creation method of the shared knowledge base, the retrieval method of the shared knowledge base, the management method of the shared knowledge base, and the sharing method of the shared knowledge base. As shown in Figure 2, the knowledge base processing method may include, but is not limited to, the following steps S201-S202:

[0082] S201, in response to the login operation of the sharing object, display the shared knowledge base of the sharing object, the shared knowledge base including at least one knowledge file.

[0083] In step S201, in response to the login operation of the sharing object in the sharing object's smart knowledge client, the sharing object's shared knowledge base can be displayed. Any shared knowledge base displayed can include a shared knowledge base created by the sharing object or a shared knowledge base joined by the sharing object. Creating a shared knowledge base means that the sharing object is the creator of the shared knowledge base, and joining a shared knowledge base means that the sharing object becomes a member of the shared knowledge base.

[0084] As shown in Figure 3, after logging into the smart knowledge client, the sharing object can display its knowledge base interface 30 in response to the triggering operation of the knowledge base entry on the homepage of the smart knowledge client. This embodiment does not limit the method by which the sharing object logs into its smart knowledge client; for example, it can log in by entering an account and password, by entering an SMS verification code, or by scanning a QR code, etc. The sharing object's knowledge base interface 30 can include a knowledge base list 301, which can include various knowledge bases of the sharing object. These can include the sharing object's personal knowledge base (e.g., the personal knowledge base named "My Knowledge Base" in Figure 3), shared knowledge bases created by the sharing object (e.g., the shared knowledge base named "Physics Knowledge Collection" in Figure 3), and shared knowledge bases joined by the sharing object (e.g., the shared knowledge base named "Drawing Tutorials," the shared knowledge base named "Photography Techniques," and the shared knowledge base named "Reference Weekly" in Figure 3). The embodiments of this application do not limit the operation method of triggering the operation. For example, the triggering operation may include a single click operation, a double click operation, and a long press operation, etc., which will be described here.

[0085] Here, we will first introduce how to create a shared knowledge base. The process of creating a shared knowledge base may include: creating a blank shared knowledge base, and adding knowledge files to the blank shared knowledge base. The sources of the added knowledge files may include at least one of the following: local files on the sharing object's terminal device, or knowledge files in the sharing object's personal knowledge base.

[0086] During the creation of a shared knowledge base, the process of creating a blank shared knowledge base is shown in Figure 4a. The knowledge base interface 30 can include a shared knowledge base creation entry 302. In response to the triggering operation of the creation entry 302, a shared knowledge base creation window 40 can be displayed. In the creation window 40, the knowledge base information of the shared knowledge base can be customized. The knowledge base information can include, for example, the knowledge base name, the knowledge base cover, and the knowledge base description. As shown in Figure 4a, a shared knowledge base named "Collection of Physics Knowledge" is created. The knowledge base cover is an image containing the text "Physics," and the knowledge base description is "covering basic concepts such as mechanics, electromagnetism, and optics...".

[0087] The process of adding local files to a shared knowledge base during its creation is shown in Figure 4b. The created shared knowledge base (e.g., the one named "Collection of Physics Knowledge" in Figure 4b) is displayed in the knowledge base display area 303 of the knowledge base interface 30. The knowledge base display area 303 may include a knowledge file import control 304. In response to a trigger operation on the knowledge file import control 304, an import path selection window 41 can be displayed. This window 41 may include controls for selecting between two import paths: local file import and personal knowledge base import. If local file import is selected, a local file window 42 can be displayed, allowing selection of local files to import into the created shared knowledge base. For example, Figure 4b shows the selection of local files named "A Brief Discussion of Quantum Mechanics" and "Introduction to Quantum Mechanics" to import into the created shared knowledge base.

[0088] During the creation of the shared knowledge base, the process of adding knowledge files from a personal knowledge base to the shared knowledge base is shown in Figure 4c. In response to the triggering operation of the knowledge file import control 304, an import path selection window 41 can be displayed. This window 41 can include selection controls for two import paths: local file import and personal knowledge base import. If personal knowledge base import is selected, a file list 43 of the personal knowledge base can be displayed. Knowledge files from the personal knowledge base can be selected from this list to import into the created shared knowledge base. For example, in Figure 4c, a local file named "Quantum Physicist" and a knowledge file named "A Brief Discussion of the History of Quantum Physics" from the personal knowledge base are selected for import into the created shared knowledge base. Furthermore, as shown in Figure 4d, when selecting knowledge files from the shared knowledge base to import from the file list 43 of the personal knowledge base, keyword search can be used to quickly filter the knowledge files to be imported within the personal knowledge base. For example, entering the keyword "quantum" in Figure 4d can quickly locate the knowledge files to import, improving the import speed.

[0089] Based on the above introduction to the creation methods of shared knowledge bases, it is easy to see that the sources of knowledge files in shared knowledge bases are very extensive, including local files uploaded locally and knowledge files in personal knowledge bases. Knowledge files in personal knowledge bases are usually collected web pages and collected AI search Q&A, etc. By importing knowledge files from multiple sources, the rapid creation of multimodal shared knowledge bases is achieved.

[0090] Next, we will introduce the retrieval methods for knowledge files in the shared knowledge base. Retrieval methods for knowledge files in the shared knowledge base can include either keyword retrieval or tag retrieval. Keyword retrieval refers to quickly searching for knowledge files in the shared knowledge base that match keywords. Knowledge files matching keywords can be those whose file names contain the keywords. For example, the keyword retrieval method is shown in Figure 5a: the knowledge base display area 303 displays the created shared knowledge bases (e.g., the shared knowledge base named "Collection of Physics Knowledge" in Figure 5a). The knowledge base display area 303 may also contain a search control 305. In response to the triggering operation of the search control 305, a search box 306 can be displayed. Keywords can be entered in the search box 306 (e.g., the keyword entered in Figure 5a is "Mechanics"), and knowledge files matching the keywords can be retrieved from the shared knowledge base (e.g., the knowledge file named "A Brief Discussion of Quantum Mechanics" and the knowledge file named "Introduction to Quantum Mechanics" in Figure 5a).

[0091] Tag retrieval refers to the rapid retrieval of knowledge files in a shared knowledge base that match the input file tags. Each knowledge file in the shared knowledge base has file tags. A knowledge file matching the input file tags can have the same file tags as the input file tags. These file tags can be obtained through content recognition during the import process (e.g., when the knowledge file in the shared knowledge base is imported from a local file), or they can be user-defined (e.g., when the knowledge file in the shared knowledge base is imported from a personal knowledge base). The file tags of knowledge files can be edited. For example, in Figure 5b, the file tag of a knowledge file named "Quantum Physicist" can be changed from "Quantum Physics" to "Physicist". The tag retrieval method is shown in Figure 5c: The knowledge base display area 303 displays the created shared knowledge base (for example, the shared knowledge base named "Collection of Physics Knowledge" in Figure 5c). The knowledge base display area 303 may also contain a search control 305. In response to the trigger operation of the search control 305, a search box 306 can be displayed. File tags can be entered in the search box 306 (for example, the file tag entered in Figure 5c is "Quantum Mechanics"). Knowledge files that match the file tags can be retrieved in the shared knowledge base (for example, the knowledge file named "A Brief Discussion of Quantum Mechanics" and the knowledge file named "Introduction to Quantum Mechanics" in Figure 5c).

[0092] Based on the above introduction to the retrieval methods for knowledge files in the shared knowledge base, it is easy to see that by using keywords or file tags, knowledge files that match the keywords or file tags can be quickly located in the shared knowledge base, enabling rapid retrieval of knowledge files and improving the efficiency of knowledge file retrieval in the shared knowledge base.

[0093] S202, in response to the sharing operation of the shared knowledge base, the shared knowledge base is shared, so that the receiving end can generate and process answers to the question information based on the knowledge files in the shared knowledge base.

[0094] In step S202, in response to a sharing operation on the shared knowledge base, the shared knowledge base can be shared. After sharing, the recipient can conduct question-and-answer sessions based on the shared knowledge base; that is, the recipient can ask questions about the shared knowledge base and generate answers based on the knowledge files in the shared knowledge base. The shared knowledge base to which the sharing operation is performed can include shared knowledge bases created by the sharing object or shared knowledge bases joined by the sharing object. In other words, the sharing object can share shared knowledge bases it has created or shared knowledge bases it has joined. By sharing shared knowledge bases, knowledge sharing can be achieved. Conducting question-and-answer sessions based on the shared knowledge base can improve the reuse rate of the shared knowledge and enhance the value of knowledge sharing.

