Knowledge acquisition method, device, medium, program product and system

Abstract

The invention relates to the technical field of computers, and discloses a knowledge acquisition method and device, a medium, a program product and a system. In the method, a first terminal device generates answer content based on knowledge information related to question information determined by a cloud side device and an end side user portrait. Compared with the knowledge obtained directly based on the question information, the correlation between the answer content and the content in which the user is interested is stronger. The user does not need to perform other operations such as re-acquisition and page turning to acquire interested contents, so that the use experience of the user can be improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a knowledge acquisition method, device, medium, program product, and system. Background Technology

[0002] Users can acquire knowledge through applications that include knowledge acquisition features. For example, users can obtain corresponding encyclopedic knowledge by asking questions in the form of text, voice, etc., in encyclopedia applications (hereinafter referred to as encyclopedia applications). The encyclopedia application obtains and returns multiple encyclopedia entries and / or encyclopedia content related to the user's question input for the user to view.

[0003] Current knowledge acquisition typically involves matching user input or semantic information related to the user's input with data in a database and returning encyclopedic knowledge that matches the user's input as the answer to the user's question.

[0004] However, different users may have different interests, and the same question input from different users may result in completely identical answers. This leads to the application returning rather monotonous answers to the same question input from different users, resulting in a poor user experience. Summary of the Invention

[0005] The purpose of this application is to provide a knowledge acquisition method, device, medium, program product, and system.

[0006] The first aspect of this application provides a knowledge acquisition method, comprising: a first terminal device receiving a question input for acquiring knowledge; the first terminal device sending question information to a cloud-side device based on the question input; the first terminal device receiving knowledge information from the cloud-side device, wherein the knowledge information is generated based on the question information; and the first terminal device generating answer content based on the knowledge information and a first user profile, wherein the first user profile is determined based on a first historical record of a user operating at at least one terminal device.

[0007] In this embodiment, the question information can be the question input itself, or it can be the feature information of the question input. The feature information includes, but is not limited to, semantic features, image features, audio features, or feature vectors obtained based on the above features. The first user profile can be a client-side user profile. The user can be a specific user, such as the first user.

[0008] It's understandable that the answers are generated based on knowledge and user profiles on the device. Compared to knowledge not determined by user profiles, the answers are more relevant to content that users are interested in. This eliminates the need for users to re-obtain or perform other actions like flipping through pages to find content of interest, thus improving the user experience.

[0009] In one possible implementation of the first aspect above, the first terminal device generates answer content based on knowledge information and a first user profile, including: the first terminal device generates answer content through a generative artificial intelligence model based on the knowledge information and the first user profile; or, the first terminal device filters and / or sorts multiple contents in the knowledge information based on the first user profile, thereby generating answer content.

[0010] In this embodiment, the first terminal device determines a knowledge acquisition plan (or strategy) based on the question information and the user profile, and then inputs the knowledge acquisition plan and knowledge information into a generative artificial intelligence model, which generates the answer content. Alternatively, the first terminal device determines the priority of multiple content modules in the knowledge information based on the user profile, and sorts the multiple content modules in the knowledge information according to the priority (e.g., descending order) to obtain the answer content. The higher the priority, the higher the degree of matching between the corresponding content module and the user profile, that is, the greater the probability that it is content that the user is interested in.

[0011] In one possible implementation of the first aspect above, the content of the first historical record includes at least one of the following: at least one application's knowledge acquisition record, stored files, and content related to interactive operations; wherein the functions of the interactive operations include at least one of the following: viewing, sharing, favorites, comments, and downloading.

[0012] In one possible implementation of the first aspect above, the first historical record is information recorded by at least one terminal device while the user is logged into at least one terminal device.

[0013] In one possible implementation of the first aspect described above, at least one terminal device includes a first terminal device.

[0014] In this embodiment of the application, the terminal device may be a first terminal device, or other devices associated with the first terminal device (e.g., in the same collaborative system (e.g., connected to the same wireless local area network)), or a device associated with a user account logged in to the first terminal device (e.g., the same user account, user accounts bound to the same authentication information, etc.).

[0015] In one possible implementation of the first aspect above, the form of the query input includes at least one of the following: text, image, audio, and video.

[0016] In one possible implementation of the first aspect above, the knowledge information is generated by a generative artificial intelligence model based on question information and a second user profile, wherein the second user profile is determined based on a second history of the user's operation of at least one terminal device and / or the user's user information, and the second history is different from the first history.

[0017] In one possible implementation of the first aspect above, the first historical record is used to record the history of a user's use of at least one terminal device within a first time interval, and the second historical record is used to record the history of a user's use of at least one terminal device within a second time interval, wherein the second time interval includes the first time interval.

[0018] It's understandable that if the second time interval includes the first time interval—that is, the second time interval is longer than the first—it indicates that the cloud-based user profile is more stable than the client-side user profile. Therefore, by first generating knowledge information highly relevant to the user's interests based on the question information and the cloud-based user profile, and then generating answers based on that knowledge information and the client-side user profile, the relevance to the user's interests is even stronger, thus further improving the user experience.

[0019] In one possible implementation of the first aspect above, the content of the second historical record includes at least one of the following: at least one application's knowledge acquisition record, stored files, and content related to interactive operations; wherein the functions of the interactive operations include at least one of the following: viewing, sharing, favorites, comments, and downloading.

[0020] In one possible implementation of the first aspect above, the second historical record is a historical record obtained by the cloud-side device from at least one terminal device, wherein the historical record is information recorded by at least one terminal device while the user is logged into at least one terminal device.

[0021] In one possible implementation of the first aspect described above, at least one terminal device includes a first terminal device.

[0022] A second aspect of this application provides a terminal device, comprising: a memory for storing computer program instructions; and a processor for executing the computer program instructions to support the terminal device in implementing any of the methods described in the first aspect above.

[0023] A third aspect of this application provides a computer-readable storage medium storing computer program instructions that, when executed by a processing circuit, implement any of the methods described in the first aspect above.

[0024] The fourth aspect of this application provides a computer program product including computer program instructions that, when executed on a device, cause the device to implement any of the methods described in the first aspect above.

[0025] The fifth aspect of this application provides a knowledge acquisition system, including a first terminal device and a cloud-side device; the first terminal device is used to implement any one of the methods in the first aspect above; the cloud-side device is used to acquire knowledge information based on the query information and send the knowledge information to the first terminal device.

[0026] In one possible implementation of the fifth aspect above, the cloud-side device is also used to: generate knowledge information through a generative artificial intelligence model based on the question information and the second user profile, wherein the second user profile is determined based on a second historical record of the user's operation of at least one terminal device and / or the user's user information, and the second historical record is different from the first historical record.

[0027] In this embodiment of the application, the second user profile can be a cloud-based user profile.

[0028] In one possible implementation of the fifth aspect above, the first historical record is used to record the history of a user's use of at least one terminal device within a first time interval, and the second historical record is used to record the history of a user's use of at least one terminal device within a second time interval, wherein the second time interval includes the first time interval.

