Information processing method and device, electronic equipment, computer readable storage medium and computer program product

Abstract

The invention provides an information processing method and device, electronic equipment, a computer readable storage medium and a computer program product. The method comprises the steps that an information processing interface is displayed, and the information processing interface comprises an interaction control based on artificial intelligence; in response to an input operation in the interaction control, the input first information is displayed, and at least one first knowledge base is displayed, and the at least one first knowledge base comprises knowledge related to the first information. According to the method and the device, the flow and propagation of knowledge can be promoted, so that the knowledge acquisition efficiency is effectively improved.

Description

Technical Field

[0001] This application relates to the field of Internet technologies, and in particular, to an information processing method, apparatus, electronic device, computer-readable storage medium, and computer program product. Background Art

[0002] In recent years, the development of artificial intelligence (AI) has made great progress, especially language models based on large models. These large models have powerful natural language processing capabilities and knowledge expression capabilities, perform well in context learning, long-distance understanding, etc., and can understand and generate human language. However, in related technologies, there is no linkage between the AI question-and-answer process and the knowledge base, resulting in the inability of users to discover good content during the AI question-and-answer process, thereby restricting the flow and dissemination of knowledge and reducing the efficiency of users to obtain knowledge. Summary of the Invention

[0003] Embodiments of this application provide an information processing method, apparatus, electronic device, computer-readable storage medium, and computer program product, which can promote the flow and dissemination of knowledge, thereby effectively improving the efficiency of obtaining knowledge.

[0004] The technical solution of the embodiments of this application is implemented as follows:

[0005] Embodiments of this application provide an information processing method, including:

[0006] Display an information processing interface, where the information processing interface includes an artificial intelligence-based interaction control;

[0007] In response to an input operation in the interaction control, display the input first information, and

[0008] Display at least one first knowledge base, where the at least one first knowledge base includes knowledge related to the first information.

[0009] Embodiments of this application provide an information processing apparatus, including:

[0010] A display module, configured to display an information processing interface, where the information processing interface includes an artificial intelligence-based interaction control;

[0011] The display module is further configured to, in response to an input operation in the interaction control, display the input first information and display at least one first knowledge base, where the at least one first knowledge base includes knowledge related to the first information.

[0012] Embodiments of this application provide an electronic device, including:

[0013] A memory for storing computer-executable instructions;

[0014] A processor, when executing the computer-executable instructions stored in the memory, implements the information processing method provided by the embodiments of the present application.

[0015] The embodiments of the present application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the information processing method provided by the embodiments of the present application.

[0016] The embodiments of the present application provide a computer program product including a computer program or computer-executable instructions, which, when executed by a processor, implement the information processing method provided by the embodiments of the present application.

[0017] The embodiments of the present application have the following beneficial effects:

[0018] By introducing at least one first knowledge base including knowledge related to the input first information in the Q&A process, the linkage between the Q&A process and the knowledge base is realized. Users can directly view the knowledge in the knowledge base they are interested in during the Q&A process, promoting the flow and dissemination of knowledge, and at the same time effectively improving the efficiency of users to obtain knowledge. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a schematic diagram of the architecture of the information processing system 100 provided by the embodiments of the present application;

[0020] Figure 2 is a schematic diagram of the structure of the electronic device 500 provided by the embodiments of the present application;

[0021] Figure 3 is a first flowchart of the information processing method provided by the embodiments of the present application;

[0022] Figure 4A is a first application scenario diagram of the information processing method provided by the embodiments of the present application;

[0023] Figure 4B is a second application scenario diagram of the information processing method provided by the embodiments of the present application;

[0024] Figure 4C is a third application scenario diagram of the information processing method provided by the embodiments of the present application;

[0025] Figure 4D is a fourth application scenario diagram of the information processing method provided by the embodiments of the present application;

[0026] Figure 4E is a fifth application scenario diagram of the information processing method provided by the embodiments of the present application;

[0027] Figure 5 It is the second process schematic diagram of the information processing method provided by the embodiments of the present application;

[0028] Figure 6 It is the third process schematic diagram of the information processing method provided by the embodiments of the present application;

[0029] Figure 7 It is the sixth application scenario schematic diagram of the information processing method provided by the embodiments of the present application;

[0030] Figure 8 It is the seventh application scenario schematic diagram of the information processing method provided by the embodiments of the present application;

[0031] Figure 9 It is the schematic diagram of available functions and permissions in the default state provided by the embodiments of the present application;

[0032] Figure 10 It is the schematic diagram of available functions and permissions in the state of being published to the square provided by the embodiments of the present application;

[0033] Figure 11 It is the schematic diagram of available functions and permissions in the state of being converted to the private state provided by the embodiments of the present application;

[0034] Figure 12 It is the schematic diagram of the organizational structure of the index provided by the embodiments of the present application. Detailed implementation manners

[0035] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0036] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0037] It can be understood that in the embodiments of the present application, data related to user information and the like are involved. When the embodiments of the present application are applied to specific products or technologies, user permission or consent needs to be obtained, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards.

[0038] In the following description, the terms "first / second / ..." only distinguish similar objects and do not represent a specific order for the objects. It can be understood that "first / second / ..." can be interchanged with a specific order or sequence when permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0039] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0040] Before further elaborating on the embodiments of this application, the nouns and terms involved in the embodiments of this application are described. The nouns and terms involved in the embodiments of this application are applicable to the following explanations.

[0041] 1) In response to: used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more executed operations can be real-time or can have a set delay; without special instructions, there is no limitation on the execution order of the multiple executed operations.

[0042] 2) Knowledge Base: refers to the organic combination and arrangement of various computer-stored information through specific methods to form a structured, easy-to-operate, easy-to-utilize, and comprehensive organized knowledge cluster to facilitate people's knowledge retrieval and utilization. That is to say, a knowledge base is an information collection, including various forms of data and information, such as including text, images, audio, and video, etc. Divided from the content, the knowledge base can include: enterprise-level knowledge base, industry knowledge base, public knowledge base, professional knowledge base, etc.; divided from the ownership and access rights, the knowledge base can include personal knowledge base (i.e., private domain knowledge base) and shared knowledge base (shared knowledge base).

[0043] 3) Embedding: is a "vectorization" or "vector representation" technology, mainly referring to mapping discrete high-dimensional data (such as text, pictures, audio, etc.) to a low-dimensional continuous vector space. This process will generate vectors composed of real numbers to capture the potential relationships and structures of the original data.

[0044] 4) Index: is a structure used to quickly search for and access data in a database table, including keyword index and vector index. Among them, the keyword index is obtained by performing word segmentation on the knowledge in multiple pre-created knowledge bases, and the vector index is obtained by performing embedding processing on the knowledge in multiple pre-created knowledge bases.

[0045] Embodiments of the present application provide an information processing method, apparatus, electronic device, computer-readable storage medium, and computer program product, which can promote the flow and dissemination of knowledge, thereby effectively improving the efficiency of users to obtain knowledge. The electronic device provided by the embodiments of the present application will be described below. The electronic device provided by the embodiments of the present application can be implemented as a terminal device, or can also be jointly implemented by a server and a terminal device. The following takes the case where the information processing method provided by the embodiments of the present application is jointly implemented by a server and a terminal device as an example for description.

[0046] Exemplarily, referring to Figure 1 , Figure 1 is a schematic diagram of the architecture of the information processing system 100 provided by the embodiments of the present application. To implement an application that supports and promotes the flow and dissemination of knowledge, such as Figure 1 As shown, the information processing system 100 provided by the embodiments of the present application includes: a server 200, a network 300, and a terminal device 400. Among them, the network 300 can be a local area network or a wide area network, or a combination of the two. The terminal device 400 is a terminal device associated with the user. An application program 410 is running on the terminal device 400. The application program 410 can be an application program based on artificial intelligence. For example, it can be an intelligent workbench application program. Among them, the intelligent workbench application program is an efficiency tool that integrates search, reading, and collaboration, and its functions cover knowledge acquisition, information processing, and content output. For example, specifically, it can include: knowledge base management, content generation, AI Q&A, multi-modal text creation, image generation, writing assistance, multi-language translation, automatic generation of mind maps, and intelligent arrangement of outlines, etc. The intelligent workbench can access various artificial intelligence models and support users to select. For example, taking the application program 410 as an intelligent workbench application program as an example, an information processing interface (such as the home page of the intelligent workbench) can be displayed on the intelligent workbench application program. Among them, the information processing interface can include interaction controls based on artificial intelligence; the intelligent workbench application program responds to the input operation of the user in the interaction control, displays the first information input by the user, and sends the first information input by the user to the server 200 through the network 300, so that the server 200 queries at least one first knowledge base including knowledge related to the first information. Subsequently, the intelligent workbench application program can display at least one first knowledge base returned by the server 200. In this way, the linkage between the Q&A process and the knowledge base is realized, promoting the flow and dissemination of knowledge, thereby effectively improving the efficiency of users to obtain knowledge.

[0047] It should be noted that Figure 1The server 200 therein may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery network (CDN), and big data and artificial intelligence platforms. The terminal device 400 may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, a vehicle-mounted terminal, etc., but is not limited thereto. The terminal device 400 and the server 200 may be directly or indirectly connected through wired or wireless communication means, which are not limited in the embodiments of the present application.

[0048] In some embodiments, the terminal device may also implement the information processing method provided in the embodiments of the present application by running various computer-executable instructions or computer programs. For example, the computer-executable instructions may be commands at the microprogram level, machine instructions, or software instructions. The computer program may be a native program or software module in the operating system; it may be a local (Native) application (APPlication, APP), that is, a program that needs to be installed in the operating system to run, such as the smart workbench APP; or it may be a small program that can be embedded in any APP, that is, a program that only needs to be downloaded to the browser environment to run. In short, the above computer-executable instructions may be instructions in any form, and the above computer programs may be application programs, modules, or plugins in any form.

[0049] Next, the structure of the electronic device provided in the embodiments of the present application will be further described. Taking the electronic device as the terminal device as an example, see Figure 2 , Figure 2 which is a schematic structural diagram of the electronic device 500 provided in the embodiments of the present application. Figure 2 The electronic device 500 shown includes: at least one processor 510, a memory 550, at least one network interface 520, and a user interface 530. Each component in the electronic device 500 is coupled together through a bus system 540. It can be understood that the bus system 540 is used to realize the connection and communication between these components. In addition to including a data bus, the bus system 540 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clear illustration, in Figure 2 all kinds of buses are labeled as the bus system 540.

[0050] The processor 510 can be an integrated circuit chip with the ability to process signals, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0051] The user interface 530 includes one or more output devices 531 that enable the presentation of media content, including one or more speakers and / or one or more visual display screens. The user interface 530 also includes one or more input devices 532, including user interface components that facilitate user input, such as a keyboard, a mouse, a microphone, a touch screen display, a camera, other input buttons, and controls.

[0052] The memory 550 can be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state memory, hard disk drives, optical disc drives, etc. The memory 550 optionally includes one or more storage devices that are physically located far from the processor 510.