[0095] The following describes the sharing methods for shared knowledge bases. As shown in Figure 6, the knowledge base display area 303 displays created shared knowledge bases (for example, a shared knowledge base named "Collection of Physics Knowledge" in Figure 6). The knowledge base display area 303 may also contain a sharing control 307. Sharing a shared knowledge base refers to triggering the sharing control 307. In response to triggering the sharing control 307, a sharing settings window 60 is displayed. Sharing settings can be configured in the sharing settings window 60, and sharing can proceed according to the settings. The sharing settings may include at least one of the following: sharing method settings and sharing permission settings. In detail:

[0096] For sharing methods, the shared knowledge base can be shared using either the following methods: link sharing and image code (e.g., QR code, barcode, etc.) sharing. As shown in Figure 6, selecting the button named "Copy Link" indicates that link sharing is selected as the sharing method for the shared knowledge base. The sharing recipient can send the shared knowledge base link to the recipient sharing the knowledge base. The recipient who receives the link can view the shared knowledge base and engage in Q&A based on the shared knowledge base. Selecting the button named "Generate QR Code" indicates that image code sharing is selected as the sharing method for the shared knowledge base. The sharing recipient can send the generated QR code to the recipient sharing the knowledge base. The recipient who receives the QR code can scan the QR code to view the shared knowledge base and engage in Q&A based on the shared knowledge base. As can be seen, there are various ways to share a shared knowledge base. The recipient can choose the sharing method as needed. For example, in offline sharing scenarios, a graphic code sharing method can be selected, which can be printed and has higher sharing efficiency. In online sharing scenarios, a link sharing method can be selected, which is very convenient to open. Therefore, setting multiple optional sharing methods helps to improve the convenience of sharing a shared knowledge base.

[0097] Regarding sharing permission settings, the system can accept settings for sharing permissions on a shared knowledge base. These permissions can include at least one of the following: visibility permissions for knowledge files within the shared knowledge base, permission to join the shared knowledge base, and permission to hide the shared knowledge base. Specifically:

[0098] (1) Setting the visibility permission for knowledge files in a shared knowledge base refers to setting whether the knowledge files in the shared knowledge base are visible or invisible after sharing. If set to visible, the recipient can view the file content of the knowledge files in the shared knowledge base. If set to invisible, the recipient can only view the file information (e.g., file information may include file name, file tags, and file type) and file summary of the knowledge files in the shared knowledge base, but cannot view the file content of the knowledge files in the shared knowledge base. However, regardless of whether the knowledge files in the shared knowledge base are set to visible or invisible after sharing, it does not affect the Q&A function based on the shared knowledge base after sharing. (2) Setting the access permission for joining a shared knowledge base refers to setting whether joining the shared knowledge base requires confirmation or not. If joining the shared knowledge base requires confirmation, the recipient's application to join the shared knowledge base needs the consent of the creator or administrator of the shared knowledge base. If joining the shared knowledge base does not require confirmation, the recipient can directly join the shared knowledge base without confirmation. (3) Setting hidden permissions for a shared knowledge base refers to setting it as a private or non-private database. A private database is one that only the sharing recipient can access, while a non-private database is one that both the sharing recipient and the recipient can access. Accessibility here means that the recipient can use the question-and-answer function based on the shared knowledge base, view the knowledge files within the shared knowledge base, and view the knowledge base information. It's easy to see that setting a shared knowledge base as a private database before sharing is equivalent to setting it as a protected personal knowledge base. Even if the link or image code is illegally leaked, those other than the sharing recipient cannot access the shared knowledge base, thus improving its security. Optionally, setting a shared knowledge base as a private database can also be done after a certain period of time since sharing, which effectively limits the access time and also improves its security.

[0099] The sharing permission settings are shown in Figure 7. The sharing settings window 60 can include a sharing permission settings entry 601. In response to triggering the sharing permission settings entry 601, a sharing permission settings window 70 can be displayed. This window 70 can include switches for setting the visibility permission of knowledge files in the shared knowledge base (701), the joining permission of the shared knowledge base (702), and the hiding permission of the shared knowledge base (703). Through these switches, the sharing permissions of the shared knowledge base can be configured. For example, enabling the visibility permission setting switch 701 for knowledge files in the shared knowledge base will make the knowledge files invisible after sharing. Similarly, enabling the joining permission setting switch requires the creator or administrator of the shared knowledge base to approve any joining request from the recipient. Furthermore, enabling the hiding permission setting switch will set the shared knowledge base as a private knowledge base.

[0100] Based on the above introduction to the methods of setting sharing permissions, it is easy to see that providing the ability to set sharing permissions essentially provides the right to protect the shared knowledge base, allowing the sharing recipient to protect the shared knowledge base through sharing permission settings; for example, protecting the knowledge files in the shared knowledge base from being visible to the sharing recipient, protecting the shared knowledge base from being illegally added, etc. In this way, the knowledge content assets in the shared knowledge base are protected, which can improve the security of the shared knowledge base.

[0101] Optionally, after the shared knowledge base is shared, it can receive member management operations, meaning the sharing object can manage the members of the shared knowledge base through member management operations. Member management operations can include at least one of the following: member addition management operations, member role management operations, and member deletion operations. Specifically, member addition management operations refer to processing member addition requests to the shared knowledge base; member role management operations refer to managing the roles of members in the shared knowledge base; and member deletion operations refer to deleting members from the shared knowledge base.

[0102] Figure 8 illustrates the member management method for a shared knowledge base. The knowledge base display area 303 shows created shared knowledge bases (e.g., a shared knowledge base named "Physics Knowledge Collection" in Figure 8). The knowledge base display area 303 may also include a member management entry point 308. In response to a trigger operation on the member management entry point 308, a knowledge base member management window 80 is displayed. The knowledge base member management window 80 may include a member joining management option 801 (e.g., an option named "Application" in Figure 8) and an already joined member management option 802 (e.g., an option named "Member" in Figure 8). The sub-settings window 803 under the member joining management option 801 can process member joining applications for the shared knowledge base, such as approving or rejecting the application. The sub-settings window 804 under the "Members Management" option 802 allows for the management of roles for members who have joined the shared knowledge base. For example, the option named "Member" can be used to set a member to the role of a regular member, or the option named "Administrator" can be used to set a member to the role of an administrator. Both regular members and administrators have the right to view the shared knowledge base and ask questions based on it. The difference between administrators and regular members is that administrators can edit the shared knowledge base, but regular members cannot. Editing the shared knowledge base here refers to file management (e.g., adding, deleting, and modifying file information such as title, tags, and type). The sub-settings window 804 under the "Members Management" option 802 also allows for the deletion of members from the shared knowledge base using the "Remove" option.

[0103] Based on the above introduction to the member management methods of shared knowledge bases, it can be seen that providing the ability to manage members of a shared knowledge base is essentially providing the right to protect the shared knowledge base, allowing sharing objects to protect the shared knowledge base through member management operations; for example, deleting untrusted members, rejecting untrusted joining requests, etc.; thus protecting the knowledge content assets in the shared knowledge base can improve the security of the shared knowledge base.

[0104] The above step S202 describes the sharing method for the entire shared knowledge base. Optionally, some knowledge files in the shared knowledge base can be shared. Specifically, it can receive a file selection operation performed on at least one knowledge file in the shared knowledge base, and in response to the sharing operation on the shared knowledge base, it can share the knowledge file selected by the file selection operation in the shared knowledge base, so that the receiving end can generate and process answers to questions based on the knowledge file selected by the file selection operation.

[0105] As shown in Figure 9, the knowledge base display area 303 displays created shared knowledge bases (e.g., a shared knowledge base named "Collection of Physics Knowledge" in Figure 9). The knowledge base display area 303 may also contain a sharing control 307. In response to a trigger operation on the sharing control 307, a sharing file selection window 90 is displayed, where file selection operations can be performed. For example, the selected knowledge files are a file named "A Brief Introduction to Quantum Mechanics" and a file named "Introduction to Quantum Mechanics." The sharing file selection window 90 may also include a sharing confirmation option (e.g., an option named "Confirm Share" in Figure 9). Sharing a shared knowledge base refers to triggering the sharing confirmation option. In response to this trigger operation, a sharing settings window 60 is displayed, where sharing settings can be made and the sharing can proceed according to the settings. By selecting and sharing specific knowledge files from a shared knowledge base, sharing can be achieved at the knowledge file level, better meeting the diverse sharing needs of different users and thus enhancing the sharing experience.