[0029] In one possible implementation of the fifth aspect above, the content of the second historical record includes at least one of the following: at least one application's knowledge acquisition record, stored files, and content related to interactive operations; wherein the functions of the interactive operations include at least one of the following: viewing, sharing, collecting, commenting, and downloading.

[0030] In one possible implementation of the fifth aspect above, the second historical record is a historical record obtained by the cloud-side device from at least one terminal device, wherein the historical record is information recorded by at least one terminal device while the user is logged into at least one terminal device.

[0031] In one possible implementation of the fifth aspect above, at least one terminal device includes a first terminal device. Attached Figure Description

[0032] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0033] Figures 1A to 1B An embodiment of this application illustrates a variation of the user interface of an encyclopedia application;

[0034] Figure 2A An embodiment of this application illustrates a user interface diagram displaying encyclopedic knowledge obtained by user A;

[0035] Figure 2B An embodiment of this application illustrates a user interface diagram displaying encyclopedic knowledge obtained by user B;

[0036] Figures 3A to 3E An illustration of another user interface variation is shown according to an embodiment of this application;

[0037] Figure 4 An interactive flowchart of a knowledge acquisition method is shown according to an embodiment of this application;

[0038] Figure 5 An interactive flowchart of a knowledge acquisition method for obtaining answer content based on question information and client-side user profile is shown according to an embodiment of this application.

[0039] Figure 6 A flowchart of a knowledge acquisition method is shown according to an embodiment of this application;

[0040] Figure 7 An interactive flowchart of a knowledge acquisition method for obtaining answer content based on question information, client-side user profile, and cloud-side user profile is shown according to an embodiment of this application.

[0041] Figure 8 An interactive flowchart of a knowledge acquisition method based on question information and cloud-side user profile is shown in the embodiments of this application.

[0042] Figure 9 A flowchart of another knowledge acquisition method is shown according to an embodiment of this application;

[0043] Figure 10 A schematic diagram of the structure of a knowledge acquisition system 1000 is shown according to an embodiment of this application;

[0044] Figure 11 An interactive schematic diagram of various modules of a knowledge acquisition system 1000 is shown according to an embodiment of this application;

[0045] Figure 12 A schematic diagram illustrating a process for generating response content is shown according to an embodiment of this application;

[0046] Figure 13A schematic diagram of the structure of an electronic device 100 is shown according to an embodiment of this application;

[0047] Figure 14 A schematic diagram of the hardware structure of a server 200 is shown according to an embodiment of this application. Detailed Implementation

[0048] The illustrative embodiments of this application include, but are not limited to, a knowledge acquisition method, apparatus, medium, program product, and system.

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described clearly and in detail below with reference to the accompanying drawings.

[0050] For ease of explanation, the following uses an encyclopedia application with knowledge acquisition capabilities as an example to provide a detailed introduction to some of the technical solutions in this application.

[0051] As mentioned earlier, users can access encyclopedic knowledge through encyclopedia applications by asking questions via text, voice, or other means. Specifically, users can enter questions related to the encyclopedic knowledge they wish to obtain into the application's search box. The encyclopedia application matches the user's question input and / or its corresponding semantic features with features corresponding to encyclopedia entries and / or content in its database, returning relevant encyclopedia entries and / or content as answers to the user's question for their viewing.

[0052] For example, Figures 1A to 1B An embodiment of this application illustrates a variation of the user interface of an encyclopedia application.

[0053] For example, Figure 1A The user interface 10a of the encyclopedia application shown includes a search box 101 and a search control 1011, as well as a "Hot Searches" display area 102 and a "Today's Recommendations" display area 103. The "Hot Searches" display area 102 includes encyclopedia entries such as "Mid-Autumn Festival" and "Mooncake". The "Today's Recommendations" display area 103 includes encyclopedia entries such as "Mid-Autumn Festival" and "Typhoon", along with corresponding portions of encyclopedia content and detail selection controls. For example, the display area corresponding to the encyclopedia entry "Mid-Autumn Festival" includes a detail selection control 1031.

[0054] It is understandable that users can obtain the desired encyclopedia knowledge by entering relevant statements in the search box 101 and clicking the search control 1011.

[0055] If electronic device 100 detects that the user is Figure 1AWhen a user enters "Mid-Autumn Festival" in the search box 101 and performs a knowledge retrieval operation (e.g., clicking the search control 1011), the electronic device 100 displays the following in response to the user's action: Figure 1B The user interface shown is 10b.

[0056] like Figure 1B As shown, the user interface 10b includes an entry overview display area 104 and a detailed encyclopedia content display area 105. The entry overview display area 104 includes an introductory video for the encyclopedia entry "Mid-Autumn Festival," titled "The Origin of the Mid-Autumn Festival," and questions such as "What is the Mid-Autumn Festival?", along with a text overview stating, "The Mid-Autumn Festival is a traditional Chinese festival, celebrated on the 15th day of the eighth lunar month… The history of the Mid-Autumn Festival is as follows:… Cultural works related to the Mid-Autumn Festival include…" The detailed encyclopedia content display area 105 includes detailed encyclopedia information such as "Festival Name" and "Festival History."

[0057] It is understandable that in some cases, the encyclopedia entry may contain a large amount of content, making it often impossible to determine the summary content based on all the encyclopedia entries. Therefore, the text summary displayed in user interface 10b may not include content that the user is most interested in (such as customs related to the Mid-Autumn Festival). Thus, the user can retrieve encyclopedia knowledge that includes content of interest by entering a new phrase such as "Mid-Autumn Festival customs".

[0058] In other cases, the detailed encyclopedia entries may be extensive, making it impossible to display all the content simultaneously on the user interface 10b. In such cases, users can view all the encyclopedia content by swiping through the pages. For example, users can swipe through pages to view encyclopedia entries on "Festival Customs" (not shown in the image) and "Festival Culture" (not shown in the image).

[0059] However, as mentioned earlier, different users may have different interests. In some cases, if the encyclopedia app doesn't present content of interest to the user, the user needs to repeatedly adjust their input to retrieve and present the information until they find something of interest. In other cases, if the encyclopedia app makes it difficult for the user to find relevant content—for example, due to the limited display area of ​​an electronic device that doesn't prioritize displaying content of interest—the user needs to perform other actions (such as continuously scrolling through the page) to obtain the relevant information. Understandably, both of these situations can negatively impact the user experience.

[0060] For example, User A is more interested in content related to festival customs, while User B is more interested in content related to festival culture. If User A... Figure 1AThe user enters "Mid-Autumn Festival" in the search box 101 of the encyclopedia application shown and clicks the search control 1011 to obtain encyclopedic knowledge related to the festival customs of the Mid-Autumn Festival. User B, through methods such as... Figure 1A The encyclopedia application shows the operation of entering "Mid-Autumn Festival" in search box 101 and clicking search control 1011 to obtain encyclopedia knowledge related to the history of the Mid-Autumn Festival. The encyclopedia application presents identical encyclopedia knowledge related to "Mid-Autumn Festival" to both user A and user B; that is, user A and user B receive the same encyclopedia knowledge, as described above. Figure 1B The user interface 10b shown depicts content that users A and B need to adjust their input to obtain encyclopedic knowledge containing information of interest, or to search for content by swiping through pages, resulting in a poor user experience.