[0053] The memory 550 includes volatile memory or non-volatile memory, and can also include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), and the volatile memory can be random access memory (RAM). The memory 550 described in the embodiments of the present application is intended to include any suitable type of memory.

[0054] In some embodiments, the memory 550 is capable of storing data to support various operations. Examples of such data include programs, modules, and data structures, or subsets or supersets thereof, which are illustratively described below.

[0055] The operating system 551 includes system programs for processing various basic system services and performing hardware-related tasks, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks;

[0056] The network communication module 552 is used to reach other computing devices via one or more (wired or wireless) network interfaces 520. Exemplary network interfaces 520 include: Bluetooth, Wi-Fi (Wireless Fidelity), and USB (Universal Serial Bus), etc.;

[0057] A presentation module 553 for enabling the presentation of information (e.g., a user interface for operating peripheral devices and displaying content and information) via one or more output devices 531 associated with the user interface 530 (e.g., a display screen, a speaker, etc.);

[0058] An input processing module 554 for detecting and translating one or more user inputs or interactions from one of one or more input devices 532.

[0059] In some embodiments, the device provided by the embodiments of the present application may be implemented in software. Figure 2 Shown is an information processing device 555 stored in the memory 550, which may be software in the form of programs and plugins, etc., including the following software modules: a display module 5551, a summarization module 5552, a search module 5553, an aggregation module 5554, a subscription module 5555, and a generation module 5556. These modules are logical, so they can be combined arbitrarily or further split according to the functions implemented. It should be noted that, Figure 2 For the convenience of expression, all the above modules are shown at once, but it should not be regarded as excluding the implementation where the information processing device 555 may only include the display module 5551. The functions of each module will be described below.

[0060] The information processing method provided by the embodiments of the present application will be specifically described below in combination with the exemplary applications and implementations of the terminal device provided by the embodiments of the present application.

[0061] Exemplarily, refer to Figure 3 , Figure 3 which is the first flowchart of the information processing method provided by the embodiments of the present application, and will be described in combination with Figure 3 the steps shown.

[0062] It should be noted that, Figure 3 the method shown can be executed by various forms of computer programs running on the terminal device, not limited to the client. For example, it can also be the operating system, software modules, scripts, and applets described above. Therefore, the examples of the client in the following should not be regarded as a limitation to the embodiments of the present application. In addition, for the convenience of expression, the terminal device and the client running on the terminal device will not be specifically distinguished in the following.

[0063] In step 101, an information processing interface is displayed.

[0064] Here, the information processing interface may include AI-based interaction controls. Among them, the interaction controls may include a text input box, an image upload entry (such as a screenshot button), and a file upload entry, etc. That is to say, the user can upload different types of information such as text, images, and files through the interaction controls.

[0065] It should be noted that the above-mentioned AI-based interaction controls mean that the intelligent workbench APP provided in the embodiments of the present application can access an AI model (such as a large language model). After the user inputs information in the interaction control, the large language model can reply to the information input by the user, or can also search for a knowledge base that matches the information input by the user.

[0066] In some embodiments, the intelligent workbench APP can be installed on the user's terminal device. When a click operation on the icon of the intelligent workbench APP is received, the intelligent workbench APP can be run. For example, the interaction controls can be displayed in the information processing interface of the intelligent workbench APP (such as the home page of the intelligent workbench). Among them, the interaction controls can support multiple types of information input. For example, the user can input text in the interaction control (for example, a text input box can be displayed in the interaction control, and the user can input text through the text input box), can also input an image in the interaction control (for example, an image upload entry can be displayed in the interaction control, and the user can upload an image through the image upload entry), or can also input a file in the interaction control (for example, a file upload entry can be displayed in the interaction control, and the user can upload a file through the file upload entry). The embodiments of the present application do not make specific limitations on this.

[0067] In step 102, in response to an input operation in the interaction control, the input first information is displayed, and at least one first knowledge base is displayed.

[0068] Here, the first information can be a problem description text (for example, the user can input a problem in the interaction control), or can also be a statement text (for example, the user can input a solution in the interaction control); at least one first knowledge base can include knowledge related to the first information. Among them, the first knowledge base can be a personal knowledge base (i.e., a private domain knowledge base), or can also be a shared knowledge base (i.e., a public domain knowledge base). The embodiments of the present application do not make specific limitations on the type of the first knowledge base.

[0069] It should be noted that the type of the first information can be not only text, but also images, files, etc. For example, the above-mentioned interaction control can also include an image upload entry (such as a screenshot button), and the user can upload an image to be consulted through the image upload entry; of course, the above-mentioned interaction control can also include a file upload entry, and the user can upload a file to be consulted through the file upload entry. The embodiments of the present application do not make specific limitations on this.

[0070] In addition, it should be noted that while displaying at least one first knowledge base, reply information for replying to the first information can also be displayed. The reply information can be obtained by the conversation model searching network resources based on the first information and integrating the found resources. For example, after the input of the first information is completed, the interface can jump to the tab page of the conversation model, and the first information input by the user, the reply information for replying to the first information output by the conversation model, and at least one first knowledge base matching the first information can be displayed on the tab page of the conversation model.

[0071] In some embodiments, the above step 102 can be implemented in the following manner: in response to an input operation in the interaction control, display the input first information; in response to the completion of the input of the first information, display at least one first knowledge base. That is to say, in the embodiments of the present application, relevant knowledge bases can be recommended after the question is input.

[0072] For example, taking the first information as the first question description text (such as "What must be known about ***") as an example, see Figure 4A , Figure 4A is a schematic diagram of the first application scenario of the information processing method provided by the embodiments of the present application. As Figure 4A shown, the user can input the first question description text in the text input box included in the interaction control 402 displayed on the information processing interface 401. During the process of inputting the first question description text, the first question description text input by the user can be displayed in real time in the text input box of the interaction control 402. After the input of the first question description text is completed (for example, receiving the user's click operation on the enter key, or receiving the period input by the user, or receiving the voice command output by the user indicating the completion of the question input), it can jump to the tab page 403 of the conversation model (such as the Q&A interface of the conversation model), and at least one knowledge base including knowledge related to the question described by the first question description text can be displayed on the tab page 403 of the conversation model.

[0073] In other embodiments, the above step 102 can also be implemented in the following manner: in response to an input operation in the interaction control, display the input first information; during the input process of the first information, in response to the completion of the input of each information unit in the first information, refresh and display at least one first knowledge base matching the currently input information unit. That is to say, in the embodiments of the present application, knowledge bases matching the currently input content can also be searched and displayed in real time during the input process of the question, and when a new knowledge base is found, it can replace the previously displayed knowledge base.

[0074] Exemplarily, taking the first information as the first statement text as an example, assuming that the first statement text includes multiple semantic units (for example, sentences or words), then refer to Figure 4B , Figure 4B is a schematic diagram of a second application scenario of the information processing method provided by an embodiment of the present application. As Figure 4B shown, every time the user inputs a semantic unit in the text input box of the interaction control 402 displayed on the information processing interface 401, a search can be performed based on the semantic unit currently input by the user, and at least one first knowledge base obtained by the current search (that is, a knowledge base including knowledge related to the semantic unit currently input by the user) can be refreshed and displayed below the interaction control 402; for example, when the user inputs the first semantic unit in the text input box of the interaction control, a search can be immediately performed based on the first semantic unit input by the user, and at least one first knowledge base (such as 6 matching knowledge bases) matching the first semantic unit can be displayed below the interaction control 402; then, when the user inputs the second semantic unit in the text input box of the interaction control 402, a search can be performed again based on the second semantic unit, and at least one first knowledge base matching the second semantic unit can be refreshed and displayed below the interaction control. Of course, when the user inputs the second semantic unit in the text input box of the interaction control 402, the first semantic unit and the second semantic unit can also be combined, and a search can be performed based on the combination result, and then at least one first knowledge base (such as 5 matching knowledge bases) matching the combination result can be refreshed and displayed below the interaction control. Among them, the number of first knowledge bases displayed below the interaction control 402 can generally show a trend of decreasing from more to less. This is because as the content input by the user gradually improves, the search intention of the user becomes clearer, so that more accurate search results can be obtained. In this way, by providing a real-time search function, the search cost can be reduced. For example, the user can adjust the input content according to the real-time search results, so as to find the required information faster and more accurately.

[0075] Continue to refer to Figure 4B , after the user finishes inputting the first statement text in the text input box of the interaction control 402 (for example, after the first statement text is input and the user clicks the enter key is received), the information processing interface 401 can be jumped to the tab page 403 of the session model, and at least one first knowledge base (for example, 4 knowledge bases displayed in the tab page 403 of the session model) obtained by searching based on the complete first statement text can be displayed in the tab page 403 of the session model. That is to say, when the first statement text is input, the tab page of the session model (that is, the Q&A page) can be jumped to, and the finally recommended knowledge base can be displayed in the tab page of the session model.

[0076] In some embodiments, the above step 102 may also be implemented in the following manner: in response to an input operation in the interactive control, display the input first information; during the input process of the first information, according to a preset interval duration (for example, assumed to be 4 seconds), refresh and display at least one first knowledge base that matches the information unit currently input in the first information. That is to say, in the embodiments of the present application, during the process of the user inputting a question, every time a set duration elapses, a search is performed and displayed based on the content that has been input currently.

[0077] For example, taking the first information as the first question description text, the user can input the first question description text in the text input box of the interactive control. During the process of the user inputting the first question description text, a search can be performed every set duration (for example, 4 seconds), that is, based on the content currently input by the user (such as a sentence or a word), and at least one first knowledge base obtained by the search (i.e., the knowledge base that matches the content currently input by the user) is displayed below the interactive control. In this way, by providing a real-time search function, the waiting time of the user when inputting a question can be reduced, and the search efficiency is improved. In addition, the user can also adjust the input question according to the search results, so that the information needed can be found more quickly.

[0078] In some embodiments, the above first information may include multiple fields, where the multiple fields may include at least two of the following: field description information (for example, it may be the name, number, resources or links of a field), the field of the historical session (for example, it may be the field of the most recent session, or the field of a session during a previous period of time), materials related to the question (for example, including images, files, etc., where the images can be uploaded through the image upload entry included in the interactive control, and the files can be uploaded through the file upload entry included in the interactive control), question description text (that is, the direct description of the question), and different fields may be separated by delimiters (for example, semicolons, paragraph numbers, etc.). Then, the above step 102 may also be implemented in the following manner: in response to an input operation in the interactive control, display the input first information; during the input process of the first information, in response to the input operation of the delimiter, display at least one first knowledge base that matches the field before the currently input delimiter.

[0079] For example, when the user enters the problem description text and a special delimiter (such as a semicolon) in the text input box of the interactive control, the system can first search based on the problem description text entered by the user and display at least one first knowledge base (i.e., the knowledge base that matches the problem description text entered by the user) below the interactive control. Then, assuming that the user uploads an image through the image upload entry (such as a screenshot button) included in the interactive control and enters a semicolon in the text input box, the system can search based on the uploaded image and refresh the display of at least one first knowledge base obtained from the current search (i.e., the knowledge base that matches the image entered by the user). That is to say, every time the user enters a delimiter, it can trigger the recommendation of knowledge bases in the corresponding field, and the user can adjust the problem according to the currently output knowledge base, so as to more accurately find the information they need.