[0106] The above step S202 describes a method for sharing a single shared knowledge base. Optionally, multiple shared knowledge bases can be shared simultaneously. Specifically, there can be multiple shared knowledge bases, each of which can include at least one knowledge file. Each shared knowledge base can be a created or joined shared knowledge base. The system can receive knowledge base selection operations performed on multiple shared knowledge bases, and in response to a sharing operation on the selected shared knowledge base, it can share the selected shared knowledge base, enabling the recipient to generate answers to questions based on the knowledge files in the selected shared knowledge base.

[0107] As shown in Figure 10, the knowledge base interface 30 may include a list of shared knowledge bases 309 for sharing objects. This list may include shared knowledge bases created by the sharing object and shared knowledge bases joined by the sharing object. It may also include a shared knowledge base joint sharing entry 310 (e.g., a control named "Shared Knowledge Base Joint Sharing" in Figure 10). In response to a trigger operation on the joint sharing entry 310, a knowledge base selection window 100 can be displayed. A knowledge base selection operation can be performed in the knowledge base selection window 100. For example, the shared knowledge base selected in the knowledge base selection operation could be a knowledge file created by the sharing object with the name "Physics Knowledge Collection" and a shared knowledge base joined by the sharing object with the name "Reference Weekly". The knowledge base selection window 100 may also include a share confirmation option (e.g., an option named "Confirm Share" in Figure 10). The share operation of the selected shared knowledge base refers to the triggering operation of the share confirmation option. In response to this triggering operation, the share settings window 60 is displayed, where sharing settings can be configured and shared according to the settings. By jointly sharing multiple shared knowledge bases, the combined value and usage rate of these knowledge bases can be increased.

[0108] In addition to receiving knowledge base selection operations performed on multiple shared knowledge bases, it can also receive file selection operations on the shared knowledge bases selected in the knowledge base selection operation. That is, after selecting the shared knowledge base to be shared, the knowledge files to be shared can also be selected in the shared knowledge base to be shared. This can further enhance the freedom of sharing shared knowledge bases, as well as the value and rate of joint use of knowledge files across knowledge bases.

[0109] This application provides an efficient method for creating and managing a shared knowledge base with multiple sources (i.e., multiple sources of knowledge files). Content can be created through various methods such as local uploads, web browsing, and AI-powered question-and-answer search. Files can be retrieved within the shared knowledge base using tags and keywords, enabling rapid construction and efficient management of a multimodal shared knowledge base. This application provides an innovative solution for knowledge content sharing and dissemination, breaking down barriers between individuals and shared knowledge bases, increasing the flow and sharing of information and knowledge, enhancing the dissemination value and influence of high-quality content through shared knowledge bases, and improving the utilization rate of shared knowledge bases.

[0110] This application provides a knowledge base processing method, which can be executed by a computer device. The computer device is the sharing end, referring to the terminal device used by any object (which can be represented as a first object) in the shared knowledge base. The knowledge base processing method includes: a question-and-answer method based on the shared knowledge base, a viewing method for the shared knowledge base, and a management method for the shared knowledge base. As shown in Figure 11, the knowledge base processing method may include, but is not limited to, the following steps S1101-S1104:

[0111] S1101, Receive the sharing credentials for the shared knowledge base.

[0112] S1102, in response to an access operation on the sharing credential, the shared knowledge base being shared is displayed, the shared knowledge base including at least one knowledge file.

[0113] In steps S1101-S1102, the terminal device used by the first object sharing the shared knowledge base can receive a sharing credential for the shared knowledge base sent by the sharing object. The sharing credential may include, for example, a sharing link or a sharing graphic code. In response to an access operation to the sharing credential, an access interface for the shared knowledge base can be displayed. The access interface may display the shared knowledge base being shared, and the shared knowledge base may include at least one knowledge file. The access operation to the sharing credential may include, for example, accessing a sharing link (e.g., receiving a sharing link in an instant messaging client and opening the sharing link in the instant messaging client, or pasting the sharing link into a browser, or pasting the sharing link into the first object's smart knowledge client and opening it), scanning a sharing graphic code, or recognizing a sharing graphic code.

[0114] Furthermore, embodiments of this application can support sharing the shared knowledge base with various types of terminal devices. In other words, embodiments of this application can support various types of terminal devices to display the shared knowledge base by accessing a link or scanning a graphic code. For example, as shown in Figure 12(a), the terminal device used by the first object can be a mobile device (e.g., a smartphone or tablet). The mobile device used by the first object can display the access interface 120 of the shared knowledge base by accessing the sharing link. The access interface 120 displays the shared knowledge base (e.g., the shared knowledge base named "Physical Knowledge Collection" in Figure 12(a)). Alternatively, as shown in Figure 12(b), the terminal device used by the first object can be a PC (Personal Computer) (e.g., a laptop or desktop computer). The PC used by the first object can display the access interface 120 of the shared knowledge base by recognizing the shared graphic code (e.g., the graphic code can be a QR code). The access interface 120 displays the shared knowledge base (e.g., the shared knowledge base named "Collection of Physics Knowledge" in Figure 12(b)). When displaying the shared knowledge base, the specific information displayed is the knowledge base information of the shared knowledge base, as well as the file information of the knowledge files in the shared knowledge base.

[0115] Taking the display of a shared knowledge base access interface on a PC as an example, as shown in Figure 13, the shared knowledge base access interface 120 may include a viewing option 1201 (e.g., an option named "View Knowledge Base" in Figure 13). In response to the triggering operation of the shared knowledge base viewing option 1201, the knowledge base interface 130 of the first object can be displayed. The knowledge base interface 130 of the first object may include a knowledge base list 1301 of the first object, a knowledge base display area 1302 of the first object, and a knowledge Q&A area 1303 of the first object. The knowledge base interface 130 of the first object is the interface in the first object's intelligent knowledge client. Displaying the knowledge base interface 130 of the first object requires the first object to be logged into the first object's intelligent knowledge client. That is, if the first object is logged into the first object's intelligent knowledge client, the knowledge base interface 130 of the first object can be displayed in the first object's intelligent knowledge client in response to the triggering operation of the viewing option 1201 of the shared knowledge base. If the first object is not logged into the first object's intelligent knowledge client, the first object will be prompted to log in to the intelligent knowledge client in response to the triggering operation of the viewing option 1201 of the shared knowledge base. After the first object logs in to the first object's intelligent knowledge client, the knowledge base interface 130 of the first object will be displayed in the first object's intelligent knowledge client.

[0116] The knowledge base list 1301 may include various knowledge bases of the first object, including the first object's personal knowledge base (e.g., the personal knowledge base named "My Knowledge Base" in Figure 13), shared knowledge bases created by the first object (e.g., the shared knowledge base named "Photography Techniques" in Figure 13), and shared knowledge bases joined by the first object (e.g., the shared knowledge base named "Drawing Tutorials" in Figure 13). The first object's knowledge base display area 1302 displays shared knowledge bases (e.g., the shared knowledge base named "Collection of Physics Knowledge" in Figure 13); the first object's knowledge base display area 1302 may also include a join request control 1304 for shared knowledge bases (e.g., a control named "Join Knowledge Base" in Figure 13), which, in response to the triggering operation of the join request control 1304, allows the first object to apply to join the shared knowledge base; after joining the shared knowledge base, the shared knowledge base can become a shared knowledge base joined by the first object. The knowledge question and answer area 1303 can be used for question and answer based on the shared knowledge base. As described above, the knowledge base interface 130 of the first object can only be triggered and displayed after the first object logs in to the first object's smart knowledge client. That is to say, the first object can only conduct question and answer based on the shared knowledge base after logging in to the first object's smart knowledge client.

[0117] S1103, Receive query information for the shared knowledge base.

[0118] In step S1103, the question information for the shared knowledge base can be targeted at the entire shared knowledge base, meaning it can receive question information for all knowledge files in the shared knowledge base. As shown in Figure 14a, the knowledge question and answer area 1303 can include a question box 1305, where question information can be directly entered (for example, the question information in Figure 14a, "What theories does this collection of physics knowledge mainly involve?"), indicating that the question information is targeted at all knowledge files in the shared knowledge base.