[0061] In view of this, embodiments of this application provide a knowledge acquisition method. In this method, a first terminal device receives a question input (e.g., a question in the form of text or voice input by a user) for acquiring knowledge, and sends the question information to a cloud-side device based on the question input. Then, the first terminal device receives knowledge information generated by the cloud-side device based on the question information, and generates answer content based on the knowledge information and a terminal-side user profile (as an instance of the first user profile). The terminal-side user profile is determined based on a first historical record of a specific user (e.g., the first user) operating at at least one terminal device.

[0062] In some embodiments, at least one terminal device includes a first terminal device. That is, the terminal-side user profile is determined based on the historical operation history of the specific user on the first terminal device. Alternatively, the terminal-side user profile is determined based on the historical operation history of the specific user on the first terminal device and the historical operation history of the specific user on other terminal devices. These other terminal devices may be terminal devices associated with the first terminal device, such as those logged into with the same user account as the first terminal device. In some embodiments, the knowledge information may be determined directly by the cloud-side device based on the query information.

[0063] In other embodiments, the knowledge information may also be determined by the cloud-side device based on the query information and the cloud-side user profile (as an example of a second user profile). The cloud-side user profile may be determined based on a second historical record (e.g., knowledge acquisition record) of the specific user's operations on at least one terminal device and / or the user's user information (e.g., gender, age, occupation, etc.). It is understood that the second historical record is different from the first historical record.

[0064] In this way, the answers displayed to different users by encyclopedia apps are generated based on the question information corresponding to each user's input and the user profile of each user. For example, for the same question input from different users, the answer content presented by the encyclopedia app may be different, and the answer content is more relevant to the content that the corresponding user is interested in (e.g., prioritizing the display of content that the user is interested in). Users do not need to make any adjustments to the sentences, turn pages, or perform other operations to obtain the content they are interested in, thereby improving the user experience.

[0065] For example, for user A who is particularly interested in content related to festival customs, if user A... Figure 1A The user enters "Mid-Autumn Festival" in the search box 101 of the encyclopedia application shown (as an example of question input) and clicks the search control 1011 to obtain encyclopedia knowledge related to the festival customs of the Mid-Autumn Festival.

[0066] Suppose that the terminal device logged into user A's account obtains the knowledge information "The Mid-Autumn Festival is a traditional Chinese festival, celebrated on the 15th day of the eighth lunar month... The history of the Mid-Autumn Festival is:... Cultural works related to the Mid-Autumn Festival include...". Then, based on this knowledge information and user A's on-device user profile (as an instance of the first user profile), the terminal device can generate the answer content "The Mid-Autumn Festival is a traditional Chinese festival, celebrated on the 15th day of the eighth lunar month... Related customs of the Mid-Autumn Festival include... Cultural works related to the Mid-Autumn Festival include..." using a multimodal AIGC model. This ensures that the answer content includes information that is more interesting to user A than the standardized answer, eliminating the need for user A to re-obtain the information and thus improving the user experience.

[0067] Suppose that the terminal device of user A's logged-in account obtains the knowledge information "festival name xxx, festival history xxx, festival customs xxx, festival culture xxx", then the terminal device can filter and / or sort the obtained knowledge information based on user A's on-device user profile, thereby generating the answer content "festival name xxx, festival customs xxx, festival history xxx, festival culture xxx", so that the content that user A is interested in is displayed first, and user A does not need to perform page turning or other operations to obtain the content that interests them, thereby improving the user experience.

[0068] For example, refer to Figure 2A The image shown is a user interface diagram displaying the encyclopedia knowledge obtained by user A. (See image for details.) Figure 2AAs shown, the user interface 20a includes an entry overview display area 201 and a detailed encyclopedia content display area 202. The entry overview display area 201 includes an introductory video for the encyclopedia entry "Mid-Autumn Festival" titled "Customs of the Mid-Autumn Festival," along with textual summaries such as "The Mid-Autumn Festival is a traditional Chinese festival, celebrated on the 15th day of the eighth lunar month… Related customs of the Mid-Autumn Festival include… Cultural works related to the Mid-Autumn Festival include…" The detailed encyclopedia content display area 202 includes detailed encyclopedia information such as "Festival Name" and "Festival Customs."

[0069] For example, for user B who is particularly interested in content related to festival culture, if user B... Figure 1A The user enters "Mid-Autumn Festival" in the search box 101 of the encyclopedia application shown (as an example of question input) and clicks the search control 1011 to obtain encyclopedia knowledge related to the festival customs of the Mid-Autumn Festival.

[0070] For example, the answer to a question from user B could be something like, "The Mid-Autumn Festival is a traditional Chinese festival, celebrated on the 15th day of the eighth lunar month... Poems related to the Mid-Autumn Festival include... Paintings related to the Mid-Autumn Festival include...". This answer contains content that user B is more interested in than a standardized answer, eliminating the need for user B to retrieve it again, thus improving user B's experience.

[0071] For example, the answer to a question from user B could be something like "Festival Name xxx Festival Culture xxx Festival History xxx Festival Customs xxx". Content that user B is interested in would be displayed first, eliminating the need for user B to flip through pages or perform other operations to access the content, thus improving user B's experience.

[0072] For example, refer to Figure 2B The image shown is a user interface diagram displaying the encyclopedia knowledge obtained by user B. Figure 2B As shown, the user interface 20b includes an entry overview display area 203 and a detailed encyclopedia content display area 204. The entry overview display area 203 includes an introductory video for the encyclopedia entry "Mid-Autumn Festival," along with related poems, calligraphy and paintings, and textual summaries such as "The Mid-Autumn Festival is a traditional Chinese festival, celebrated on the 15th day of the eighth lunar month… Related poems include… Related paintings include…" The detailed encyclopedia content display area 204 includes detailed encyclopedia information such as "Festival Name" and "Festival Culture."

[0073] It is understood that the above only introduces the technical solution of this application by taking the example of a user obtaining encyclopedia knowledge through an encyclopedia application. In other embodiments, users may also obtain knowledge through other applications with knowledge acquisition functions or applications that can call the knowledge acquisition function of encyclopedia applications (such as intelligent voice assistants).

[0074] For example, refer to Figures 3A to 3E This is a schematic diagram illustrating another change in the user interface.

[0075] Suppose user C is interested in economics-related content, and user C uses a smart voice assistant to obtain encyclopedic knowledge related to event X. For example, see [reference needed]. Figure 3A The user interface 30a of the intelligent voice assistant shown includes the user-inputted phrase "Tell me about event X" (as an example of a question input).