[0080] In some embodiments, when the number of at least one first knowledge base is multiple, the above-mentioned display of at least one first knowledge base can also be implemented in the following way: Aggregate multiple first knowledge bases to obtain multiple knowledge base sets (for example, first knowledge bases with the same or similar fields can be divided into the same knowledge base set). For example, clustering algorithms such as K-Means, Hierarchical Clustering, Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Spectral Clustering, or Gaussian Mixture Models (GMM) can be used to aggregate multiple first knowledge bases to obtain multiple knowledge base sets. Then, multiple set identifiers (such as set names or set icons) corresponding to the multiple knowledge base sets can be displayed. Each knowledge base set can include multiple first knowledge bases in the same or similar fields. Subsequently, the following processing can also be performed: In response to a trigger operation on the set identifier of the first knowledge base set, display the multiple first knowledge bases included in the first knowledge base set, where the multiple knowledge base sets include the first knowledge base set.

[0081] For example, refer to Figure 4C , Figure 4C is the third application scenario schematic diagram of the information processing method provided by the embodiments of the present application. As Figure 4CAs shown, for the case where the number of at least one first knowledge base is multiple, a clustering algorithm (such as the K-Means algorithm) can be used to aggregate multiple first knowledge bases (for example, assume 100 first knowledge bases) according to the field, obtaining multiple knowledge base sets (for example, assume 3 knowledge base sets), and display the set icons corresponding to the multiple knowledge base sets respectively in the tab page 403 of the session model. Among them, each knowledge base set can include multiple first knowledge bases in the same field or similar fields. When a click operation on the set icon 404 of the first knowledge base set (such as the AI knowledge set) is received by the user, multiple first knowledge bases included in the first knowledge base set can be displayed in the tab page 403 of the session model. In this way, by organizing and classifying a large number of knowledge bases, users can find knowledge related to a specific field more quickly, saving the time for users to screen through a large number of knowledge bases.

[0082] In some embodiments, when the number of at least one first knowledge base is multiple, the above display of at least one first knowledge base can also be achieved in the following way: taking the first knowledge base with the highest matching degree with the first information among the multiple first knowledge bases as the center, arranging and displaying the remaining first knowledge bases in different directions. Among them, the farther the first knowledge base is from the center, the greater the field span between it and the first knowledge base at the center. That is to say, the farther the first knowledge base is from the center, the smaller the matching degree with the first information.

[0083] Exemplarily, see Figure 4D , Figure 4D is the fourth application scenario schematic diagram of the information processing method provided by the embodiments of the present application. As Figure 4D shown, for the case where the number of at least one first knowledge base is multiple, multiple first knowledge bases can be arranged radially according to the field. For example, the first knowledge base with the highest matching degree with the first information (such as the first problem description text) input by the user among the multiple first knowledge bases (for example, assume the knowledge base named "XX Omnidirectional Guide") can be taken as the center, and the remaining multiple first knowledge bases can be arranged and displayed in different directions (for example, different directions can represent different field derivation directions), and the farther the first knowledge base is from the center, the greater the field span between it and the first knowledge base at the center. In this way, it is convenient for users to quickly query knowledge bases in similar fields in the tab page 403 of the session model.

[0084] In some embodiments, the above-mentioned display of at least one first knowledge base can also be implemented in the following manner: display at least one knowledge base identifier (such as a knowledge base name or a knowledge base icon, etc.) corresponding to each of the at least one first knowledge bases, and perform at least one of the following processes: display at least one piece of key information corresponding to each of the at least one first knowledge bases, where the key information can include at least one of the following: the covered field, the knowledge type, the update frequency, and the key information can be displayed automatically by default or after receiving a specific trigger operation (such as a click operation, a hover operation); display at least one preview interface corresponding to each of the at least one first knowledge bases, where the preview interface can include knowledge related to the first information and can support at least one of the following operations: a favorite operation, a copy operation, a share operation, a subscription operation.

[0085] It should be noted that the preview interface can also be displayed automatically by default or only after receiving a specific trigger operation. In addition, the key information of the knowledge base or the preview interface corresponding to the knowledge base can also be automatically displayed only around some of the first knowledge bases. For example, the key information and the preview interface of the knowledge base can be automatically displayed only around the first knowledge base with the highest matching degree with the first information (such as below or on the right side). For example, the preview interface can be displayed on the right side of the knowledge base, and the key information can be displayed below the knowledge base. For other first knowledge bases, the user needs to perform a specific trigger operation (such as a mouse hover operation, a long press operation, etc.) before the corresponding key information and preview interface will be displayed. Among them, the trigger methods of the key information and the preview interface can be the same, that is, when receiving a specific operation triggered by the user, the key information and the preview interface corresponding to the knowledge base can be displayed simultaneously. Of course, the trigger methods of the key information and the preview interface can also be different, and the embodiments of the present application do not make specific limitations on this.

[0086] Exemplarily, taking at least one first knowledge base as four first knowledge bases as an example, assuming they are knowledge bases 1 to 4 respectively, the knowledge base names corresponding to these four knowledge bases can be at least displayed in the tab page of the session model. In addition, key information such as the fields covered by the knowledge base, the depth of knowledge (such as including basic, intermediate, advanced, etc.), the type of knowledge (such as including theoretical knowledge, practical cases, operation guides, etc.), and the update frequency can also be displayed around each knowledge base. And these key information can be displayed automatically by default, or can be displayed only after being triggered by a specific method (such as long pressing, mouse hovering). Of course, a preview interface corresponding to the knowledge base can also be displayed around each knowledge base. In the preview interface, knowledge related to the first information included in the knowledge base can be displayed. At the same time, the preview interface can also support operations such as collecting, copying, sharing, and subscribing to the knowledge base. For example, users can directly collect, copy, or share the knowledge displayed in the preview interface, or users can also subscribe to the knowledge base by clicking the subscribe button displayed in the preview interface. In this way, the operation efficiency of users is effectively improved.

[0087] In some other embodiments, the above-mentioned display of at least one first knowledge base can also be implemented in any of the following ways: displaying at least one first knowledge base in the tab page of the session model; displaying at least one first knowledge base in the tab page of the knowledge base platform; displaying some of the at least one first knowledge base and a view more control in the tab page of the session model; in response to a trigger operation for the view more control, jumping from the tab page of the session model to the tab page of the knowledge base platform, and displaying the remaining first knowledge bases in the at least one first knowledge base in the tab page of the knowledge base platform.

[0088] Exemplarily, after the user inputs the first information in the interaction control included in the information processing interface (such as the home page), the user can jump from the information processing interface to the tab page of the conversation model, and display at least one first knowledge base searched based on the first information in the tab page of the conversation model; or, the user can also jump from the information processing interface to the tab page of the knowledge base platform (such as the knowledge base square), and display at least one first knowledge base searched based on the first information in the tab page of the knowledge base platform. For example, at least one first knowledge base can be prominently displayed in the tab page of the knowledge base platform; or, an entrance to the knowledge base platform can also be displayed in the information processing interface. When the user inputs the first information in the interaction control, a red circle can be displayed on the entrance to the knowledge base platform to prompt the user that there is a recommended knowledge base currently. When a click operation on the entrance to the knowledge base platform is received, the user can jump from the information processing interface to the tab page of the knowledge base platform, and display at least one first knowledge base searched based on the first information in the tab page of the knowledge base platform. In addition, for the case where the number of at least one first knowledge base is multiple (for example, assuming there are a total of 10 first knowledge bases, assumed to be knowledge base 1 to knowledge base 10 respectively), after the user inputs the first information in the interaction control, the user can first jump from the information processing interface to the tab page of the conversation model, and display a small number of first knowledge bases (such as knowledge base 1 to knowledge base 4) and a "View More" button in the tab page of the conversation model. When a click operation on the "View More" button is received, the user can jump from the tab page of the conversation model to the tab page of the knowledge base platform, and display the remaining 6 first knowledge bases (i.e., knowledge base 5 to knowledge base 10) in the tab page of the knowledge base platform. In this way, on the one hand, the page loading speed can be improved. By first displaying a small number of knowledge bases, the page loading speed can be accelerated, improving the user experience; on the other hand, the user can also choose whether to view more results according to their own needs, thereby improving the user's satisfaction and engagement.

[0089] In some other embodiments, when displaying at least one first knowledge base, the following processing can also be performed: display a plurality of direction controls (such as including a horizontal control and a vertical control), wherein different direction controls correspond to different field derivative directions; in response to a trigger operation on the first direction control (such as assuming it is a horizontal control), display the knowledge base of the field corresponding to the first field derivative direction, wherein the first field derivative direction is the field derivative direction corresponding to the first direction control, and the plurality of direction controls include the first direction control.

[0090] It should be noted that in practical applications, the domain derivative direction corresponding to the direction control can be fixed (i.e., the system default, which cannot be changed by the user); of course, the user can also manually set the domain derivative direction corresponding to the direction control. For example, when a long-press operation of the user on any direction control is received, multiple candidate domain derivative directions can be displayed for the user to select, and then the domain derivative direction selected by the user from the multiple candidate domain derivative directions can be used as the domain derivative direction corresponding to the direction control. The embodiments of the present application do not make specific limitations on this.

[0091] Exemplarily, refer to Figure 4E , Figure 4E which is a schematic diagram of the fifth application scenario of the information processing method provided by the embodiments of the present application. As Figure 4E shown, in the tab page 403 of the session model, in addition to displaying at least one first knowledge base (such as 4 knowledge bases) obtained by searching based on the first information, multiple direction controls can also be displayed, such as including a horizontal control 405 and a vertical control 406. Among them, different direction controls correspond to different domain derivative directions. When a click operation of the user on the horizontal control 405 is received, a knowledge base of the domain corresponding to the domain derivative direction corresponding to the horizontal direction (for example, assumed to be domain A) can be displayed in the tab page 403 of the session model to replace at least one first knowledge base; when a click operation of the user on the vertical control 406 is received, a knowledge base of the domain corresponding to the domain derivative direction corresponding to the vertical direction (for example, assumed to be domain B) can be displayed in the tab page 403 of the session model to replace at least one first knowledge base. In this way, the user can quickly obtain relevant knowledge bases through the direction controls, improving the acquisition efficiency of the knowledge bases.

[0092] In some embodiments, the above information processing interface may further include a domain selection control. Then, before performing Figure 3 the step 102 shown, the following processing may further be performed: in response to a domain selection operation based on the domain selection control, the selected domain is displayed, and the above display of at least one first knowledge base may further be implemented in the following manner: displaying at least one first knowledge base including knowledge related to the first information and belonging to the selected domain.