[0119] Alternatively, questions for a shared knowledge base can be targeted at a specific subset of knowledge files within the shared knowledge base; that is, questions can be received for a specified subset of knowledge files in the shared knowledge base. The method of specification can include at least one of the following: specification based on file tags, specification based on keywords, and specification based on file type. Where:

[0120] (1) File tag-based specification refers to receiving input file tags and question information. The question information is for a portion of knowledge files in the shared knowledge base whose file tags are the input file tags. As shown in Figure 14b, the knowledge question and answer area 1303 may include a question box 1305. After entering a tag call symbol (e.g., the tag call symbol may be "#" in Figure 14b) in the question box 1305, a file tag list 140 can be displayed. The file tag list 140 may include the file tags of various knowledge files in the shared knowledge base (e.g., "quantum mechanics" and "quantum physics" in Figure 14b). In response to the file tag selection operation in the file tag list 140, the file tag selected by the file tag selection operation can be displayed in the question box 1305 (e.g., the selected file tag is "quantum mechanics" in Figure 14b). Question information can be entered after the selected file tag (e.g., the question information "What is quantum mechanics?" in Figure 14b), indicating that the question information is for the knowledge file in the shared knowledge base whose file tag is "quantum mechanics".

[0121] (2) Keyword-based specification refers to receiving input keywords and question information. The question information is for a portion of the knowledge file in the shared knowledge base whose file information and / or file content contains the input keywords. As shown in Figure 14c, the knowledge question and answer area 1303 may include a question box 1305. After entering the keyword start symbol (e.g., the keyword start symbol can be "*" in Figure 14c) in the question box 1305, the keyword can be entered (e.g., the keyword entered in Figure 14c is "physicist"). After entering the keyword, the keyword end symbol can be entered (e.g., the keyword end symbol can be "*" in Figure 14c). After entering the keyword end symbol, the question information can be entered (e.g., the question information "What are some quantum physicists?" in Figure 14c), indicating that the question information is for a knowledge file in the shared knowledge base whose file information and / or file content contains "physicist".

[0122] (3) File type specification refers to receiving the input file type and question information. The question information is for a portion of the knowledge files in the shared knowledge base whose file type is the input file type. As shown in Figure 14d, the knowledge question and answer area 1303 may include a question box 1305. After entering a file type call symbol (e.g., the file type call symbol can be "@" in Figure 14d) in the question box 1305, a file type list 141 can be displayed. The file tag list 141 may include the file types of various knowledge files in the shared knowledge base (e.g., "Word file", "PDF file", and "Internet article" in Figure 14d). In response to the file type selection operation in the file type list 141, the file type selected by the file type selection operation can be displayed in the question box 1305 (e.g., the selected file type is "Internet article" in Figure 14d). Question information can be entered after the selected file type (e.g., the question information "Outstanding representatives in the history of quantum physics" in Figure 14d), indicating that the question information is for the knowledge file in the shared knowledge base whose file type is "Internet article".

[0123] Based on the above introduction to the questioning methods of the shared knowledge base, it is easy to see that questions can be asked about the entire shared knowledge base, or about specific knowledge files within the shared knowledge base. The methods for specifying specific knowledge files are diverse, which enhances the flexibility of asking questions about the shared knowledge base and the flexibility of using the shared knowledge base.

[0124] S1104 Displays the answer to the question, which is generated by processing the question based on knowledge files in the shared knowledge base.

[0125] In step S1104, the answer to the question can be displayed. The answer can be generated based on knowledge files in a shared knowledge base. As shown in Figure 15a, the knowledge question-and-answer area 1303 can include an answer display area 1306, where the answer to the question can be displayed. The answer can include answer reference information 1501 and the answer content 1502. The answer reference information 1501 refers to the file information of the knowledge files referenced in the shared knowledge base during the answer generation process. Therefore, displaying answer reference information in the answer can enhance the richness and professionalism of the answer.

[0126] Optionally, embodiments of this application can support switching knowledge bases when dissatisfied with the answer information. The same question can then be answered based on the switched knowledge base. The switched knowledge base can include the first object's personal knowledge base, a shared knowledge base created by the first object, other shared knowledge bases joined by the first object besides the shared knowledge base, or a knowledge base across the entire internet; where a knowledge base across the entire internet refers to a knowledge base composed of all knowledge files on the internet. As shown in Figure 15b, the answer display area 1306 can also include a knowledge base switching control 1503 (e.g., the control named "Switch Knowledge Base" in Figure 15b). In response to a trigger operation on the knowledge base switching control 1503, a list of switchable knowledge bases 1504 can be displayed. A knowledge base can be selected from the list of switchable knowledge bases 1504, and the same question can be answered based on the selected knowledge base. In this way, if the answer information is unsatisfactory, switching back to the source knowledge base of the information will provide a satisfactory and accurate answer.

[0127] Optionally, embodiments of this application may support displaying the source information of the answer information in a shared knowledge base. Specifically, in response to a source viewing operation on an information fragment in the answer information, the source information of the information fragment in the shared knowledge base is displayed. The source information may include the source knowledge file of the information fragment in the shared knowledge base, and the specific paragraph of the information fragment in the source knowledge file. As shown in Figure 16, the source viewing operation on an information fragment in the answer information refers to a trigger operation on the information fragment in the answer information. A source information window 160 can be displayed, which can show the file title 1601 of the source knowledge file of the information fragment in the shared knowledge base (e.g., "A Brief Introduction to Quantum Mechanics" in Figure 16), and the specific paragraph 1602 of the information fragment in the source knowledge file (e.g., "Quantum Mechanics is..." in Figure 16). In this way, the source information of the information fragment in the answer information in the shared knowledge base can be quickly indexed, which is beneficial to obtaining more comprehensive and detailed information through the source information and improving the reliability of the answer information.

[0128] Optionally, the first object to be shared in the knowledge base can apply to join the knowledge base, that is, it can join the knowledge base in response to the joining operation. Before and after joining the knowledge base, the first object's question-and-answer capabilities based on the knowledge base are different. Specifically, before joining the knowledge base, the question-and-answer capabilities of the knowledge base are restricted; after joining, the restrictions on the question-and-answer capabilities of the knowledge base are lifted. The restrictions on the question-and-answer capabilities of the knowledge base may include at least one of the following: a limit on the number of questions asked to the knowledge base, a limit on the amount of information displayed in the answers, and a limit on the source of information for the output answers. Wherein:

[0129] (1) The limitation on the number of questions to the shared knowledge base means that before joining the shared knowledge base, the first user can only ask a limited number of questions to the shared knowledge base. For example, only 3 questions can be asked, which is equivalent to 3 trial questions to the shared knowledge base. After joining the shared knowledge base, there is no limit to the number of questions to the shared knowledge base. As shown in Figure 17a, after the trial questions to the shared knowledge base are used up before joining the shared knowledge base, the question box 1305 is locked. Being locked means that it is impossible to enter question information in the question box 1305. For example, the lock symbol and the lock text "To continue asking questions, please join the knowledge base" shown in Figure 17a indicate that the question box 1305 is locked. If you want to continue asking questions to the shared knowledge base, you can apply to join the shared knowledge base by triggering the join application control 1304. Figure 17a takes the joining application to the shared knowledge base as an example where confirmation is required.

[0130] (2) The limited amount of information displayed in the answer means that before joining the shared knowledge base, only part of the answer information is displayed, while after joining the shared knowledge base, the complete answer information can be displayed. As shown in Figure 17b, part of the answer information is displayed in the answer display area 1306. If you want to view the complete answer information, you can apply to join the shared knowledge base by triggering the join application control 1304. Figure 17b takes the application to join the shared knowledge base requiring confirmation as an example.

[0131] (3) The information source of the output answer information is restricted, meaning that before joining the shared knowledge base, it is not supported to view the information source of the answer information, but after joining the shared knowledge base, it is supported to view the information source of the answer information. As shown in Figure 17c, in response to the operation of viewing the source of information fragments in the answer information, the source information window 160 can be displayed. Before joining the shared knowledge base, the source information window 160 displays the information source locked information (for example, "To view the information source, please join the knowledge base" in Figure 17c). If you want to view the information source, you can apply to join the shared knowledge base by triggering the join application control 1304. Figure 17c takes the example of joining the shared knowledge base without confirmation and joining directly. After joining, the source information of the information fragment can be displayed in the source information window 160.

[0132] It can be seen that by restricting the question-and-answer capabilities of the shared knowledge base before it is added, the knowledge assets in the shared knowledge base can be protected, the illegal misuse of knowledge in the shared knowledge base can be prevented, and the security of the shared knowledge base can be improved.

[0133] In addition to asking questions about the shared knowledge base, the first user of the shared knowledge base can also view the knowledge files within it. Furthermore, the content viewed by the first user varies depending on their viewing permissions for the knowledge files. In response to a file viewing operation on a target knowledge file in the shared knowledge base, the target knowledge file can be displayed according to its viewing permissions. Specifically, if the file viewing permission for the target knowledge file is visible, the file content can be displayed; if the file viewing permission is not visible, a summary of the target knowledge file can be displayed.