[0076] In response to user input, electronic device 100 displays as follows Figure 3B The user interface 30b is shown. (As shown) Figure 3B As shown, the user interface 30b includes the user-inputted statement "Tell me about event X" and the corresponding encyclopedic knowledge "Event X is an event that began in xxx year. Event X has an impact on the world economy... In finance, the impact of event X on the parties involved includes... The timeline of event X is... Other impacts of event X include...", as well as feedback prompts such as "Are you satisfied with the content provided by the voice assistant?" and feedback controls "like" 301 and "dislike" 302. It can be understood that the knowledge acquisition results displayed in user interface 30b include economically related content, and economically related content has a higher priority (displayed first).

[0077] Optionally, the knowledge acquisition results may also include hyperlinks corresponding to event X, for example... Figure 3C The user interface 30c shown contains hyperlinks 303. These hyperlinks, corresponding to event X, link to the encyclopedia entry page for event X or the display interface of the encyclopedia application corresponding to event X. It can be understood that the encyclopedia entry page for event X or the display interface of the encyclopedia application corresponding to event X typically includes more encyclopedic knowledge related to event X, such as images and videos related to event X.

[0078] Optionally, the display format of the feedback control can also be as follows: Figure 3C The feedback control group 304 in the user interface 30c shown is used to obtain the user's satisfaction level.

[0079] In some embodiments, if a user wants to obtain more encyclopedic knowledge related to event X, the user can click the hyperlink corresponding to event X to jump to the encyclopedia intermediate page of event X or the display interface of the encyclopedia application corresponding to event X to obtain more encyclopedic knowledge related to event X.

[0080] For example, in response to a user clicking a hyperlink 303 in the user interface 30c, the electronic device 100 displays, as shown in the image. Figure 3D The user interface shown is 30d. (As shown in the image) Figure 3D As shown, the user interface 30d includes a textual overview of Event X: "Event X is an event that began in xxx year. Event X has an impact on the world economy… In finance, the impact of Event X on the parties involved includes… The timeline of Event X is… Other impacts of Event X include…", introductory videos about Event X such as "Talking about Event X from a financial perspective" and "The impact of Event X on the economy", and detailed encyclopedic content about Event X such as "The impact of Event X on the economy… The impact of sub-events of Event X on the economy…". It can be understood that the user interface 30d displays economically relevant content in the textual overview and detailed encyclopedic content, with higher priority given to this economically relevant content.

[0081] Or, for example, in response to a user clicking a hyperlink 303 in the user interface 30c, the electronic device 100 can also jump to the display interface corresponding to the event X of the encyclopedia application, for example, displaying... Figure 3E The user interface 30e is shown. (As shown) Figure 3E As shown, user interface 30e includes an introductory video about Event X, titled "Talking about Event X from a Financial Perspective," and "The Impact of Event X on the Economy." It also includes a textual overview of Event X, stating that "Event X is an event that began in xxx year. Event X's impact on the world economy… In finance, the impact of Event X on the parties involved includes… The timeline of Event X is… Other impacts of Event X include…," and detailed encyclopedic information about Event X, such as "The impact of Event X on the economy… The impact of Event X's sub-events on the economy…." It can be understood that user interface 30e displays economically relevant content within the textual overview and detailed encyclopedic information, with these economically relevant contents given higher priority.

[0082] Specifically, in some embodiments, the knowledge acquisition method according to this application includes the following steps:

[0083] The first terminal device receives a question input for acquiring knowledge. The form of the question input includes, but is not limited to, text, images, audio, and video.

[0084] The first terminal device sends a question to the cloud-side device based on the question input. For example, the first terminal device determines the question information (such as the first user's question or semantic information related to the first user's question) based on the question input and sends the question information to the cloud-side device. The first terminal device is an electronic device associated with a specific user, such as an electronic device logged into the first user's account.

[0085] The first terminal device receives knowledge information from the cloud-side device, which is generated based on the question information.

[0086] The generation of knowledge information by the cloud-side device based on the question information can be achieved by the cloud-side device generating knowledge information based on the question information and the cloud-side user profile (or cloud-side long-term memory). The cloud-side user profile can be determined based on the user's user information (e.g., gender, age, occupation, etc.) and / or a second historical record of the specific user's operations on at least one terminal device (e.g., knowledge acquisition records within a preset time interval, such as within one month, three months, etc.).

[0087] After receiving the knowledge information, the first terminal device generates a response based on the knowledge information and the user profile (or short-term memory). The user profile can be determined based on the first historical record of the specific user's operation of at least one terminal device (e.g., historical knowledge acquisition records within 7 days).

[0088] In some embodiments, the first history record is information recorded by at least one terminal device while the specific user is logged into at least one terminal device.

[0089] In some embodiments, at least one terminal device includes a first terminal device. That is, the first historical record includes information recorded by the first terminal device when the specific user is logged into the first terminal device.

[0090] It is understood that the first historical record can be the historical record on the first terminal device, or the historical record on other devices associated with the specific user, or the historical record on other devices associated with the first terminal device. The first historical record includes, but is not limited to: documents, images, audio, video, and other files stored on the first terminal device; historical knowledge acquisition records of at least one application on the first terminal device; and content related to the user's interactive operations on the first terminal device. The interactive operations include, but are not limited to: viewing, screenshotting, saving, forwarding, commenting, liking, downloading, and disliking. This application does not impose specific limitations on this.

[0091] In some embodiments, the second history record is different from the first history record.

[0092] For example, a first historical record is used to record the history of a user using at least one terminal device within a first time interval, and a second historical record is used to record the history of a user using at least one terminal device within a second time interval, wherein the second time interval includes the first time interval.

[0093] In some embodiments, the second historical record is a historical record obtained by the cloud-side device from at least one terminal device, wherein the historical record is information recorded by at least one terminal device while the specific user is logged in. That is, the second historical record is information obtained by the cloud-side device from at least one terminal device logged in by the specific user, and recorded by at least one terminal device while the specific user is logged in.

[0094] In some embodiments, at least one terminal device includes a first terminal device. That is, the second historical record includes information recorded by the first terminal device when the specific user is logged in, obtained by the cloud-side device from the first terminal device.

[0095] It is understood that the second historical record includes, but is not limited to: stored documents, images, audio, video, and other files; knowledge acquisition records from at least one application; and content related to interactive operations. The interactive operations include, but are not limited to: viewing, taking screenshots, saving, forwarding, commenting, liking, downloading, and disliking.

[0096] In other embodiments, the knowledge acquisition method according to this application includes the following steps:

[0097] The first terminal device receives a question input for acquiring knowledge.

[0098] The first terminal device sends a question to the cloud-side device based on the question input.

[0099] The cloud-side device retrieves knowledge information based on the query information and sends the knowledge information to the first terminal device. For example, the cloud-side device generates knowledge information based on the query information and the cloud-side user profile (or cloud-side long-term memory).

[0100] The first terminal device receives knowledge information and generates answer content based on the knowledge information and the user profile on the terminal side (or terminal short-term memory).

[0101] It is understandable that the answers obtained using the above methods are determined based on the user's on-device and / or cloud-based user profiles. Compared to standardized answers, the answers determined using these methods are more relevant to the content that users are interested in. Therefore, users do not need to retrieve or perform other operations such as flipping through pages to find content of interest, thus improving the user experience.