[0093] For example, in addition to the interactive controls, the information processing interface (such as the homepage) can also display a field selection control, that is, before the user enters the first information (such as the problem description text) in the interactive control, he can also select the knowledge field involved. For example, when the user clicks on the field selection control, multiple candidate fields can be displayed for the user to choose from, and then the field selected by the user from the multiple candidate fields can be displayed in the interactive control. The user can then enter the problem description text in the text input box of the interactive control. After the problem description text is entered, a search can be performed based on the problem description text entered by the user, and the knowledge base belonging to the field selected by the user can be screened out from the multiple knowledge bases obtained by the search and displayed. In this way, by first selecting the knowledge field involved in the problem, the final recommended knowledge base can be made more accurate.

[0094] In other embodiments, before displaying at least one first knowledge base, the following processing may also be performed: when the information processing interface is in a first search mode (e.g., "whole network based" mode), at least one first knowledge base including knowledge related to the first information may be searched in order of priority of shared knowledge bases (i.e., public domain knowledge bases) and private knowledge bases (i.e., personal knowledge bases); when the information processing interface is in a second search mode (e.g., "knowledge base based" mode), at least one first knowledge base including knowledge related to the first information may be searched in order of priority of the private knowledge base of the first account (i.e., the user's own personal knowledge base), the shared knowledge base of the first account (i.e., public domain knowledge base), and the private knowledge base of the second account (i.e., other users' personal knowledge bases), wherein the first account is the current login account of the information processing interface (e.g., it may be an instant messaging account).

[0095] Exemplarily, the information processing method provided by the embodiments of the present application may include multiple search modes, such as a "network-wide based" search mode and a "knowledge base based" search mode. When the user inputs the first information in the interaction control, the user can also select a search mode in the interaction control. For example, a search mode selection entry may be displayed in the interaction control. When receiving a click operation of the user on the search mode selection entry, the above-mentioned "network-wide based" search mode and "knowledge base based" search mode can be displayed for the user to select. If the user selects the "network-wide based" search mode, when receiving the first information input by the user in the interaction control, the shared knowledge base platform can be queried first based on the first information input by the user. If there is no shared knowledge base related to the first information in the shared knowledge base platform, the private knowledge base platform can be queried based on the first information. That is to say, for the "network-wide based" search mode, the public domain knowledge base can be preferentially recommended. If there is no relevant public domain knowledge base, the relevant personal knowledge base can be recommended. In addition, if the user selects the "knowledge base based" search mode, when receiving the first information input by the user in the interaction control, if the relevance between the first information input by the user and the currently used knowledge base is relatively low, it can be recommended in the priority order of personal knowledge base (that is, preferentially recommend the knowledge base related to the problem of the user himself), public domain knowledge base, and personal knowledge bases of other users. In this way, the embodiments of the present application provide multiple search modes for the user to select, which can meet the personalized needs of different users.

[0096] It should be noted that for the recommendation of the personal knowledge base, the user's authorization can be obtained in advance (such as including whether to allow other users to use, whether to allow recommendation, etc.). In addition, the usage permission can also be applied to the creator of the knowledge base, where the permission can be subdivided into: permanent usage permission, usage permission based on duration or the number of questions.

[0097] In some embodiments, before performing Figure 3 the step 102 shown, the following processing can also be performed: display at least one second knowledge base, where the at least one second knowledge base includes: a knowledge base matching the historical information input by the current login account (i.e., the first account, such as an instant messaging account) of the information processing interface, a knowledge base matching the feature data of the object (such as user A) associated with the current login account of the information processing interface, a knowledge base used by other accounts (such as the instant messaging account of user B) having a social relationship with the current login account of the information processing interface, and a knowledge base with the number of pushes greater than the number threshold (i.e., a popular knowledge base in the knowledge base platform).

[0098] Exemplarily, before the user inputs the first information (such as a text description of a certain problem), it is also possible to recommend a knowledge base based on some prior information. For example, the user's historical Q&A records can be obtained and a knowledge base that matches the historical Q&A records can be pushed to the user (i.e., the warm start scenario); or, the user's feature data can be obtained and a knowledge base that matches the feature data can be pushed to the user. Among them, the user's feature data can be filled in when the user first registers. For example, when the user first uses the intelligent workbench, a simple questionnaire can be designed through a friendly guiding interface. For example, the user can be asked about their job type (such as student, doctor, engineer, etc.), professional field (such as including computer science, medicine, agriculture, finance, etc.), and the main purpose of using the intelligent workbench (such as including learning new knowledge, solving work problems, conducting research, etc.). Through the above feature data, the user's demand direction can be initially understood, so as to provide a reference for subsequent screening of the knowledge base. Of course, for the case where the user's feature data cannot be obtained (i.e., the cold start scenario), it is also possible to recommend the knowledge base used by friends to the user, or recommend the popular knowledge bases in the knowledge base platform (such as knowledge bases with the number of pushes greater than the number threshold, or knowledge bases with the number of subscriptions greater than the number threshold, etc.). In this way, by analyzing the user's prior information, the knowledge that the user may need can be predicted, thereby reducing the search time and effort of the user when looking for information, improving the efficiency of information acquisition. In addition, by pushing relevant knowledge bases through prior information, the user can be helped to pre-understand relevant knowledge, thereby improving the accuracy of subsequent retrieval.

[0099] In some embodiments, the first information may include a first text. Before displaying at least one first knowledge base, the following processing may also be performed: performing word segmentation on the first text to obtain a plurality of word elements; querying a keyword index based on the plurality of word elements to obtain keywords that match the plurality of word elements (i.e., the keywords that hit the word elements). Among them, the keyword index may be constructed based on a plurality of keywords, and the plurality of keywords are obtained by performing word segmentation on the knowledge in a plurality of knowledge bases pre-created in the knowledge base platform; the knowledge bases in the plurality of knowledge bases that include the matching keywords are used as at least one first knowledge base.

[0100] For example, for each knowledge base stored in the knowledge base platform, a tokenizer can be used to tokenize the knowledge in the knowledge base to obtain multiple words. Then, keywords can be selected from the multiple words (for example, they can be the title, introduction, etc. of the knowledge base). Subsequently, a keyword index can be constructed based on the multiple keywords corresponding to the multiple knowledge bases (for example, indexes can be established separately for the title, content, tags, and pinyin of the title of the knowledge base). In this way, when the first text entered by the user in the text input box of the interaction control is received, the first text can be tokenized to obtain multiple tokens. Then, the keyword index can be queried based on the multiple tokens to obtain the keywords of the tokens that are hit. Finally, the knowledge base including the keywords of the tokens that are hit can be used as the knowledge base that matches the first text. In this way, by establishing a keyword index, the knowledge base related to the first text entered by the user can be quickly recalled.

[0101] It should be noted that the above keyword index can include text indexes in multiple dimensions. For example, for the knowledge bases in the knowledge base platform, multiple indexes can be established from multiple dimensions such as the title, content, tags, and pinyin of the title of the knowledge base. In this way, different search strategies can also be adopted for different fields during subsequent searches. For example, for the search of tags, exact match can be selected (that is, only search the tag index); while for the search of the title, pinyin search can be used at the same time (that is, the title index and the pinyin search of the title can be searched at the same time). In this way, through the integration of multiple search strategies, the recall rate can be further improved.

[0102] In some embodiments, for the case where there are multiple text indexes, embodiments of the present application can also create an index alias for the multiple text indexes. By using this alias to replace the actual name of the index, it can be used in query and index operations. Among them, the index alias is a technology used in a database management system to simplify queries and improve maintainability. By creating an alias for the index, it is more convenient to reference and manage these indexes, especially in the case of multiple indexes or complex queries. That is to say, in order to handle large-scale data, embodiments of the present application can also construct a multi-index mechanism. The multi-index mechanism refers to creating multiple indexes for the same data set in the database. Each index can be designed according to different query requirements, data characteristics, or performance optimization goals. For example, embodiments of the present application can create corresponding multiple indexes for multiple dimensions such as the title, content, and tags of the knowledge base respectively. These indexes can be distributed on different physical nodes or shards to support horizontal expansion and high-concurrency access. Subsequently, an index alias can be created for these indexes. The index alias can be used as a reference pointing to multiple actual indexes, so that only queries can be made through the alias without caring about the specific index distribution and structure at the bottom layer. In addition, the index alias can distribute query requests to multiple indexes to achieve load balancing of queries and improve the overall performance of the system.

[0103] In some other embodiments, following the above, when the field corresponding to the first text is the target field (such as a certain emerging field), the above-mentioned tokenization processing of the first text to obtain multiple tokens can also be achieved in the following manner: collect the vocabulary included in the target field (such as including new words, technical terms, etc.) from the materials of the target field (such as including professional literature, industry reports, news reports, etc.); integrate the collected multiple vocabularies (such as removing duplicates, unifying different descriptions, etc.) to obtain a dictionary corresponding to the target field; perform tokenization processing on the first text according to the vocabulary included in the dictionary to obtain multiple tokens.

[0104] Exemplarily, to meet the retrieval of various new words, embodiments of the present application may also create a private domain noun dictionary. The private domain noun dictionary is a collection of specialized vocabulary constructed for a specific field or business scenario. These words are usually noun terms closely related to the professional knowledge, business processes, specific concepts, etc. of that field. For example, in the medical field, the private domain noun dictionary may include various disease names, drug names, medical device names, etc.; in the financial field, the private domain noun dictionary may include various financial product names, trading terms, policy nouns, etc. With the continuous development and innovation in each field, new concepts, terms, and nouns are constantly emerging. To address this issue, embodiments of the present application create a private domain noun dictionary and include these new words through the private domain noun dictionary, so as to ensure that the dictionary can accurately reflect the latest knowledge and information in the field and provide more comprehensive and accurate vocabulary support for relevant language processing tools (such as the ik word segmenter. The ik word segmenter is an open-source Chinese word segmenter that can segment continuous Chinese text into meaningful word sequences according to certain rules and algorithms, providing basic support for subsequent text analysis, information retrieval, machine learning, etc.). For example, new noun terms can be collected, screened, and sorted by domain experts, professional editors, or those familiar with the field. Specifically, new words can be obtained from channels such as professional literature, industry reports, and academic research according to the development trends of the field, accurately defined and classified, and then added to the private domain noun dictionary; or, natural language processing techniques and text mining algorithms can be used to detect and analyze a large amount of field-related text, automatically extract possible new words by identifying newly emerging words, counting the occurrence frequency and semantic information of words, etc., and adding them to the private domain noun dictionary after manual review. In this way, when a user wants to query knowledge in a certain emerging field, the ik word segmenter can perform word segmentation processing on the first text input by the user through the private domain noun dictionary. That is to say, through the private domain noun dictionary, the ik word segmenter can accurately identify professional terms and new words in the vertical field, so that they can be segmented and processed as a whole, avoiding splitting professional terms into meaningless single characters or words, thereby improving the accuracy and recall rate of indexing.

[0105] In some other embodiments, before displaying at least one first knowledge base, the following processing may also be performed: performing embedding processing on the first information to obtain a first vector; querying a vector index based on the first vector to obtain a second vector that matches the first vector (for example, a second vector whose similarity to the first vector is greater than a similarity threshold), where the vector index may be constructed based on multiple second vectors, and the multiple second vectors are obtained by performing embedding processing on the knowledge in multiple pre-created knowledge bases respectively; using the knowledge bases in the multiple knowledge bases corresponding to the matching second vector as at least one first knowledge base.