[0134] As shown in Figure 18, by triggering file information 1801 of a target knowledge file in the shared knowledge base (e.g., the file title in Figure 18 is "A Brief Discussion of the History of Quantum Physics"), a file viewing operation can be performed on the target knowledge file. If the file viewing permission for the target knowledge file is set to visible (specifically, when the sharing object sets the permissions for the shared knowledge base, it sets the knowledge files in the shared knowledge base to be visible after sharing), the file content 1802 of the target knowledge file can be displayed; if the file viewing permission for the target knowledge file is set to invisible (specifically, when the sharing object sets the permissions for the shared knowledge base, it sets the knowledge files in the shared knowledge base to be invisible after sharing), the file summary 1803 of the target knowledge file can be displayed. The file summary of the target knowledge file can be generated by using artificial intelligence technology (e.g., large language model technology) to understand the content of the target knowledge file.

[0135] It is easy to see that by setting viewing permissions for knowledge files in a shared knowledge base, the knowledge assets in the shared knowledge base can be protected, thus protecting the ownership of the knowledge assets within the shared knowledge base by the creator or sharing object and improving the security of the shared knowledge base.

[0136] Optionally, if the first object sharing the knowledge base is set as the administrator of the knowledge base by the sharing object, then the first object can manage the knowledge base. Specifically, the first object, as the administrator of the knowledge base, can receive management operations on the knowledge base and synchronize the management results to the second object that has joined the knowledge base, i.e., to other members of the knowledge base. These management operations include at least one of the following: adding knowledge files to the knowledge base, deleting knowledge files from the knowledge base, and updating the file information of knowledge files in the knowledge base (e.g., updating file titles, updating file tags, etc.). In other words, the object sharing the knowledge base can become the administrator of the knowledge base and manage the knowledge files within it. The management rights of the knowledge base are not limited to the creator of the knowledge base, achieving collaborative management and thus improving the management efficiency. Furthermore, the first object, as the administrator, can also manage members of the knowledge base, similar to the sharing object, such as managing member additions, member roles, and member deletions, further improving the management efficiency of the knowledge base.

[0137] Steps S1101-S1103 above describe receiving content related to a shared knowledge base shared by a sharing object. Optionally, multiple shared knowledge bases shared by the sharing object can be received, and a shared knowledge base can be selected from multiple shared knowledge bases for question-and-answer processing. Specifically, question information can be received for a target shared knowledge base selected from multiple shared knowledge bases, and answer information for the question information can be displayed. The answer information is generated by processing the question information based on the knowledge files in the target shared knowledge base.

[0138] As shown in Figure 19, the terminal device used by the first object can access the shared knowledge base through a link or by scanning a graphic code to display the access interface 120. The access interface 120 can display the knowledge base information of multiple shared knowledge bases received (for example, the shared knowledge base named "Collection of Physics Knowledge" and the shared knowledge base named "Reference Weekly" in Figure 19). The access interface 120 can also include a viewing option 1201 for the shared knowledge base. The access interface 120 can receive a selection operation for the target shared knowledge base (for example, the target shared knowledge base selected in Figure 19 is the shared knowledge base named "Collection of Physics Knowledge"). In response to the triggering operation of the viewing option 1201 for the shared knowledge base, the target shared knowledge base can be displayed in the knowledge base display area 1302 of the first object, and a question information for the target shared knowledge base can be received in the question box 1305 of the knowledge question and answer area 1303 (for example, "What theories does this collection of physics knowledge mainly involve?" in Figure 19). In other words, when receiving multiple shared knowledge bases from a sharing object, users can select the shared knowledge base for question and answer as needed, thus improving the flexibility of knowledge question and answer.

[0139] In this embodiment of the application, an innovative method is provided in which the objects of the shared knowledge base can perform search and question answering based on the shared knowledge base after sharing, thereby improving the utilization rate of the shared knowledge base. There can be multiple objects of the shared knowledge base, enabling multiple people to perform AI search and question answering based on their own proprietary knowledge base content, so as to obtain more professional, focused and targeted question and answer results and meet personalized search and question answering needs.

[0140] The technical architecture of the knowledge base processing method is shown in Figure 20. The knowledge base functional area (knowledge base tab) of the intelligent knowledge platform involves knowledge bases and knowledge base lists. The intelligent knowledge platform also involves knowledge base member management (including member addition and removal management) and member permission (i.e., member role) management. The intelligent instruction platform also provides a sharing function for shared knowledge bases. During the sharing process, security is ensured through security proxy services and security logic services. The sharing process also transmits links and graphic codes through messages via a message module. The intelligent knowledge platform stores knowledge files in the knowledge base through a storage module and provides knowledge retrieval capabilities through the storage module. The storage module can include Redis and Elasticsearch. Redis is a database, and Elasticsearch (ES) is a distributed, highly scalable, and real-time search and data analysis engine that has the characteristics of a search engine and is often regarded as a database.

[0141] Based on the technical structure shown in Figure 20, the technologies involved in the knowledge base processing method may include:

[0142] I. Analysis of Knowledge Documents:

[0143] Knowledge document parsing refers to the standardization process of knowledge documents. Standardization can include extracting text and images from knowledge documents (e.g., web pages, PDF files, and Word documents), and performing data cleaning, organization, and removal of irrelevant information. Standardization improves the cleanliness, accuracy, and security of knowledge documents.

[0144] The knowledge file parsing process, as shown in Figure 21, can include four stages: In stage one, the knowledge file is retrieved using a file crawling tool, such as the Distiller tool. In stage two, the knowledge file's content (e.g., title, body text) is parsed according to path rules (e.g., XPath rules). In stage three, content filtering is performed on the knowledge file, removing files with limited content (e.g., no more than 50 lines) to ensure the richness of the knowledge file's content. In stage four, data cleaning is performed on the knowledge file, removing abnormal content and determining whether the knowledge file originates from a blacklisted site; if so, it is deleted to ensure the security of the knowledge file.

[0145] II. Updates to knowledge files in the knowledge base (including personal and shared knowledge bases):

[0146] Updating knowledge files in a knowledge base refers to updating the file information (e.g., file title, file tags, and file type) of the knowledge files in the storage module. The update process is shown in Figure 22 and may include:

[0147] 1. The storage module (e.g., Elasticsearch) writes a new version of the knowledge file to a shard, and sets its state to write-ahead (at this time, the new version of the data is completely invisible).

[0148] 2. After the storage module writes all the shards of the new version of the knowledge file, it changes the storage status (ES status) to committed (at this time, the shards can be retrieved, for example, in knowledge base question answering, but the knowledge file cannot be retrieved because the knowledge base version has not been updated), and changes the version of the knowledge file to the latest (at this time, the knowledge file can also be retrieved). The process from commit to modification is only the file version, which is very efficient.

[0149] a. Change the status to "Submission Timeout": Retry. If it keeps timing out: Abandon.

[0150] i. If you give up, the modification may not have been successful, and the new version of the knowledge file will not be visible.

[0151] ii. If you give up, the modification may be successful. However, since the knowledge base version has not been updated, the questions and answers based on the knowledge base may access both the new and old versions of the data at the same time. The knowledge file being modified can only access the old version.

[0152] b. Submission to modify knowledge base timed out: Same as above ii.

[0153] 3. Delete old version data in the storage module. Timeout: Same as above ii.

[0154] III. Index Reconstruction of Knowledge Bases (including Personal Knowledge Bases and Shared Knowledge Bases):

[0155] Knowledge base index reconstruction refers to updating the index of a knowledge file after the file information of the knowledge file has been updated. The prerequisite for updating is that access to the index is done via routing (e.g., alias), which is a prerequisite for index reconstruction. The update scheme employs a dual-write approach. The update process, as shown in Figure 23, may include:

[0156] 1. Record the start time ts;

[0157] 2. Start writing to the new index while simultaneously writing to the old index;

[0158] 3. Asynchronously rewrite files smaller than ts to a new index;

[0159] 4. After the asynchronous process ends, delete the old index reference and point to the new reference (atomic), and pause writing new data (at this time, the new data and the underlying alias are the same data, and if the upstream specifies a unique identifier (id) for writing, it will definitely not be duplicated).

[0160] 5. The client remains unaware of this, provided that the index is accessed via an alias.