[0102] To better understand the technical solutions of the embodiments of this application, the relevant technical solutions of the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0103] For example, Figure 4 An interactive flowchart of a knowledge acquisition method is shown according to an embodiment of this application.

[0104] like Figure 4 As shown, this process includes, but is not limited to:

[0105] S401: The first terminal device receives a question input for acquiring knowledge.

[0106] It is understandable that the first terminal device is a device associated with the first user. For example, the first terminal device is logged into the first user's account.

[0107] In some embodiments, the first application may be a knowledge acquisition application, or it may be an application that can invoke the knowledge acquisition function of an application with knowledge acquisition capabilities. For example, the first application may be the aforementioned encyclopedia application or a smart voice assistant, and this application does not impose any limitations on this.

[0108] It is understood that the input for the question can be in text form, or other forms, such as audio, image, or video. This application does not impose any restrictions on this.

[0109] S402: The first terminal device sends a question to the cloud-side device based on the question input.

[0110] In some embodiments, the question information may be the question input itself.

[0111] In other embodiments, the question information may also be feature information of the question input. This feature information includes, but is not limited to, semantic features, image features, audio features, or feature vectors obtained based on the aforementioned features.

[0112] S403: Cloud-side devices obtain knowledge information based on the query information.

[0113] In some embodiments, the cloud-side device determines knowledge information directly based on the query information.

[0114] S404: The cloud-side device sends knowledge information to the first terminal device.

[0115] S405: The first terminal device generates the answer content based on knowledge information and the terminal user profile (as an instance of the first user profile).

[0116] In some embodiments, the first terminal device generates response content using a generative artificial intelligence model (AIGC model) based on knowledge information and a user profile. The user profile is determined by the first terminal device based on the first user's historical records (as an example of the first historical record) over a period such as one month or several months (as an example of a first time interval).

[0117] It is understood that the content of the historical records includes, but is not limited to: documents, images, audio, video, and other files stored on the first terminal device; historical knowledge acquisition records of at least one application on the first terminal device; and content corresponding to the user's interactive operations on the first terminal device. The interactive operations include, but are not limited to: viewing, taking screenshots, saving, forwarding, commenting, liking, downloading, and disliking.

[0118] It is understood that AIGC models include, but are not limited to, neural network models such as generative adversarial networks (GANs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), and deep belief networks (DBNs), as well as various variants thereof.

[0119] In other embodiments, the first terminal device filters and / or sorts multiple items in the knowledge information based on the user profile on the device side, thereby generating answer content.

[0120] For example, taking the example of a first terminal device reordering multiple items in knowledge information based on the user profile on the device side to generate answer content, if the knowledge information includes multiple content modules (such as those mentioned above) Figure 3D The content modules shown (such as "impact on the economy...", "timeline...", etc.) each correspond to a semantic meaning. The first terminal device can then reorder these content modules based on the user profile on the client side to obtain the answer. For example, the first terminal device can determine the priority of multiple content modules in the knowledge information based on the user profile on the client side, and then sort these modules based on the priority (e.g., in descending order) to obtain the answer.

[0121] It is understandable that a higher priority indicates a higher degree of matching between the corresponding content module and the user profile on the client side, that is, a greater probability that it is content that the user is interested in.

[0122] As you can understand, the above explanation uses the example of a single user account logged in on the first terminal device. In some cases, a device may not only have one user account logged in, but also associated accounts of that user account. That is, multiple accounts of a single user may be logged in on the first terminal device. For example, the first terminal device may have a first user account and associated accounts of the first user account logged in. The first user account and associated accounts of the first user account may be used for different scenarios.

[0123] The associated account can be a user account that is bound to the same authentication information as the first user account. Authentication information includes, but is not limited to: subscriber identity module (SIM) card number, email address, identity verification documents (such as ID card, passport, etc.).

[0124] For example, the first terminal device has account 1 logged in for work scenarios and account 2 logged in for daily entertainment scenarios. The first user can manually switch between the logged-in accounts to use different user accounts in different scenarios. The user can also set automatic switching rules for the logged-in accounts, such as using account 1 from 8:00 to 18:00 on weekdays, using account 2 from 19:00 to 8:00 the next day on weekdays, and using account 2 24 hours a day on holidays, etc.

[0125] In some embodiments, if the first terminal device is logged into multiple accounts of the first user, the terminal-side user profile can be determined based on the historical records corresponding to the user account selected by the first user, or the historical records corresponding to the user account currently used by the first terminal device. For example, if the user account used by the first user when using the encyclopedia application is account 1, then the first terminal device determines the corresponding terminal-side user profile based on the historical records of the first terminal device using account 1 during the time period.

[0126] It is understandable that different user accounts of the same user may have different usage scenarios, so the historical records of different accounts may be different, and the user profiles determined based on the historical records of different accounts may also be different.

[0127] For example, if the first user's job is related to economics, the knowledge acquisition records and browsing records corresponding to account 1 used for work scenarios are likely mostly related to economics. Therefore, the user profile determined based on the historical records corresponding to account 1 may indicate that the first user is interested in economic-related content.

[0128] For example, the knowledge acquisition and browsing history of account 2, which is used for daily entertainment, are mostly related to food. Therefore, the user profile determined based on the historical records of account 2 may be that the first user is interested in food-related content.

[0129] Suppose the first user asks "Tell me about city C" through a smart voice assistant, for example. If the first terminal device determines that the user selected account 1, it might present the user with answers such as "City C is... The economic development of city C... Famous attractions of city C include...". If the first terminal device determines that the user selected account 2, it might present the user with answers such as "City C is... The local cuisine of city C... Famous attractions of city C include...".

[0130] It is understandable that if the first terminal device is logged into multiple accounts belonging to the first user, a user profile can be determined based on the user account selected by the first user or the historical records corresponding to the user account currently used by the first terminal device. The answer content generated based on the question information, knowledge information, and this user profile may be closer to the knowledge that the first user expects to obtain. In other words, the answer content determined by this method may be more relevant to the content that the first user is interested in in the current usage scenario.

[0131] It is understood that the above explanation is based solely on the example of determining that the historical records of the user profile on the device side are the historical records of the first user on the first terminal device.

[0132] In some embodiments, the historical records used to determine the user profile on the device side can also be historical records collected and accumulated from other electronic devices associated with the first user. For example, other electronic devices associated with the first user can be other electronic devices different from the current device that are logged into the first user's account, such as electronic devices logged into the same user account as the first terminal device, such as mobile phones, wearable devices, tablets, etc.

[0133] Assuming the first user uses a mobile phone logged into account 1 to acquire knowledge, the phone can retrieve the historical records of account 1 locally and on a tablet logged into account 1. These historical records are then used to determine the user profile on the device side. Thus, the historical records of the first user obtained may be richer in content than the historical records of account 1 locally on the phone, potentially improving the accuracy of the user profile on the device side.