[0106] Exemplarily, in view of the diversity of the knowledge included in the knowledge base, embodiments of the present application can also construct a vector index. For example, for each knowledge base stored in the knowledge base platform, an embedding model (such as the bge-m3 model) can be used to perform embedding processing on the knowledge in the knowledge base, so as to convert it into a 1024-dimensional vector. After performing embedding processing on the knowledge in multiple knowledge bases respectively to obtain corresponding multiple 1024-dimensional vectors, a vector index can be established in a database (such as the ElasticSearch database) based on the Hierarchical Navigable Small World (HNSW) algorithm. Among them, the HNSW algorithm is an efficient algorithm for approximate nearest neighbor search in high-dimensional spaces. It is based on the concept of small-world networks and realizes fast search by constructing a hierarchical graph structure. Specifically, a small-world network is a network structure with high clustering and short path lengths. In HNSW, each level of the graph is a small-world network, which means that the connections between nodes are both locally dense and the path length of the entire network is short through a small number of long-distance connections. In addition, HNSW also constructs a graph structure composed of multiple levels. Each level contains a part of the nodes. The top level contains fewer nodes, while the bottom level contains all the nodes. Each node may have different connections in different levels, which enables the search to navigate between different levels. The search starts from a random node at the top level. By comparing the distance between the query point (i.e., the node corresponding to the first vector, where the first vector is obtained by performing embedding processing on the first information input by the user) and the current node, the node with the shortest distance is selected and moved down to the next level in this direction. This process is continuously repeated until the bottom level is reached, so as to find the nearest neighbor (i.e., the second vector that matches the first vector mentioned above). In this way, by converting the knowledge in the knowledge base into vectors, the problem of the diversity of the knowledge included in the knowledge base can be addressed. At the same time, by using the HNSW algorithm to construct the vector index, a high recall rate can be provided while maintaining a high search speed.

[0107] In addition, when inserting a new node (for example, when a user uploads a new knowledge base in the knowledge base platform, the knowledge in the new knowledge base can be embedded to obtain a corresponding 1024-dimensional vector), the HNSW algorithm starts from the top level, finds a suitable insertion position, and establishes connections for the new node in different levels, thus ensuring the dynamicity and scalability of the graph structure.

[0108] In some embodiments, to help users discover more knowledge bases, after receiving the first information input by the user in the interaction control, multi-channel retrieval can also be performed based on the first information. For example, the first information can be embedded to obtain the corresponding first vector, and the vector index can be retrieved based on the first vector. At the same time, a tokenizer can be used to tokenize the first information to obtain multiple tokens, and the keyword index can be retrieved based on the multiple tokens. Subsequently, the knowledge bases recalled based on the vector index and the knowledge bases recalled based on the keyword index can be deduplicated, and the deduplicated knowledge bases can be sorted. For example, the relevance between the knowledge in the deduplicated knowledge bases and the first information can be calculated, and the sorting can be performed in descending order of the relevance scores. Finally, at least one knowledge base ranked at the top in the sorting result can be recommended to the user.

[0109] In other embodiments, to handle the storage and retrieval of large-scale user data, the embodiments of the present application also provide an organizational structure based on multiple clusters and multiple indexes. For example, clusters can be dynamically allocated according to the growth of business data, routed to different indexes according to the user identifier, and then sharded routing can be performed through the user identifier, so as to ensure that the performance of writing and retrieval is within a controllable range to cope with the high-speed growth in the later stage of the business.

[0110] In some embodiments, refer to Figure 5 , Figure 5 is the second process schematic diagram of the information processing method provided by the embodiments of the present application. As Figure 5 shown, after performing the step 102 shown in Figure 3 , the step 103 and step 104 shown in Figure 5 can also be performed, which will be described in combination with the steps shown in Figure 5 .

[0111] In step 103, in the case where none of the at least one first knowledge base is selected, in response to a re-input operation in the interaction control, the re-input second information is displayed.

[0112] In some embodiments, after a user inputs first information (i.e., makes the first query) in an interaction control included in an information processing interface (such as a home page), the user can jump from the information processing interface to a tab page of a conversation model, and at least one first knowledge base (such as four first knowledge bases by assumption) searched based on the first information and an interaction control are displayed in the tab page of the conversation model. If the user is not satisfied with the four first knowledge bases recommended for the first time (for example, the user does not select any of the four first knowledge bases), the user can make a second query in the interaction control displayed in the tab page of the conversation model. For example, after receiving the second information (i.e., making the second query) input by the user in the interaction control displayed in the tab page of the conversation model, the second information input by the user can be displayed in the tab page of the conversation model.

[0113] In step 104, at least one third knowledge base is displayed to replace at least one first knowledge base.

[0114] Here, at least one third knowledge base may include knowledge related to the intention reflected by the information sequence, where the information sequence includes the first information and the second information sorted in the order of input.

[0115] In some embodiments, for the case where the user makes multiple queries. For example, after the user's first query, for the knowledge base recommended for the first time (i.e., the above-mentioned at least one first knowledge base), the user does not select it but makes a second query. In this case, it can be considered that the user does not need the knowledge base recommended for the first time. In this situation, embodiments of the present application can predict the user's query intention through the sequence of intention changes in multiple queries (such as a sequence composed of the question 1 input by the user for the first time and the question 2 input by the user for the second time), and then can push a knowledge base matching the query intention to the user to replace the previously displayed knowledge base. Specifically, for the case where the user makes multiple queries, the questions input by the user multiple times can be combined in the order of input to obtain a corresponding question sequence, and then a sequence prediction algorithm (such as including a hidden Markov model, a recurrent neural network, a long short-term memory network, etc.) can be used to predict the question sequence, so as to obtain the user's true query intention. Subsequently, a search can be performed based on the user's true query intention, and thus a knowledge base that conforms to the user's true query intention can be obtained. In this way, the recommended knowledge base can highly conform to the user's needs.

[0116] In some other embodiments, refer to Figure 6 , Figure 6 is the third process schematic diagram of the information processing method provided by embodiments of the present application. As Figure 6 shown, after executing the Figure 3 steps shown in Figure 6Steps 105 and 106 shown will be described in conjunction with Figure 6 the steps shown.

[0117] In step 105, multiple knowledge bases respectively matched with multiple pieces of information input multiple times are summarized to obtain a summary table.

[0118] Here, the summary table may record the number of times each knowledge base hits the input information. Among them, the multiple pieces of information include the first piece of information, and the multiple knowledge bases include at least one first knowledge base.

[0119] In step 106, the knowledge bases with the hit times greater than the times threshold are highlighted in the summary table, or a prompt message is displayed in the summary table.

[0120] Here, the prompt message can be used to prompt the user to subscribe to the knowledge base with the hit times greater than the times threshold.

[0121] In some embodiments, for the situation where the user asks questions continuously for multiple times, multiple knowledge bases respectively matched with multiple questions input by the user can also be summarized to obtain a summary table. Among them, the summary table may record the number of times each knowledge base hits the questions input by the user. For the knowledge base with the hit times greater than the times threshold (for example, assumed to be 5 times), a prompt for subscribing to the knowledge base can also be given to prompt the user that they can manually subscribe to the knowledge base, or the knowledge base can also be automatically subscribed for the user. In this way, by summarizing multiple knowledge bases and prompting the knowledge bases with high hit times in the summary table, the efficiency of the user in screening knowledge bases is effectively improved.

[0122] In some embodiments, when displaying at least one first knowledge base, the following processing can also be performed: displaying a negative feedback control (such as a "dislike" button) for at least one first knowledge base, where the negative feedback control is used to represent negative emotions (that is, if the user clicks the "dislike" button, it means that the user is not satisfied with the currently recommended knowledge base); in response to a trigger operation on the negative feedback control, displaying at least one fourth knowledge base to replace at least one first knowledge base, where at least one fourth knowledge base includes knowledge related to the first information and different from at least one first knowledge base (for example, the type of knowledge in the fourth knowledge base is different from the type of knowledge in the first knowledge base).

[0123] Exemplarily, taking at least one first knowledge base as four first knowledge bases, for example, knowledge bases 1 to 4, it can be assumed that these four knowledge bases can be displayed in the tab page of the session model. At the same time, a "dislike" button can also be displayed below these four knowledge bases. When a click operation on the "dislike" button is received from the user, it indicates that the user is not satisfied with the currently recommended four knowledge bases. Then, other knowledge bases (for example, assume knowledge bases 5 to 8, where the knowledge included in knowledge bases 5 to 8 is also related to the first information input by the user) can be updated and displayed in the tab page of the session model to replace these four knowledge bases. In this way, by providing a feedback function and responding to the user's feedback in real time, the user experience can be effectively improved.

[0124] In some other embodiments, when displaying at least one first knowledge base, the following processing can also be performed: display a feedback control for the at least one first knowledge base; in response to an input operation in the feedback control, display the input feedback information, where the feedback information can characterize the domain matching degree of the at least one first knowledge base (for example, the user can provide feedback on the domain matching situation of the currently recommended knowledge base in the feedback control, such as whether the domain is too broad or too narrow); display at least one fourth knowledge base to replace the at least one first knowledge base, where the at least one fourth knowledge base is determined based on the feedback information.

[0125] Exemplarily, taking at least one first knowledge base as knowledge base 1, after the user inputs the first information in the interaction control displayed on the information processing interface (such as the home page), the user can jump from the information processing interface to the tab page of the session model, and knowledge base 1 obtained by searching based on the first information is displayed in the tab page of the session model. In addition, a feedback control can also be displayed around (for example, below) knowledge base 1, and the user can input feedback information for knowledge base 1 in the feedback control. The feedback information can be whether the domain matches, whether the domain is too broad or too narrow, etc. After receiving the feedback information input by the user in the feedback control, the feedback information input by the user can be displayed in the tab page of the session model, and then search is performed again by combining the feedback information input by the user and the first information to obtain at least one fourth knowledge base (for example, assume knowledge base 2). Subsequently, knowledge base 2 can be displayed in the tab page of the session model to replace the previously displayed knowledge base 1. In this way, the correction of the recommended knowledge base can be achieved, so that the finally recommended knowledge base can highly meet the user's needs.

[0126] In some embodiments, after displaying at least one first knowledge base, the following processing can also be performed: in response to a selection operation on the at least one first knowledge base, display a detail page of the selected first knowledge base, where the detail page can include a subscription control; in response to a trigger operation on the subscription control, subscribe to the selected first knowledge base.