[0161] IV. Question and answering based on knowledge bases (including personal knowledge bases and shared knowledge bases):

[0162] The knowledge-based question-answering process is the RAG (Retrieval Augmented Generation) process. RAG is a solution that combines a large language model and information retrieval technology, aiming to enhance the generative capabilities of a large language model by retrieving external knowledge. The knowledge-based question-answering process, as shown in Figure 24, can be summarized as follows: knowledge files in the knowledge base can be divided into file fragments for storage; similarity calculations can be performed between the question information and each file fragment obtained from the knowledge file division in the knowledge base to obtain the similarity between the question information and each file fragment; based on the similarity between the question information and each file fragment, file fragments matching the question information can be selected; the file fragments matching the question information can be rearranged; the question information and the rearranged file fragments can be input into the large language model, and the output of the large language model serves as the answer information for the question. It is easy to see that by leveraging the generative capabilities of the large language model and combining it with the file fragments matching the question information in the knowledge base to generate the answer information, the generated answer information can accurately respond to the question.

[0163] Furthermore, to improve the recall rate of file segments matching the query information, vector and text indexes can be established for each file segment. Similarity calculation can include both vector and text similarity calculations. For vector similarity calculation: the query information can be vectorized to obtain a query vector. The vector similarity between the query vector and the vector indices of the various file segments obtained from the knowledge files in the knowledge base can be calculated. Therefore, based on the vector similarity between the query vector and the vector indices of each file segment, vector indices matching the query vector can be selected. The file segments corresponding to the vector indices matching the query vector are the file segments matching the query information. For text similarity calculation: the query information can be segmented to obtain segmented information. The text similarity between the segmented information and the text indices of the various file segments obtained from the knowledge files in the knowledge base can be calculated. Therefore, based on the text similarity between the segmented information and the text indices of each file segment, text indices matching the segmented information can be selected. The file segments corresponding to the text indices matching the segmented information are the file segments matching the query information.

[0164] The knowledge base's RAG solution is primarily based on Elasticsearch and consists of two parts: index building and retrieval. To handle large-scale data storage, we designed a scalable index management scheme for easy future expansion. Simultaneously, to improve recall, we established two indexing methods: vector indexing and text indexing, corresponding to two retrieval methods: vector retrieval and text retrieval. For the file fragments retrieved and recalled, to further improve the relevance between file fragments and question information, we used a re-ranking model (e.g., a rerank model) for re-ranking and filtering. Finally, in the original text index, we used the re-ranking model to calculate the similarity between information fragments in the answer information and knowledge files in the knowledge base, and then indexed the original text content of the information fragments, thus determining the source information of the information fragments. The technical details of the knowledge base-based question-answering process are described below:

[0165] 1. Establishing vector indices:

[0166] To address the diversity of knowledge base data, we adopted a multilingual vectorization model (embedding model) fine-tuned based on business data corpora. For example, the vectorization model could be bge-m3 (a multi-functional, multilingual, and multi-dimensional language vector model), which performs embedding vectorization on the file fragments of knowledge documents, transforming them into 1024-dimensional vectors. A vector index was built in the Elasticsearch database based on the HNSW (Hierarchical Navigable Small Word) algorithm. To maintain speed for large-scale vector searches, int8 quantization (8-bit integer quantization) was used. The process of building the vector index using the HNSW algorithm is shown in Figure 25, which is an index building process that reduces the feature radius.

[0167] 2. Creating a text index:

[0168] To further improve recall, we also built a text-based inverted index. By segmenting file shards into words, we created a multi-field, multi-dimensional inverted index in Elasticsearch, specifically indexing file pairs, titles, content, tags, and the pinyin of file titles.

[0169] 3. Index organization:

[0170] To handle large-scale terabyte (TB) level data storage and retrieval, we designed a multi-cluster, multi-index-based organizational structure that dynamically allocates clusters based on business data growth. The allocation process, as shown in Figure 26, routes users to different index clusters based on their UID (User Identification), and then routes file shards using the user UID. This ensures that write and retrieval performance remains within a controllable range to cope with rapid business growth in the later stages.

[0171] 4. The process of retrieving file fragments:

[0172] The process of file fragment retrieval is shown in Figure 27. The query information can be processed (e.g., vectorization and word segmentation). A multi-way recall strategy can be used to recall file fragments that match the query information. The recalled file fragments can be rearranged according to a rearrangement strategy to obtain the final file fragments. Specifically:

[0173] The vector retrieval process can include: Vector retrieval mainly empowers the knowledge base with semantic search capabilities. By vectorizing the query information, the vectorized query vector is used to retrieve file fragments related to the semantics of the query information on the ElasticSearch vector index based on the HNSW algorithm.

[0174] The text retrieval process can include: Text retrieval is mainly used to supplement the fragmented recall pool and further improve the recall rate. We perform word segmentation on the query information, and calculate the similarity between the segmented information and the text index of the file fragments. Here, the similarity calculation is performed by matching the segmented information with the text index of the file fragments. Optionally, different fields can adopt different processing and search strategies. For example, for tag search, we will choose complete matching, and for title-based search, we will also use pinyin search. By combining multiple strategies, we can further improve the recall rate by performing text retrieval of file fragments based on text relevance scoring algorithms (e.g., BM25).

[0175] The reordering process may include: for recalling multi-path file fragments, the final relevance calculation and scoring of the file fragments are required to obtain the final reordered sequence. Our reordering model is a multilingual version fine-tuned based on the business corpus. For example, the reordering model could be bge-reranker-v2-m3, which is a retrieval ranking model mainly used to optimize the result ranking in the information retrieval process and improve the relevance and diversity of retrieval results.

[0176] Specifically, the reordering strategy for text search scenarios can include: In text search scenarios, for file fragments retrieved based on semantic vectors, the relevance requirements for file fragments are higher, so the reordering strategy will be more stringent. However, for file fragments retrieved for text search, as long as there is a match, they need to be displayed. Therefore, in text search scenarios, the reordering strategy is roughly designed to reorder, score, and filter the content of vector search, and directly display the fragments retrieved for text search.

[0177] Reordering strategies for question-answering scenarios can include: In question-answering scenarios, the goal is to allocate the most relevant content to the most relevant segments within the context of a large language model. Therefore, in this scenario, our strategy is to use multi-way recall, unified reordering, and finally select the Top N (i.e., the scores are ranked in the top N positions, where N is a positive integer) file segments based on the scenario.

[0178] 5. Original text index calculation (i.e., determining the source of the answer information):

[0179] For the large language model's response to the question, we need to parse the response flow of the large language model in the original knowledge file in the knowledge base to obtain response points. We then use these points and the paragraphs of the file segments to calculate the relevance. The calculation model is a multilingual reordering model that has been fine-tuned based on the business corpus (for example, the reordering model could be bge-reranker-v2-m3). Finally, we locate the relevant information in the original knowledge file based on the coordinate information corresponding to the file segments.

[0180] V. Sharing of the Shared Knowledge Base:

[0181] Each knowledge base has a unique knowledge base identifier, and each knowledge file has a unique file identifier. When sharing a shared knowledge base, sharing information can be generated based on the knowledge base identifier and the file identifiers of each knowledge file within the shared knowledge base. This sharing information can be encapsulated as a sharing credential (i.e., a sharing link or a sharing graphic code), which can then be shared by the sharing end.

[0182] Upon receiving the sharing credential, the receiving end can parse it to obtain the sharing information. Based on the knowledge base identifier of the shared knowledge base in the sharing information, the receiving end can retrieve the knowledge base information of the shared knowledge base from the server of the intelligent knowledge platform. Furthermore, based on the file identifiers of each knowledge file in the sharing information, the receiving end can retrieve the file information of each knowledge file from the server of the intelligent knowledge platform. Thus, the receiving end can display the shared knowledge base based on the knowledge base information of the shared knowledge base and the file information of each knowledge file in the shared knowledge base.

[0183] Please refer to Figure 28, which is a schematic diagram of a knowledge base processing device provided in an embodiment of this application. This knowledge base processing device can be installed in a computer device provided in this embodiment. The computer device can be a sharing endpoint, i.e., a terminal device used by an object sharing the knowledge base. The knowledge base processing device shown in Figure 28 can be a computer program running on the computer device. This knowledge base processing device can be used to execute some or all of the steps in the method embodiment shown in Figure 11. Please refer to Figure 28; the knowledge base processing device may include the following units:

[0184] Processing unit 2802 is used to receive sharing credentials for the shared knowledge base;

[0185] Display unit 2801 is used to display the shared knowledge base being shared in response to an access operation on the sharing credential, wherein the shared knowledge base includes at least one knowledge file;

[0186] Processing unit 2802 is used to receive query information for the shared knowledge base;

[0187] The display unit 2801 is also used to display the answer information to the question information, which is generated by processing the question information based on the knowledge files in the shared knowledge base.