[0134] In other embodiments, the historical records used to determine the user profile on the endpoint can also be historical records from other electronic devices associated with the first terminal device. For example, other electronic devices associated with the first terminal device can be electronic devices within the same collaborative system as the first terminal device. These electronic devices in the collaborative system establish communication connections via wireless local area networks (WLANs) (such as wireless fidelity (Wi-Fi) networks) or other wireless communication methods such as Bluetooth (BT). Other electronic devices associated with the first terminal device can also be electronic devices connected to the first terminal device via a wired connection. This application does not impose specific limitations on other electronic devices associated with the first user or other electronic devices associated with the first terminal device.

[0135] For example, other electronic devices associated with the first terminal device may be other electronic devices that are different from the first terminal device but are connected to the same wireless local area network as the first terminal device, such as mobile phones, smart screens, smart speakers, etc. connected to the same Wi-Fi network.

[0136] In a smart home scenario, if a user accesses encyclopedic knowledge through a smart speaker, the smart speaker's local storage of knowledge acquisition records might be limited. Therefore, the smart speaker can access the knowledge acquisition records stored on a mobile phone connected to the same Wi-Fi network. Based on this historical record, a user profile can be determined on the device side. In this way, the acquired knowledge acquisition records for the first user are likely to be more comprehensive and richer than those stored locally on the smart speaker, improving the accuracy of the user profile.

[0137] It is understood that the historical records used to determine the user profile on the device side can be one or more of the historical records in the first terminal device, the historical records in other electronic devices associated with the first user, and the historical records in other electronic devices associated with the first terminal device. This application does not impose any specific restrictions on this.

[0138] Understandable, through Figure 4 The knowledge acquisition method shown generates answers based on question information, knowledge information, and user profiles on the client side. Compared to standardized results (knowledge not determined based on user profiles), the answers are more relevant to content of interest to the user. Thus, users do not need to re-acquire information or perform other operations such as page flipping to find content of interest, thereby improving the user experience.

[0139] In other embodiments, the aforementioned knowledge information can also be generated by the cloud-side device using a generative artificial intelligence model based on the query information and the cloud-side user profile (as an instance of the second user profile). That is, in this scenario, it is similar to the above... Figure 4 The difference in the illustrated embodiment is that the cloud-side device does not directly forward knowledge related to the question information from the database. Instead, based on the question information and the cloud-side user profile, it generates knowledge information that is highly relevant to the content of interest to the first user through a generative artificial intelligence model, and then sends the knowledge information to the first terminal device. Next, the first terminal device generates the answer content based on the knowledge information and the terminal-side user profile.

[0140] It is understandable that cloud-side user profiles can differ from edge-side user profiles.

[0141] On the one hand, assuming that the user profile on the device side is determined based on historical records within a single recording interval of the first terminal device, the user profile on the cloud side can be determined based on historical records collected and accumulated from multiple recording intervals of the first terminal device. For example, the user profile on the device side is determined based on historical records within one month corresponding to the first terminal device, while the user profile on the cloud side is determined based on historical records collected and accumulated from the first terminal device within one year. Since the user profile on the cloud side is recorded in the cloud, it is essentially generated based on the user profiles on the device side from multiple different recording intervals collected and aggregated from the first terminal device.

[0142] For example, a user profile on the device side determined based on one month of historical records may differ from a user profile on the cloud side determined based on one year of historical records. This is especially true if the user's interests have changed multiple times within one year. The more times the user's interests have changed, the more significant the difference between the user profile on the device side and the user profile on the cloud side may be.

[0143] It is understandable that the content of interest of the first user may change at different times (e.g., different record intervals), which will lead to changes in the user profile of the first user at different times. For example, the user profile determined by the historical records within the past month may be different from the user profile determined based on the previous month. Therefore, the cloud-side user profile determined by the cloud device based on the historical records within multiple record intervals of the first terminal device is a user profile determined based on the commonalities of the historical records within multiple record intervals, and the cloud-side user profile is more stable than the terminal-side user profile.

[0144] On the other hand, assuming that the user profile on the terminal side is determined based on the knowledge acquisition records of the first application in the first terminal device, the user profile on the cloud side can be determined based on the knowledge acquisition records of the first application in the first terminal device and other electronic devices associated with the first user.

[0145] It is understandable that the knowledge acquisition records of the first application within the first terminal device may contain richer content than those within the first terminal device itself. Therefore, cloud-based user profiles may be more stable than those on the device side.

[0146] In some cases, if the cloud-side device is the system server corresponding to the first terminal device, the owner of the terminal-side user profile and the owner of the cloud-side user profile are the same user.

[0147] For example, a second user associated with a cloud-side device and a first user associated with a first terminal device may correspond to the same user account.

[0148] It is understandable that the cloud-side user profile differs from the one described above. It is not collected from the first terminal device, but rather the cloud-side device independently acquires and stores the first user's historical records within the second time interval (as instances of the second historical records), and is determined based on the first user's historical records within the second time interval. The second time interval includes the first time interval.

[0149] For example, if the user account logged in on the first terminal device is account a, and the user account logged in on the cloud device is also account a, then the cloud device can use the historical records from multiple electronic devices associated with account a as the second historical records, and determine the cloud user profile based on the second historical records.

[0150] For example, the second user associated with the cloud-side device and the first user associated with the first terminal device may correspond to the same user account. For instance, the user account logged in on the first terminal device and the user account logged in on the cloud-side device may be users bound to the same authentication information.

[0151] It is understandable that the cloud-side user profile may not be collected from the first terminal device, but rather the first user's historical records within the second time interval are independently acquired and stored by the cloud-side device, and determined based on the first user's historical records within the second time interval.

[0152] For example, user account 'a' is logged in on the first terminal device, and user account 'b' is logged in on the cloud device. Accounts 'a' and 'b' are bound to the same SIM card number. Then, the cloud device can obtain the historical records of logged-in account 'b' and the electronic devices associated with logged-in account 'b' (e.g., account 'a') as a second historical record, and determine the cloud user profile based on the second historical record.

[0153] In other scenarios, if the cloud-side device is the application cloud-side device of the first application, the owner of the user profile on the client side and the owner of the user profile on the cloud side are nominally different users. This situation generally occurs when the knowledge acquisition service is a third-party service relative to the terminal system. The cloud-side user profile is determined based on the application account of the third-party knowledge acquisition service. However, the application account of the knowledge acquisition service is associated with the user of the first terminal device, and this association can be determined based on the login and usage records of that application account on the terminal.

[0154] It is understandable that there can be one or more application accounts associated with the user of the first terminal device.

[0155] Therefore, the cloud-side device can obtain the application account logged in during the third time interval in the first terminal device (i.e., the electronic device currently used by the user) that sent the question information, and use the knowledge acquisition record corresponding to the user account logged in during the third time interval in the first terminal device as the second historical record.

[0156] For example, if the user account logged in on the first terminal device 1 is account 'a', and the application account logged in within the past month in the encyclopedia application on the first terminal device 1 is 'baike1', then the cloud-side device of the encyclopedia application can obtain the knowledge acquisition records corresponding to 'baike1' as a second historical record, and determine the cloud-side user profile based on the second historical record.