[0127] Exemplarily, taking the first information as the first problem description text as an example, after the user enters the first problem description text in the text input box included in the interactive control on the information processing interface, the user can jump from the information processing interface to the tab page of the conversation model, and at least one first knowledge base (for example, assumed to be 4 first knowledge bases, namely knowledge bases 1 to 4) searched based on the first problem description text is displayed on the tab page of the conversation model for the user to select. If the user is interested in knowledge base 1, the user can click on knowledge base 1. For example, when receiving the click operation of the user on knowledge base 1, the user can jump from the tab page of the conversation model to the detail page of knowledge base 1, and a subscription control (such as an "Add to Knowledge Base" button) can be displayed on the detail page of knowledge base 1. When receiving the click operation of the user on the "Add to Knowledge Base" button, the user can be added to knowledge base 1 (that is, control the user's account to subscribe to knowledge base 1). In this way, the user can quickly select the knowledge base they are interested in for subscription, thereby improving the efficiency of subscribing to the knowledge base.

[0128] It should be noted that the detail page of knowledge base 1 may also include a Q&A area. If the user has not subscribed to knowledge base 1, the user can only ask a limited number of questions (for example, at most 3 questions) in the Q&A area; if the user has subscribed to knowledge base 1, the user can ask an unlimited number of questions in the Q&A area.

[0129] In some other embodiments, after displaying at least one first knowledge base, the following processing may also be performed: in response to a selection operation on at least one first knowledge base, generate third information for replying to the first information based on the knowledge included in the selected first knowledge base, and display the third information; or, in response to an input operation based on the interactive control, display the input demand information; automatically select a first knowledge base that matches the demand information from at least one first knowledge base; generate third information for replying to the first information based on the knowledge included in the matched first knowledge base, and display the third information.

[0130] Exemplarily, taking the first information as the first problem description text as an example, after the user inputs the first problem description text in the text input box included in the interactive control displayed on the information processing interface, the user can jump from the information processing interface to the tab page of the session model, and display at least one first knowledge base (for example, assuming there are 4 first knowledge bases, namely knowledge base 1 to knowledge base 4) searched based on the first problem description text in the tab page of the session model for the user to select. Suppose the user selects knowledge base 1 from these 4 knowledge bases. For example, upon receiving the click operation of the user on knowledge base 1, a first reply text for replying to the first problem description text can be generated based on the knowledge in knowledge base 1, and the first reply text can be displayed in the tab page of the session model. For example, the first reply text can be displayed below these 4 first knowledge bases. In this way, a more accurate reply can be output in combination with the knowledge base selected by the user. Of course, corresponding knowledge bases can also be automatically selected from the above 4 knowledge bases in combination with the demand information input by the user. For example, in addition to the above 4 knowledge bases being displayed in the tab page of the session model, an interactive control can also be displayed. The user can input demand information in the interactive control. For example, after the user finishes inputting the demand information in the interactive control, the demand information input by the user can be displayed in the tab page of the session model, and a knowledge base (for example, assuming it is knowledge base 3) that matches the demand information can be automatically selected from the above 4 knowledge bases. Subsequently, a first reply text for replying to the first problem description text can be generated based on the knowledge in knowledge base 3, and the first reply text can be displayed in the tab page of the session model. In this way, the further demands of the user can be met. For example, the demand of the user to obtain more in-depth knowledge can be met.

[0131] Next, an exemplary application of the embodiments of the present application in an actual application scenario will be described.

[0132] With the development of artificial intelligence technology, a large number of AI applications have emerged. They provide the capabilities of collecting knowledge bases and AI question answering, but do not support the linkage between the two. That is to say, users cannot discover good knowledge bases during the process of AI question answering. This is because on the one hand, the linkage relationship between the AI question answering and the knowledge base product module is weak, and on the other hand, the discovery of knowledge bases mostly depends on logics such as search and recommendation, but is decoupled from the large model question answering, resulting in that users still rely on traditional search and recommendation for the discovery of knowledge bases and cannot naturally discover good knowledge bases during the process of AI question answering.

[0133] In view of this, the embodiments of the present application provide an information processing method. By connecting (or linking) the relationship between AI question answering and knowledge base discovery, relevant high-quality knowledge bases can be recommended during the process of AI question answering. In this way, the flow and dissemination of knowledge can be promoted, and knowledge can naturally emerge during each question-and-answer process. That is to say, the embodiments of the present application effectively solve the above technical problems by recommending relevant high-quality knowledge bases during the process of AI question answering, thereby improving the efficiency of users publishing information and obtaining information.

[0134] The following specifically describes the information processing method provided by the embodiments of the present application.

[0135] In some embodiments, refer to Figure 7 , Figure 7 which is the sixth application scenario schematic diagram of the information processing method provided by the embodiments of the present application. As Figure 7 shown, the user can publish the shared knowledge base 702 created by himself / herself to the knowledge base square 703 through the knowledge base permission management component 701. For example, three states (such as including the "default" state, the "publish to the square" state, and the "convert to private" state) can be displayed on the knowledge base permission management component 701 for the user to select. When the user selects the "publish to the square" state on the knowledge base permission management component 701, the currently selected shared knowledge base (i.e., the shared knowledge base 702) among the multiple shared knowledge bases created by the user can be published to the knowledge base square 703. At this time, other users can see the shared knowledge base 702 published by the current user in the knowledge base square 703. In addition, the user can also enter the knowledge base square 703 through the "discovery" entrance and find the shared knowledge base that he / she wants to join through the recommendation function.

[0136] In other embodiments, during the AI question answering process in multiple scenarios such as the home page or sidebar of the intelligent workbench, the public domain knowledge base can be recommended according to the content of the AI question answering. In this way, by recommending the knowledge base during the answering process, the user can discover the knowledge base closely related to the question and answer, and then can better obtain the answer related to the question. And through this natural recommendation method, high-quality and interesting knowledge base content can be better discovered.

[0137] For example, refer to Figure 8 , Figure 8 which is the seventh application scenario schematic diagram of the information processing method provided by the embodiments of the present application. As Figure 8 shown, in addition to the question 801 input by the user and the answer 802 output by the AI model (such as a large language model) (for example, the large language model can perform a full-network search based on the question 801 input by the user, such as including public domain knowledge bases, personal knowledge bases, websites, etc.) displayed on the question-and-answer page, multiple recommended knowledge bases (such asFigure 8 Four knowledge bases shown in the figure) for the user to select. If the user is interested in knowledge base 803 among these four knowledge bases, the user can click on knowledge base 803. When receiving the click operation of the user on knowledge base 803 displayed on the Q&A page, it is possible to jump from the Q&A page to the details page 804 of knowledge base 803. A "Join Knowledge Base" button can be displayed on the details page 804 of knowledge base 803, and the user can join knowledge base 803 by clicking the "Join Knowledge Base" button (that is, controlling the user's account to subscribe to knowledge base 803, so as to obtain the permission to view the knowledge in knowledge base 803). If the user has not joined knowledge base 803 (that is, the user has not clicked the "Join Knowledge Base" button), then only a limited number of rounds of questions can be asked on the details page 804 of knowledge base 803 (for example, only 3 rounds of questions are supported when not joining the knowledge base); if the user has joined knowledge base 803, the limit on the number of questions can be cancelled, that is, the user can ask an unlimited number of questions on the details page 804 of knowledge base 803.

[0138] In some embodiments, the solutions for knowledge base publishing and discovery can be divided into two parts, for example, including publishing permission management and discovery page retrieval. In the publishing permission management module, in order to implement permission control for sharing knowledge bases with different granularities, the embodiments of the present application design three different shared knowledge base states and corresponding permission control mechanisms. In addition, in the discovery page retrieval module, in order to cope with the storage of a large amount of user data, the embodiments of the present application design an extensible index management scheme, so as to facilitate subsequent expansion. At the same time, in order to improve the matching accuracy, the embodiments of the present application also establish a text index for knowledge bases related to tag recall.

[0139] The publishing permission management process will be described first below.

[0140] In some embodiments, for shared knowledge bases, the embodiments of the present application design three different states, including: default, published to the square, and transferred to private. Under different states, different users have different available functions and permissions for shared knowledge bases. For example, as Figure 9 shown, when the shared knowledge base is in the default state, the creator of the shared knowledge base has the permissions of knowledge base core write, knowledge base write, knowledge write, knowledge base read, and knowledge read. The administrator of the knowledge base square has the permissions of knowledge base write, knowledge write, knowledge base read, and knowledge read, while the members of the knowledge base square only have the permissions of knowledge base read and knowledge read; as Figure 10 shown, when the shared knowledge base is in the state of being published to the square, the creator of the shared knowledge base has the permissions of knowledge base core write, knowledge base write, knowledge write, knowledge base read, and knowledge read. The administrator of the knowledge base square has the permissions of knowledge base write, knowledge write, knowledge base read, and knowledge read, while the members of the knowledge base square only have the permissions of knowledge base read and knowledge read; asFigure 11 As shown, when the shared knowledge base is in the process of being converted to a private state, the creator of the shared knowledge base has the permissions of core knowledge base writing, knowledge base writing, knowledge writing, knowledge base reading, and knowledge reading. The administrator of the knowledge base square has no permissions, and the members of the knowledge base square also have no permissions.

[0141] It should be noted that a shared knowledge base in the default state cannot be discovered by other users in the knowledge base square, that is, other users cannot search for a shared knowledge base in the default state in the knowledge base square. However, the creator of the shared knowledge base can manually share the shared knowledge base with other users through links or QR codes. A shared knowledge base in the state of being published to the square can be discovered by other users, and after the user publishes the shared knowledge base to the square, it will no longer occupy the personal space capacity. In addition, a shared knowledge base in the state of being transferred to a private state is only visible to the creator himself / herself, and other users have no permissions.

[0142] The following continues to describe the process of index establishment.

[0143] In some embodiments, for the scenario of recommending knowledge bases during the AI question-and-answer process, for the shared knowledge bases in the knowledge base square, in order to meet various retrieval requirements in the business scenario, such as filtering, keyword retrieval, sorting, and recommendation, etc., the embodiments of the present application can respectively construct corresponding multiple indexes on multiple dimensions such as the title, introduction, and other metadata of the knowledge base. Among them, the index scheme can include keyword inverted index, vector index, etc. In addition, in order to meet the retrieval of various new words, the embodiments of the present application also maintain a private domain noun dictionary to improve the index construction effect of the ik tokenizer in the vertical field. In addition, in order to support the retrieval efficiency of a large number of vectors, the embodiments of the present application can also use the HNSW algorithm to construct corresponding vector indexes for all vector data, so as to achieve millisecond-level retrieval of data at the billion level while meeting the business recall effect.

[0144] In other embodiments, refer to Figure 12 , Figure 12 is a schematic diagram of the organizational structure of the index provided by the embodiments of the present application. As Figure 12 shown, in order to cope with large-scale (such as TB-level) data, the embodiments of the present application construct a multi-index mechanism, which is accessed through an index alias (one index alias can point to multiple indexes). The multi-index mechanism can cope with the subsequent rapid growth of data, and the embodiments of the present application also set a relatively large shard capacity for the index to ensure that the subsequent cluster bearing capacity can be increased through node expansion.

[0145] In some embodiments, to help users discover more knowledge bases, the embodiments of the present application can perform multi-channel retrieval based on the questions (queries) input by users. For example, keyword and vector retrieval can be used to recall relevant content such as titles and introductions in the knowledge bases, and then the recalled knowledge bases are sorted. The knowledge bases ranked at the top in the sorting results are pushed to the users. In this way, the exploration boundary of users for knowledge bases is expanded, which also helps the circulation and consumption of knowledge bases on the platform, and at the same time can generate positive incentives for the creators of knowledge bases.