[0188] In one implementation, when the processing unit 2802 receives query information about the shared knowledge base, it is specifically configured to perform any of the following:

[0189] Receive query information for all knowledge files in the shared knowledge base;

[0190] Receive query information for a specified portion of knowledge files in a shared knowledge base; wherein the specification method includes at least one of the following: specification based on file tags, specification based on keywords, and specification based on file type.

[0191] In one implementation, the display unit 2801 is further configured to perform the following steps:

[0192] In response to a file viewing operation on a target knowledge file in a shared knowledge base, the target knowledge file is displayed according to the file viewing permissions.

[0193] In one implementation, when displaying the target knowledge file according to the file viewing permissions of the target knowledge file, the display unit 2801 specifically performs the following steps:

[0194] If the file viewing permission for the target knowledge file is set to file visibility, then the file content of the target knowledge file will be displayed;

[0195] If the file viewing permission for the target knowledge file is set to invisible, then the file summary of the target knowledge file will be displayed.

[0196] In one implementation, the processing unit 2802 is further configured to perform the following steps:

[0197] In response to the "join" operation of the shared knowledge base, apply to join the shared knowledge base;

[0198] Before joining the shared knowledge base, the question-and-answer capabilities of the shared knowledge base are restricted. After joining the shared knowledge base, the restrictions on the question-and-answer capabilities of the shared knowledge base are lifted. The restrictions on the question-and-answer capabilities of the shared knowledge base include at least one of the following: the number of questions asked to the shared knowledge base is limited, the amount of information displayed in the answer is limited, and the source of information for the output answer is limited.

[0199] In one implementation, the device is located in the terminal device of a first object sharing the shared knowledge base, the first object being the administrator of the shared knowledge base; the processing unit 2802 is further configured to perform the following steps:

[0200] Receive management operations on the shared knowledge base;

[0201] The management results of the shared knowledge base will be synchronized to the second object that has been added to the shared knowledge base;

[0202] The management operations for the shared knowledge base include at least one of the following: adding knowledge files to the shared knowledge base, deleting knowledge files in the shared knowledge base, and updating the file information of knowledge files in the shared knowledge base.

[0203] In one implementation, the display unit 2801 is further configured to perform the following steps:

[0204] In response to an operation to view the source of an information fragment in the answer information, display the source information of the information fragment in the shared knowledge base.

[0205] In one implementation, there are multiple shared knowledge bases; the processing unit 2802, when receiving query information about the shared knowledge base, specifically performs the following steps:

[0206] Receive query information for a target shared knowledge base selected from multiple shared knowledge bases;

[0207] The answer information is generated by processing the question information based on the knowledge files in the target shared knowledge base.

[0208] According to one embodiment of this application, the units in the knowledge base processing apparatus shown in FIG28 can be individually or entirely merged into one or more other units, or some of the units can be further divided into multiple functionally smaller units. This achieves the same operation without affecting the technical effect of the embodiment of this application. The above units are based on logical function division. In practical applications, the function of one unit can also be implemented by multiple units, or the function of multiple units can be implemented by one unit. In other embodiments of this application, the knowledge base processing apparatus may also include other units. In practical applications, these functions can also be implemented with the assistance of other units, and can be implemented collaboratively by multiple units.

[0209] According to another embodiment of this application, a processing apparatus for the knowledge base shown in FIG28, and a knowledge base processing method of the embodiments of this application, can be constructed by running a computer program capable of performing some or all of the steps involved in the method shown in FIG11 on a general-purpose computing device, such as a computer, which includes processing elements and storage elements such as a central processing unit (CPU), random access storage medium (RAM), and read-only storage medium (ROM). The computer program can be recorded on, for example, a computer-readable storage medium, loaded into the aforementioned computing device through the computer-readable storage medium, and run therein.

[0210] In this embodiment, a sharing credential for a shared knowledge base can be received. In response to an access operation to the sharing credential, the shared knowledge base can be displayed. Therefore, the shared knowledge base possesses sharing attributes and can be shared, thus improving its shareability. Furthermore, the recipient can conduct question-and-answer sessions based on the knowledge files in the shared knowledge base. Thus, this embodiment provides two methods for using a shared knowledge base: sharing and question-and-answer sessions, thereby increasing its utilization rate.

[0211] Please refer to Figure 28, which is a schematic diagram of a knowledge base processing device provided in an embodiment of this application. This knowledge base processing device can be located in a computer device provided in this embodiment. The computer device can be a sharing terminal, i.e., a terminal device used by objects sharing the knowledge base. The knowledge base processing device shown in Figure 28 can be a computer program running on the computer device. This knowledge base processing device can be used to execute some or all of the steps in the method embodiment shown in Figure 2. Please refer to Figure 28; the knowledge base processing device may include the following units:

[0212] Display unit 2801 is used to display the shared knowledge base of the sharing object in response to the login operation of the sharing object. The shared knowledge base includes at least one knowledge file.

[0213] The processing unit 2802 is used to share the shared knowledge base in response to the sharing operation, so that the receiving end can generate an answer to the question based on the knowledge files in the shared knowledge base.

[0214] In one implementation, the device is located in the terminal device of the sharing object of the shared knowledge base; the shared knowledge base includes shared knowledge bases created by the sharing object or shared knowledge bases joined by the sharing object.

[0215] The process of creating a shared knowledge base includes:

[0216] Create a blank shared knowledge base;

[0217] Add knowledge files to a blank shared knowledge base;

[0218] The sources of the added knowledge files include at least one of the following: local files on the terminal device of the sharing object, and knowledge files in the personal knowledge base of the sharing object; the shared knowledge base is a knowledge base with sharing attributes, while the personal knowledge base is a knowledge base without sharing attributes.

[0219] In one implementation, the processing unit 2802 is further configured to perform the following steps:

[0220] Receive settings for sharing permissions to the shared knowledge base;

[0221] The sharing permissions for the shared knowledge base include at least one of the following: visibility permissions for knowledge files in the shared knowledge base, permission to join the shared knowledge base, and permission to hide the shared knowledge base.

[0222] In one implementation, the processing unit 2802 is further configured to perform the following steps:

[0223] Receive member management operations for the shared knowledge base;

[0224] The member management operations include at least one of the following: member addition management operations, member role management operations, and member deletion operations.

[0225] In one implementation, there are multiple shared knowledge bases, each of which includes at least one knowledge file, and each shared knowledge base is either a created shared knowledge base or a joined shared knowledge base; the processing unit 2802 is further configured to perform the following steps:

[0226] Receive knowledge base selection operations performed on multiple shared knowledge bases;

[0227] Processing unit 2802 is configured to respond to a sharing operation on a shared knowledge base, specifically performing the following steps when sharing the shared knowledge base:

[0228] In response to the sharing operation of the shared knowledge base selected by the knowledge base selection operation, the shared knowledge base selected by the knowledge base selection operation is shared.

[0229] In one implementation, the processing unit 2802 is further configured to perform the following steps:

[0230] Receive a file selection operation performed on at least one knowledge file;

[0231] Processing unit 2802 is configured to respond to a sharing operation on a shared knowledge base, specifically performing the following steps when sharing the shared knowledge base:

[0232] In response to a sharing operation on a shared knowledge base, the knowledge file selected by the file selection operation in the shared knowledge base is shared.

[0233] According to one embodiment of this application, the units in the knowledge base processing apparatus shown in FIG28 can be individually or entirely merged into one or more other units, or some of the units can be further divided into multiple functionally smaller units. This achieves the same operation without affecting the technical effect of the embodiment of this application. The above units are based on logical function division. In practical applications, the function of one unit can also be implemented by multiple units, or the function of multiple units can be implemented by one unit. In other embodiments of this application, the knowledge base processing apparatus may also include other units. In practical applications, these functions can also be implemented with the assistance of other units, and can be implemented collaboratively by multiple units.

[0234] According to another embodiment of this application, a processing apparatus for the knowledge base shown in FIG28, and a knowledge base processing method of the embodiments of this application, can be constructed by running a computer program capable of performing some or all of the steps involved in the method shown in FIG2 on a general-purpose computing device including processing elements and storage elements such as a central processing unit (CPU), random access storage medium (RAM), and read-only storage medium (ROM). The computer program can be recorded on, for example, a computer-readable storage medium, loaded into the aforementioned computing device through the computer-readable storage medium, and run therein.

[0235] In this embodiment, a sharing credential for a shared knowledge base can be received. In response to an access operation to the sharing credential, the shared knowledge base can be displayed. Therefore, the shared knowledge base possesses sharing attributes and can be shared, thus improving its shareability. Furthermore, the recipient can conduct question-and-answer sessions based on the knowledge files in the shared knowledge base. Thus, this embodiment provides two methods for using a shared knowledge base: sharing and question-and-answer sessions, thereby increasing its utilization rate.