[0157] For example, if the user account logged in on the first terminal device 1 is account a, and the application accounts logged in within the encyclopedia application on the first terminal device 1 within a month are baike1 and baike2, then the cloud-side device of the encyclopedia application can obtain the knowledge acquisition records corresponding to baike1 and baike2 as a second historical record, and determine the cloud-side user profile based on the second historical record.

[0158] It is understandable that, since there may be electronic devices logged into the same user account as the first terminal device, the cloud-side device can also obtain the application accounts logged in during the third time interval from the first terminal device (i.e., the electronic device currently used by the user) that sent the query information, and / or the application accounts logged in during the third time interval from electronic devices logged into the same user account as the first terminal device. The knowledge acquisition records of the obtained application accounts will be used as a second historical record.

[0159] For example, if the user account logged in on the first terminal device 1 is account a, and the application account logged in on the first terminal device 1 within a month is baike1, and the user account logged in on the first terminal device 2 is also account a, and the application account logged in on the first terminal device 2 within a month is baike2, then the cloud-side device of the encyclopedia application can obtain the knowledge acquisition records corresponding to baike1 and baike2 as the second historical records, and determine the cloud-side user profile based on the second historical records.

[0160] It is understandable that, since there may be electronic devices with user accounts associated with the first terminal device, the cloud-side device can also obtain the application accounts logged in during the third time interval from the first terminal device (i.e., the electronic device currently used by the user) that sent the query information, and / or the application accounts logged in during the third time interval from the electronic devices with user accounts associated with the first terminal device. The knowledge acquisition records of the obtained application accounts will be used as a second historical record.

[0161] For example, if the user account logged in on the first terminal device 1 is account a, and the application account logged in on the first terminal device 1 within the past month is baike1, and the user account logged in on the first terminal device 2 is also account b, and the application account logged in on the first terminal device 2 within the past month is baike2, and accounts a and b are associated user accounts, then the cloud-side device of the encyclopedia application can obtain the knowledge acquisition records corresponding to baike1 and baike2 as a second historical record, and determine the cloud-side user profile based on the second historical record.

[0162] It is understood that the durations of the first, second, and third time intervals mentioned above can be the same or different. The durations of each time interval include, but are not limited to, 24 hours, 48 ​​hours, 3 days, 7 days, 30 days, 60 days, and 90 days.

[0163] Thus, in this embodiment, the answer content is generated by the first terminal device based on the question information, knowledge information, and the user profile on the terminal side. Furthermore, the knowledge information is generated by the cloud-side device based on the question information and the user profile on the cloud side. Compared to directly obtaining knowledge from the database based on the question information as the answer content, the answer content in this embodiment is more relevant to the content of interest to the first user.

[0164] Furthermore, if the second time interval includes the first time interval (i.e., the second time interval is longer than the first time interval), it indicates that the user profile corresponding to the cloud-side user profile is more stable than the client-side user profile. Thus, compared to the method described above, which determines knowledge information solely based on the question information and generates answers based on the question information, knowledge information, and client-side user profile, first generating knowledge information with strong relevance to the first user's interests based on the question information and cloud-side user profile, and then generating answers based on the question information, knowledge information, and client-side user profile, results in even stronger relevance to the first user's interests, thereby further improving the user experience.

[0165] The following example uses electronic device 100 as the first terminal device, server 200 as the cloud-side device, intelligent voice assistant as the first application, and terminal user profile as the first user profile, combined with the above. Figures 3A to 3E The schematic diagram showing the changes in the user interface is used to provide a detailed description of the relevant technical solutions in the embodiments of this application.

[0166] Example 1

[0167] This application provides a knowledge acquisition method in which an electronic device 100 generates answer content based on knowledge information obtained by a server 200 based on question information and a user profile on the device side.

[0168] For example, Figure 5 An embodiment of this application illustrates an interactive flow diagram of a knowledge acquisition method that generates answer content based on question information, knowledge information obtained based on the question information, and a user profile on the client side.

[0169] like Figure 5 As shown, this process includes, but is not limited to:

[0170] S501: Electronic device 100 acquires and stores the first historical record of the first user within a first time interval.

[0171] It can be understood that electronic device 100 is a device associated with a first user. For example, the first user's first user account is logged into electronic device 100, or an associated account of the first user is logged into electronic device 100. The associated account can be a user account bound to the same authentication information as the first user account. Authentication information includes, but is not limited to: SIM card number, email address, identity verification documents (such as ID card, passport, etc.).

[0172] In some embodiments, the content of the historical record includes, but is not limited to: documents, pictures, audio, video and other files stored in the electronic device 100, historical knowledge acquisition records of the user in at least one application in the electronic device 100 (e.g., historical dialogues in a smart voice assistant), and content corresponding to the user's interactive operations in the electronic device 100.

[0173] The interactive functions include, but are not limited to: viewing, screenshotting, saving, forwarding, commenting, liking, downloading, and disliking. For example, an interactive function could be a user clicking on one of the above. Figure 3B The operation of the "like" control 301 and the "dislike" control 302 in the user interface 30b shown.

[0174] In some embodiments, the first historical record includes, but is not limited to, the historical record of the first user in electronic device 100, and the historical records in other electronic devices associated with the first user, and the historical records in other electronic devices associated with electronic device 100. For example, other electronic devices associated with the first user may be electronic devices logged into the first user's account, or electronic devices logged into associated accounts of the first user's account. Other electronic devices associated with electronic device 100 may be electronic devices that have established a wireless communication connection with electronic device 100 via wireless communication methods such as WLAN and Bluetooth, or electronic devices that have established a communication connection with electronic device 100 via wired means. This application does not impose specific limitations on other electronic devices associated with the first user and other electronic devices associated with electronic device 100.

[0175] In other words, the first historical record can be information recorded by at least one terminal device while the first user is logged into at least one terminal device. Specifically, the first historical record can be the historical record of the electronic device 100 itself, or it can be the historical record of other electronic devices associated with the first user. The first historical record can also consist of at least two of the aforementioned types of historical records. This application does not impose specific limitations on this.

[0176] S502: Electronic device 100 determines and stores the end-side user profile based on the first historical record.

[0177] In some embodiments, the electronic device 100 may determine a user profile based on files in a first history record. For example, keywords appearing more frequently than a frequency threshold in documents stored in the device that relate to the first history record may be used as the user profile. Or, for example, keywords with related relationships in documents stored in the device that relate to the first history record may be used as the user profile.

[0178] For example, suppose the first historical record includes the following files stored on the device: a document titled "The Impact of Event Y on the US Dollar," a document titled "Current Situation of International Commodities," and screenshots of Q&A about "What economic impact will Event X bring?" Event Y describes the type of event X, such as a social event or a natural phenomenon. If the electronic device 100 determines that the keyword "economy" appears more frequently than a frequency threshold in the stored files, then the electronic device 100 can determine that the first user is interested in economy-related content and can use "economy" as a user profile on the device side.