[0146] In other embodiments, the embodiments of the present application can also perform full-link detection on the recommended knowledge bases. For example, objective indicators such as click-through rate, first-screen hit ratio (which refers to the proportion of users clicking on documents related to the query term on the first page of search results and is usually used to measure user satisfaction with search results and the relevance of search engines), recall rate, and accuracy rate can be used to automatically evaluate the overall quality of a single knowledge base. The quality score is used to affect the recommendation and recall for users, realizing a recommendation system with a self-feedback adjustment mechanism.

[0147] In summary, the information processing method provided by the embodiments of the present application has the following beneficial effects:

[0148] Through the linkage between the AI Q&A and the knowledge base module, the flow and dissemination of knowledge are promoted, enabling knowledge to emerge naturally in each Q&A process, effectively improving the efficiency of users in obtaining knowledge.

[0149] Next, the implementation of the information processing device 555 provided by the embodiments of the present application as a software module will be further described. In some embodiments, as Figure 2 shown, the software module in the information processing device 555 stored in the memory 550 may include: a display module 5551.

[0150] The display module 5551 is used to display an information processing interface, where the information processing interface includes interaction controls based on artificial intelligence; the display module 5551 is further used to, in response to an input operation in the interaction control, display the input first information and display at least one first knowledge base, where at least one first knowledge base includes knowledge related to the first information.

[0151] In some embodiments, the display module 5551 is further configured to perform any one of the following processes: in response to an input operation in the interaction control, display the input first information, and in response to the completion of the input of the first information, display at least one first knowledge base; in response to an input operation in the interaction control, display the input first information, and during the input process of the first information, in response to the completion of the input of each information unit in the first information, refresh and display at least one first knowledge base that matches the currently input information unit; in response to an input operation in the interaction control, display the input first information, and during the input process of the first information, according to a preset interval duration, refresh and display at least one first knowledge base that matches the currently input information unit in the first information.

[0152] In some embodiments, before the display module 5551, in response to an input operation in the interaction control, displays the input first information and displays at least one first knowledge base, it is further configured to display at least one second knowledge base, where the at least one second knowledge base includes: a knowledge base that matches the historical information input by the first account, a knowledge base that matches the feature data of the object associated with the first account, a knowledge base used by other accounts that have a social relationship with the first account, a knowledge base whose pushed times are greater than a times threshold, and the first account is the currently logged-in account of the information processing interface.

[0153] In some embodiments, after the display module 5551 displays at least one first knowledge base, in the case where none of the at least one first knowledge bases is selected, it is further configured to, in response to a re-input operation in the interaction control, display the re-input second information, and display at least one third knowledge base to replace the at least one first knowledge base, where the at least one third knowledge base includes knowledge related to the intention reflected by the information sequence, and the information sequence includes the first information and the second information arranged in the order of input.

[0154] In some embodiments, the information processing device 555 further includes a summarization module 5552, configured to, after the display module 5551 displays at least one first knowledge base, summarize multiple knowledge bases respectively matching multiple inputs of multiple information to obtain a summary table, where the summary table records the number of times each knowledge base hits the input information, the multiple information includes the first information, and the multiple knowledge bases include at least one first knowledge base; the display module 5551 is further configured to highlight in the summary table the knowledge bases whose hit times are greater than the times threshold, or display a prompt message in the summary table, where the prompt message is used to prompt subscribing to the knowledge bases whose hit times are greater than the times threshold.

[0155] In some embodiments, when the display module 5551 displays at least one first knowledge base, it is further configured to display a negative feedback control for the at least one first knowledge base, where the negative feedback control is used to represent negative emotions; the display module 5551 is further configured to, in response to a trigger operation on the negative feedback control, display at least one fourth knowledge base to replace the at least one first knowledge base, where the at least one fourth knowledge base includes knowledge related to the first information and different from the at least one first knowledge base.

[0156] In some embodiments, when the display module 5551 displays at least one first knowledge base, it is further configured to display a feedback control for the at least one first knowledge base; in response to an input operation in the feedback control, display the input feedback information, where the feedback information characterizes the domain matching degree of the at least one first knowledge base; display at least one fourth knowledge base to replace the at least one first knowledge base, where the at least one fourth knowledge base is determined based on the feedback information.

[0157] In some embodiments, the first information includes multiple fields, and the multiple fields include at least two of the following: domain description information, the domain of the historical session, materials related to the question, question description text; and different fields are separated by a delimiter; the display module 5551 is further configured to, in response to an input operation in the interaction control, display the input first information, and during the input process of the first information, in response to an input operation of the delimiter, display at least one first knowledge base that matches the field located before the currently input delimiter.

[0158] In some embodiments, the information processing interface further includes a domain selection control; before the display module 5551 displays the input first information in response to an input operation in the interaction control, it is further configured to, in response to a domain selection operation based on the domain selection control, display the selected domain; the display module 5551 is further configured to display at least one first knowledge base that includes knowledge related to the first information and belongs to the selected domain.

[0159] In some embodiments, the information processing device 555 further includes a search module 5553, which is configured to perform the following processing before the display module 5551 displays at least one first knowledge base: in the case where the information processing interface is in the first search mode, search for at least one first knowledge base that includes knowledge related to the first information in the priority order of the shared knowledge base and the private knowledge base; in the case where the information processing interface is in the second search mode, search for at least one first knowledge base that includes knowledge related to the first information in the priority order of the private knowledge base of the first account, the shared knowledge base of the first account, and the private knowledge base of the second account, where the first account is the currently logged-in account of the information processing interface.

[0160] In some embodiments, the information processing device 555 further includes an aggregation module 5554, configured to aggregate multiple first knowledge bases to obtain multiple knowledge base sets when the number of at least one first knowledge base is multiple; and a display module 5551, further configured to display multiple set identifiers respectively corresponding to the multiple knowledge base sets, where each knowledge base set includes multiple first knowledge bases in the same or similar fields.

[0161] In some embodiments, after the display module 5551 displays the multiple set identifiers respectively corresponding to the multiple knowledge base sets, it is further configured to display multiple first knowledge bases included in the first knowledge base set in response to a trigger operation on the set identifier of the first knowledge base set, where the multiple knowledge base sets include the first knowledge base set.

[0162] In some embodiments, when the number of at least one first knowledge base is multiple, the display module 5551 is further configured to arrange and display the remaining first knowledge bases in different directions with the first knowledge base having the highest matching degree with the first information among the multiple first knowledge bases as the center, where the greater the field span between the first knowledge base farther from the center and the first knowledge base at the center.

[0163] In some embodiments, when the display module 5551 displays at least one first knowledge base, it is further configured to display multiple direction controls, where different direction controls correspond to different field derivation directions; and to display the knowledge bases in the field corresponding to the first field derivation direction in response to a trigger operation on the first direction control, where the first field derivation direction is the field derivation direction corresponding to the first direction control, and the multiple direction controls include the first direction control.

[0164] In some embodiments, the display module 5551 is further configured to display at least one knowledge base identifier respectively corresponding to at least one first knowledge base, and perform at least one of the following processes: display at least one key information respectively corresponding to at least one first knowledge base, where the key information includes at least one of the following: the covered field, knowledge type, update frequency, and the key information is displayed by default automatically or after receiving a trigger operation; display at least one preview interface respectively corresponding to at least one first knowledge base, where the preview interface includes knowledge related to the first information, and supports at least one of the following operations: favorite operation, copy operation, share operation, subscription operation.

[0165] In some embodiments, the display module 5551 is further configured to perform one of the following processes: display at least one first knowledge base in a tab page of a session model; display at least one first knowledge base in a tab page of a knowledge base platform; display a partial first knowledge base in at least one first knowledge base and a view more control in a tab page of a session model; in response to a trigger operation on the view more control, jump from the tab page of the session model to the tab page of the knowledge base platform, and display the remaining first knowledge bases in at least one first knowledge base in the tab page of the knowledge base platform.

[0166] In some embodiments, after the display module 5551 displays at least one first knowledge base, it is further configured to, in response to a selection operation on at least one first knowledge base, display a detail page of the selected first knowledge base, where the detail page includes a subscription control; the information processing device 555 further includes a subscription module 5555, configured to subscribe to the selected first knowledge base in response to a trigger operation on the subscription control.

[0167] In some embodiments, the information processing device 555 further includes a generation module 5556, configured to, after the display module 5551 displays at least one first knowledge base, in response to a selection operation on at least one first knowledge base, generate third information for replying to the first information based on the knowledge included in the selected first knowledge base; the display module 5551 is further configured to display the third information; or, the display module 5551 is further configured to, in response to an input operation based on an interaction control, display the input demand information; the generation module 5556 is further configured to select a first knowledge base matching the demand information from at least one first knowledge base, and generate third information for replying to the first information based on the knowledge included in the matching first knowledge base; the display module 5551 is further configured to display the third information.

[0168] In some embodiments, the first information includes first text; the search module 5553 is further configured to, before the display module 5551 displays at least one first knowledge base, perform word segmentation on the first text to obtain a plurality of tokens; query a keyword index based on the plurality of tokens to obtain keywords matching the plurality of tokens, where the keyword index is constructed based on a plurality of keywords, and the plurality of keywords are obtained by performing word segmentation on the knowledge in a plurality of pre-created knowledge bases; use the knowledge bases in the plurality of knowledge bases that include the matching keywords as at least one first knowledge base.

[0169] In some embodiments, when the field corresponding to the first text is a target field, the search module 5553 is further configured to collect the vocabulary included in the target field from the materials of the target field; integrate the collected plurality of vocabularies to obtain a corresponding dictionary; perform word segmentation on the first text according to the vocabulary included in the dictionary to obtain a plurality of tokens.

[0170] In some embodiments, the search module 5553 is further configured to, before the display module 5551 displays at least one first knowledge base, perform embedding processing on the first information to obtain a first vector; query a vector index based on the first vector to obtain a second vector that matches the first vector, where the vector index is constructed based on a plurality of second vectors, and the plurality of second vectors are obtained by performing embedding processing on the knowledge in a plurality of pre-created knowledge bases respectively; and use the knowledge bases corresponding to the matching second vectors in the plurality of knowledge bases as the at least one first knowledge base.

[0171] It should be noted that the description of the device in the embodiments of the present application is similar to the description of the above method embodiments, and has similar beneficial effects as the method embodiments, so details will not be repeated. For the technical details not described in the information processing device provided in the embodiments of the present application, reference can be made to Figure 3 , Figure 5 , or Figure 6 any of the descriptions of the accompanying drawings.

[0172] The embodiments of the present application provide a computer program product, which includes a computer program or computer-executable instructions, and the computer program or computer-executable instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer-executable instructions from the computer-readable storage medium, and the processor executes the computer-executable instructions, so that the computer device executes the information processing method described above in the embodiments of the present application.