[0236] Please refer to Figure 29, which is a schematic diagram of a computer device provided in an embodiment of this application. The computer device shown in Figure 29 includes at least a processor 2901, an input interface 2902, an output interface 2903, and a computer-readable storage medium 2904. The processor 2901, input interface 2902, output interface 2903, and computer-readable storage medium 2904 can be connected via a bus or other means.

[0237] Computer-readable storage medium 2904 can be stored in the memory of a computer device. Computer-readable storage medium 2904 is used to store computer programs, including computer instructions. Processor 2901 is used to execute the computer program stored in computer-readable storage medium 2904. Processor 2901 (or CPU (Central Processing Unit)) is the computing and control core of the computer device, suitable for implementing computer programs, specifically for loading and executing computer programs to achieve corresponding method flows or functions.

[0238] This application also provides a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both built-in storage media in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space for storing the operating system of the computer device. Furthermore, the storage space also stores computer programs suitable for loading and execution by a processor. It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device; optionally, it can also be at least one computer-readable storage medium located remotely from the aforementioned processor.

[0239] The computer device can be the sharing end, that is, the terminal device used by the object (first object) of the shared knowledge base. Specifically, the processor 2901 can load and execute the computer program stored in the computer-readable storage medium 2904 to implement the corresponding steps in the method shown in Figure 11. The computer device can be the sharing end, that is, the terminal device used by the object of the shared knowledge base. Specifically, the processor 2901 can load and execute the computer program stored in the computer-readable storage medium 2904 to implement the corresponding steps in the method shown in Figure 2.

[0240] In this embodiment, a sharing credential for a shared knowledge base can be received. In response to an access operation to the sharing credential, the shared knowledge base can be displayed. Therefore, the shared knowledge base possesses sharing attributes and can be shared, thus improving its shareability. Furthermore, the recipient can conduct question-and-answer sessions based on the knowledge files in the shared knowledge base. Thus, this embodiment provides two methods for using a shared knowledge base: sharing and question-and-answer sessions, thereby increasing its utilization rate.

[0241] This application also 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 executes the computer program, causing the computer device to perform the aforementioned knowledge base processing method.

Claims

1. A method for processing a knowledge base, characterized in that, The method is executed by a terminal device of the first object of the shared knowledge base, and the method includes: Receive sharing credentials for the shared knowledge base; In response to an access operation to the sharing credential, the shared knowledge base being shared is displayed, the shared knowledge base including at least one knowledge file; Receive query information for the shared knowledge base; The system displays the answer to the question, which is generated based on knowledge files in the shared knowledge base.

2. The method as described in claim 1, characterized in that, The receiving of query information for the shared knowledge base includes any of the following: Receive query information for all knowledge files in the shared knowledge base; Receive query information for a specified portion of knowledge files in the shared knowledge base; wherein the specification method includes at least one of the following: specification based on file tags, specification based on keywords, and specification based on file type.

3. The method as described in claim 1 or 2, characterized in that, The method further includes: In response to a file viewing operation on a target knowledge file in the shared knowledge base, the target knowledge file is displayed according to the file viewing permissions for the target knowledge file.

4. The method according to any one of claims 1-3, characterized in that, The step of displaying the target knowledge file according to the file viewing permissions of the target knowledge file includes: If the file viewing permission for the target knowledge file is set to file visibility permission, then the file content of the target knowledge file will be displayed; If the file viewing permission for the target knowledge file is set to invisible, then the file summary of the target knowledge file will be displayed.

5. The method according to any one of claims 1-4, characterized in that, The method further includes: In response to the join operation of the shared knowledge base, a request is made to join the shared knowledge base; Before being added to the shared knowledge base, the question-and-answer capabilities of the shared knowledge base were restricted; after being added to the shared knowledge base, the restrictions on the question-and-answer capabilities of the shared knowledge base were lifted. The restrictions on the question-and-answer capabilities of the shared knowledge base included at least one of the following: the number of questions that can be asked to the shared knowledge base was limited, the amount of information displayed in the answer was limited, and the source of information for the output answer was limited.

6. The method according to any one of claims 1-5, characterized in that, The first object is the administrator of the shared knowledge base; the method further includes: Receive management operations on the shared knowledge base; The management results of the shared knowledge base will be synchronized to the second object that has been added to the shared knowledge base; The management operations of the shared knowledge base include at least one of the following: adding knowledge files to the shared knowledge base, deleting knowledge files in the shared knowledge base, and updating the file information of knowledge files in the shared knowledge base.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: In response to a source viewing operation on an information fragment in the answer information, the source information of the information fragment in the shared knowledge base is displayed.

8. The method according to any one of claims 1-7, characterized in that, The number of shared knowledge bases is multiple; receiving query information for the shared knowledge base includes: Receive query information for a target shared knowledge base selected from multiple shared knowledge bases; The answer information is obtained by generating an answer to the question information based on knowledge files in the target shared knowledge base.

9. A method for processing a knowledge base, characterized in that, The method is executed by the terminal device of the sharing object, and the method includes: In response to the login operation of the sharing object, the shared knowledge base of the sharing object is displayed, the shared knowledge base including at least one knowledge file; In response to the sharing operation of the shared knowledge base, the shared knowledge base is shared, enabling the recipient to generate an answer to the question based on the knowledge files in the shared knowledge base.

10. The method as described in claim 9, characterized in that, The shared knowledge base includes shared knowledge bases created by the sharing object or shared knowledge bases joined by the sharing object; The creation process of the shared knowledge base includes: Create a blank shared knowledge base; Add knowledge files to the blank shared knowledge base; The sources of the added knowledge files include at least one of the following: local files on the terminal device of the sharing object, and knowledge files in the personal knowledge base of the sharing object; the shared knowledge base is a knowledge base with sharing attributes, and the personal knowledge base is a knowledge base without sharing attributes.

11. The method as described in claim 9 or 10, characterized in that, The method further includes: Receive settings for sharing permissions of the shared knowledge base; The sharing permissions for the shared knowledge base include at least one of the following: visibility permissions for knowledge files in the shared knowledge base, permission to join the shared knowledge base, and permission to hide the shared knowledge base.

12. The method according to any one of claims 9-11, characterized in that, The method further includes: Receive member management operations for the shared knowledge base; The member management operations include at least one of the following: member addition management operation, member role management operation, and member deletion operation.

13. The method according to any one of claims 9-12, characterized in that, The number of shared knowledge bases is multiple, each of which includes at least one knowledge file, and each of which is a created shared knowledge base or a joined shared knowledge base; The method further includes: Receive knowledge base selection operations performed on multiple shared knowledge bases; The step of sharing the shared knowledge base in response to a sharing operation includes: In response to a sharing operation of the shared knowledge base selected by the knowledge base selection operation, the shared knowledge base selected by the knowledge base selection operation is shared.

14. The method according to any one of claims 9-13, characterized in that, The method further includes: Receive a file selection operation performed on the at least one knowledge file; The step of sharing the shared knowledge base in response to a sharing operation includes: In response to the sharing operation of the shared knowledge base, the knowledge files selected by the file selection operation in the shared knowledge base are shared.

15. A knowledge base processing apparatus, characterized in that, include: The processing unit is used to receive the sharing credentials for the shared knowledge base; A display unit is configured to display the shared knowledge base, which includes at least one knowledge file, in response to an access operation to the sharing credential. A processing unit is configured to receive query information for the shared knowledge base; The display unit is also used to display answer information to the question information, which is obtained by generating an answer to the question information based on knowledge files in the shared knowledge base.

16. A knowledge base processing apparatus, characterized in that, include: The display unit is used to display a shared knowledge base in response to a login operation of a sharing object. The shared knowledge base includes at least one knowledge file. The processing unit is configured to share the shared knowledge base in response to a sharing operation, enabling the recipient to generate an answer to the question based on the knowledge files in the shared knowledge base.

17. A computer device, characterized in that, The computer device includes: A processor is a tool for implementing computer programs. A computer-readable storage medium storing a computer program adapted to be loaded by the processor and executed as a knowledge base processing method as claimed in any one of claims 1-8, or as a knowledge base processing method as claimed in any one of claims 9-14.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded by a processor and executed by the processing method of the knowledge base as described in any one of claims 1-8, or the processing method of the knowledge base as described in any one of claims 9-14.

19. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the knowledge base processing method as described in any one of claims 1-8, or the knowledge base processing method as described in any one of claims 9-14.

Images