[0173] The embodiments of the present application provide a computer-readable storage medium storing computer-executable instructions, where the computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, the processor will be caused to execute the information processing method provided in the embodiments of the present application. For example, as Figure 3 , Figure 5 , or Figure 6 shown in the information processing method.

[0174] In some embodiments, the computer-readable storage medium may be a FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disc, or CD-ROM, etc.; or may be various devices including one or any combination of the above memories.

[0175] In some embodiments, the executable instructions may be in the form of a program, software, software module, script, or code, and may be written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including being deployed as an independent program or being deployed as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0176] As an example, the executable instructions can be deployed to execute on an electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed at multiple locations and interconnected by a communication network.

[0177] As described above, the above are only embodiments of the present application and are not intended to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and scope of the present application are all included in the protection scope of the present application.

Claims

1. An information processing method, characterized in that The method includes: Displaying an information processing interface, where the information processing interface includes an interaction control based on artificial intelligence; In response to an input operation in the interaction control, displaying the input first information, and Displaying at least one first knowledge base, where the at least one first knowledge base includes knowledge related to the first information.

2. The method according to claim 1, wherein The step of, in response to an input operation in the interaction control, displaying the input first information and displaying at least one first knowledge base includes: Performing any one of the following processes: In response to an input operation in the interaction control, displaying the input first information, and in response to the completion of the input of the first information, displaying at least one first knowledge base; In response to an input operation in the interaction control, displaying the input first information, and during the input process of the first information, in response to the completion of the input of each information unit in the first information, refreshing and displaying at least one first knowledge base that matches the currently input information unit; In response to an input operation in the interaction control, displaying the input first information, and during the input process of the first information, refreshing and displaying at least one first knowledge base that matches the currently input information unit in the first information according to a preset interval duration.

3. The method according to claim 1, characterized in that Before the step of, in response to an input operation in the interaction control, displaying the input first information and displaying at least one first knowledge base, the method further includes: Displaying at least one second knowledge base, where the at least one second knowledge base includes: a knowledge base that matches the historical information input by the first account, a knowledge base that matches the feature data of the object associated with the first account, a knowledge base used by other accounts that have a social relationship with the first account, a knowledge base whose number of pushes is greater than a threshold number, and the first account is the currently logged-in account of the information processing interface.

4. The method according to claim 1, characterized in that After the step of displaying at least one first knowledge base, the method further includes: In the case where none of the at least one first knowledge bases is selected, in response to a re-input operation in the interaction control, displaying the re-input second information, and Displaying at least one third knowledge base to replace the at least one first knowledge base, where the at least one third knowledge base includes knowledge related to the intention reflected by the information sequence, and the information sequence includes the first information and the second information arranged in the order of input.

5. The method according to claim 1, wherein After the step of displaying at least one first knowledge base, the method further includes: Summarizing multiple knowledge bases respectively matched with multiple pieces of input information to obtain a summary table, where the summary table records the number of times each knowledge base hits the input information, the multiple pieces of information include the first information, and the multiple knowledge bases include the at least one first knowledge base; Highlighting in the summary table the knowledge bases whose hit times are greater than the threshold number, or displaying a prompt message in the summary table, where the prompt message is used to prompt subscribing to the knowledge bases whose hit times are greater than the threshold number.

6. The method according to claim 1, wherein When displaying at least one first knowledge base, the method further includes: Display a negative feedback control for the at least one first knowledge base, where the negative feedback control is used to represent negative emotions; In response to a trigger operation on the negative feedback control, display at least one fourth knowledge base to replace the at least one first knowledge base, where the at least one fourth knowledge base includes knowledge related to the first information and different from the at least one first knowledge base.

7. The method according to claim 1, characterized in that, When displaying the at least one first knowledge base, the method further includes: Display a feedback control for the at least one first knowledge base; In response to an input operation in the feedback control, display the input feedback information, where the feedback information characterizes the domain matching degree of the at least one first knowledge base. Display at least one fourth knowledge base to replace the at least one first knowledge base, where the at least one fourth knowledge base is determined based on the feedback information.

8. The method according to claim 1, characterized in that, The first information includes multiple fields, and the multiple fields include at least two of the following: domain description information, domain of the historical session, materials related to the question, question description text; and different fields are separated by a separator symbol; The displaying the input first information and displaying at least one first knowledge base in response to an input operation in the interaction control includes: In response to an input operation in the interaction control, display the input first information. During the input process of the first information, in response to an input operation of the separator symbol, display at least one first knowledge base that matches the field before the currently input separator symbol.

9. The method according to claim 1, wherein: The information processing interface further includes a domain selection control; Before the displaying the input first information in response to an input operation in the interaction control, the method further includes: In response to a domain selection operation based on the domain selection control, display the selected domain; The displaying at least one first knowledge base includes: Display at least one first knowledge base including knowledge related to the first information and belonging to the selected domain.

10. The method according to claim 1, characterized in that, Before the displaying at least one first knowledge base, the method further includes: In the case where the information processing interface is in the first search mode, search for at least one first knowledge base including knowledge related to the first information in the priority order of the shared knowledge base and the private knowledge base; In the case where the information processing interface is in the second search mode, search for at least one first knowledge base including knowledge related to the first information in the priority order of the private knowledge base of the first account, the shared knowledge base of the first account, and the private knowledge base of the second account, where the first account is the currently logged-in account of the information processing interface.

11. The method according to any one of claims 1 to 10, wherein: When the number of the at least one first knowledge base is multiple, the displaying at least one first knowledge base includes: Aggregate the multiple first knowledge bases to obtain multiple knowledge base sets; Display a plurality of set identifiers respectively corresponding to the plurality of knowledge base sets, wherein each of the knowledge base sets includes a plurality of first knowledge bases in the same or similar fields.

12. The method according to claim 11, characterized in that, After displaying the plurality of set identifiers respectively corresponding to the plurality of knowledge base sets, the method further includes: In response to a trigger operation on the set identifier of the first knowledge base set, display the plurality of first knowledge bases included in the first knowledge base set, wherein the plurality of knowledge base sets includes the first knowledge base set.

13. The method according to any one of claims 1 to 10, wherein When the number of the at least one first knowledge base is multiple, the displaying of the at least one first knowledge base includes: Taking the first knowledge base with the highest matching degree with the first information among the multiple first knowledge bases as the center, arranging and displaying the remaining first knowledge bases in different directions, wherein the farther the first knowledge base is from the center, the greater the field span between it and the first knowledge base at the center.

14. The method according to any one of claims 1 to 10, characterized in that, When displaying the at least one first knowledge base, the method further includes: Display a plurality of direction controls, wherein different direction controls correspond to different field derivation directions; In response to a trigger operation on the first direction control, display the knowledge bases in the field corresponding to the first field derivation direction, wherein the first field derivation direction is the field derivation direction corresponding to the first direction control, and the plurality of direction controls includes the first direction control.

15. The method according to any one of claims 1 to 10, characterized in that The displaying of the at least one first knowledge base includes: Display at least one knowledge base identifier respectively corresponding to the at least one first knowledge base, and perform at least one of the following processes: Display at least one key information respectively corresponding to the at least one first knowledge base, wherein the key information includes at least one of the following: the covered field, knowledge type, update frequency, and the key information is displayed by default automatically or after receiving a trigger operation; Display at least one preview interface respectively corresponding to the at least one first knowledge base, wherein the preview interface includes knowledge related to the first information and supports at least one of the following operations: favorite operation, copy operation, share operation, subscription operation.

16. The method according to any one of claims 1 to 10, characterized in that, The displaying of the at least one first knowledge base includes: Perform one of the following processes: Display the at least one first knowledge base in the tab page of the session model; Display the at least one first knowledge base in the tab page of the knowledge base platform; Display some of the at least one first knowledge bases and a view more control in the tab page of the session model; in response to a trigger operation on the view more control, jump from the tab page of the session model to the tab page of the knowledge base platform, and display the remaining first knowledge bases of the at least one first knowledge base in the tab page of the knowledge base platform.

17. The method according to any one of claims 1 to 10, characterized in that After displaying the at least one first knowledge base, the method further includes: In response to a selection operation on the at least one first knowledge base, display the detail page of the selected first knowledge base, wherein the detail page includes a subscription control. Subscribe to the selected first knowledge base in response to a trigger operation on the subscription control.

18. The method according to any one of claims 1 to 10, characterized in that, After displaying the at least one first knowledge base, the method further includes: In response to a selection operation on the at least one first knowledge base, generate third information for replying to the first information based on the knowledge included in the selected first knowledge base, and display the third information; or, In response to an input operation based on the interaction control, display the input demand information, select a first knowledge base that matches the demand information from the at least one first knowledge base, generate third information for replying to the first information based on the knowledge included in the matched first knowledge base, and display the third information.

19. The method according to any one of claims 1 to 10, characterized in that The first information includes first text; Before displaying the at least one first knowledge base, the method further includes: Perform word segmentation on the first text to obtain a plurality of tokens; Query a keyword index based on the plurality of tokens to obtain keywords that match the plurality of tokens, wherein the keyword index is constructed based on a plurality of keywords, and the plurality of keywords are obtained by performing word segmentation on the knowledge in a plurality of pre-created knowledge bases; Use the knowledge bases among the plurality of knowledge bases that include the matched keywords as the at least one first knowledge base.

20. The method according to claim 19, characterized in that When the field corresponding to the first text is a target field, the performing word segmentation on the first text to obtain a plurality of tokens includes: Collect the vocabulary included in the target field from the materials of the target field; Integrate the collected plurality of vocabularies to obtain a corresponding dictionary; Perform word segmentation on the first text according to the vocabulary included in the dictionary to obtain a plurality of tokens.

21. The method according to any one of claims 1 to 10, characterized in that, Before displaying the at least one first knowledge base, the method further includes: Perform embedding processing on the first information to obtain a first vector; Query a vector index based on the first vector to obtain a second vector that matches the first vector, wherein the vector index is constructed based on a plurality of second vectors, and the plurality of second vectors are obtained by performing embedding processing on the knowledge in a plurality of pre-created knowledge bases respectively; Use the knowledge bases among the plurality of knowledge bases that correspond to the matched second vectors as the at least one first knowledge base.

22. An information processing apparatus, characterized in that, The apparatus includes: A display module for displaying an information processing interface, wherein the information processing interface includes an interaction control based on artificial intelligence; The display module is further configured to display the input first information and display at least one first knowledge base in response to an input operation in the interaction control, wherein the at least one first knowledge base includes knowledge related to the first information.

23. An electronic device, characterized in that, Includes: A memory for storing computer-executable instructions; A processor for implementing the information processing method according to any one of claims 1 to 21 when executing the computer-executable instructions stored in the memory.

24. A computer-readable storage medium storing computer-executable instructions, characterized in that, When the computer-executable instructions are executed by a processor, the information processing method according to any one of claims 1 to 21 is implemented.

25. A computer program product, comprising a computer program or computer-executable instructions, characterized in that, When the computer program or computer-executable instructions are executed by a processor, the information processing method according to any one of claims 1 to 21 is implemented.

Images