[0179] In other embodiments, the electronic device 100 may identify and extract features from the contents of the first historical record, and use the identification results and / or extracted features that appear more frequently than a frequency threshold as the end-side user profile.

[0180] For example, suppose the first historical record includes a document titled "The Impact of Event Y on the US Dollar" and a document titled "Current Situation of International Commodities," along with screenshots of Q&A regarding "What economic impact will Event X have?". Electronic device 100 can use a neural network model with recognition or classification capabilities to extract features from the "Impact of Event Y on the US Dollar" document, the "Current Situation of International Commodities" document, and the Q&A screenshots regarding "What economic impact will Event X have?", obtaining the feature "economically relevant." Then, electronic device 100 can determine that the first user is interested in economically relevant content and can use "economically relevant" as a user profile on the device side.

[0181] Specifically, the user profile on the device side and the historical records for determining the user profile on the device side can be found in the above. Figure 4 The relevant descriptions in S405 are not repeated here.

[0182] S503: Electronic device 100 receives a question input for acquiring knowledge.

[0183] The input for the question can be in text form or other forms, such as audio, image, or video. This application does not impose any restrictions on this.

[0184] S504: Electronic device 100 sends a question to server 200 based on the question input.

[0185] In some embodiments, the question information may be the question input from a smart voice assistant.

[0186] In other embodiments, the question information may also be feature information of the question input in the intelligent voice assistant. This feature information includes, but is not limited to, semantic features, feature vectors, image features, and audio features.

[0187] For example, the question information could be like this Figure 3A The user input "Tell me about event X" in the user interface 30a shown can also be a question based on the characteristic "event X" determined by "Tell me about event X".

[0188] S505: Server 200 generates knowledge information based on the question information.

[0189] In some embodiments, server 200 obtains knowledge information from the database that has a similarity greater than a similarity threshold to the query information, based on the query information. For example, the knowledge information obtained by server 200 based on "Event X" could be "Event X is an event that started in xxx year. Its background is... The timeline of Event X is... The impact of Event X includes... Its impact on the world economy...".

[0190] S506: Server 200 sends knowledge information to electronic device 100.

[0191] S507: Electronic device 100 generates answer content based on knowledge information and user profiles on the device side.

[0192] In some embodiments, the electronic device 100 generates answer content through an AIGC model based on knowledge information and user profiles. For example, it determines a knowledge acquisition plan (or knowledge acquisition strategy) based on question information and user profiles, and then inputs the knowledge acquisition plan and knowledge information into the AIGC model, which generates the answer content.

[0193] For example, suppose the knowledge information is "Event X is an event that started in xxx year. Its background is... The timeline of Event X is... The impact of Event X includes... its impact on the world economy..." The user profile on the device side is determined by the "economic relevance" based on the documents "The Impact of Event Y on the US Dollar" and "Recent Developments in International Commodities" and screenshots of Q&A about "What economic impact will Event X bring?", meaning that the user is interested in economically related content. The electronic device 100 can determine the device side knowledge acquisition plan as "The Impact of Event Y on the Economy" based on the question information and the user profile on the device side, and input the device side knowledge acquisition plan and knowledge information into the AIGC model, which will then generate the answer content.

[0194] The answer can be found above. Figure 3B The statement shows that "Event X is an event that began in xxx year. Event X has an impact on the world economy... In finance, the impact of Event X on the parties involved includes... The timeline of Event X is... Other impacts of Event X include...".

[0195] In other embodiments, the electronic device 100 filters and / or sorts multiple items in the knowledge information based on the user profile on the device side, thereby generating answer content.

[0196] For example, if the knowledge information includes multiple content modules, each corresponding to a semantic meaning, the electronic device 100 can reorder the multiple content modules in the knowledge information based on the user profile on the device side to obtain the answer content. For example, the electronic device 100 determines the priority of the multiple content modules in the knowledge information based on the user profile on the device side, and sorts the multiple content modules in the knowledge information based on the priority (e.g., descending order) to obtain the answer content.

[0197] It is understandable that a higher priority indicates a higher degree of matching between the corresponding content module and the user profile on the client side, that is, a greater probability that it is content that the user is interested in.

[0198] For example, electronic device 100 can determine the priority of the content modules "Definition of Event X," "Background of Event X," "Timeline of Event X," and "Impact of Event X on the Economy" in the knowledge information based on the user profile "Economic Relevant" and "Impact of Event X on the Economy." For example, the priorities from highest to lowest are: "Definition of Event X," "Impact of Event X on the Economy," "Background of Event X," and "Timeline of Event X." Then, based on the priorities corresponding to the content modules, multiple content modules in the knowledge information are sorted to obtain the answer content. For example, the answer content could be: "Event X is an event that began in xxx year. The impact of Event X includes... its impact on the world economy... Its background is... The timeline of Event X is...."

[0199] Understandable, through Figure 5 The knowledge acquisition method shown generates answers based on the question information, the knowledge information obtained from the question, and the user profile on the client side. Compared with standardized results, the answers are more relevant to the content that users are interested in. Therefore, users do not need to perform additional operations such as re-acquiring information or flipping through pages to find content of interest, thus improving the user experience.

[0200] In other embodiments, after receiving the answer, the electronic device 100 may display the answer in the user interface corresponding to the intelligent voice assistant, or broadcast the answer in the form of voice. It is understood that the electronic device 100 may also present the answer to the user in other forms. For example, a hyperlink corresponding to the answer may be displayed in the user interface corresponding to the intelligent voice assistant, and the hyperlink may link to another display interface corresponding to the answer. This application does not impose any limitations on this.

[0201] In some embodiments, after the electronic device 100 presents the answer to the user, it can also obtain the user's satisfaction level. For example, interactive controls such as click / dislike controls can be displayed in the user interface displaying the answer to obtain the user's satisfaction with the answer. Or, for example, user satisfaction can be obtained through voice dialogue, such as a smart voice assistant asking the user, "Are you satisfied with the results of this knowledge acquisition?"

[0202] For example, refer to Figure 3B The user interface 30b shown includes a feedback prompt "Are you satisfied with the content provided by the voice assistant?" and feedback controls "like" 301 and "dislike" 302.

[0203] In some embodiments, the electronic device 100 may create a user profile based on user feedback information (e.g., satisfaction level) regarding the content of the answer.

[0204] In other embodiments, the electronic device 100 may also mark unsatisfactory answers from the user and refrain from recommending knowledge with a similarity greater than a similarity threshold to the answer within a preset time interval.

[0205] It's understandable that adding feedback controls to the answer display page to obtain user satisfaction data can further enrich the user profile. At the same time, using interactive controls can increase the fun factor.

[0206] The following example uses electronic device 100 as the first terminal device, server 200 as the cloud-side device, intelligent voice assistant as the first application, and terminal user profile as the first user profile, combined with the above. Figures 3A to 3E The schematic diagram showing the changes in the user interface is used to provide a detailed description of the relevant technical solutions in the embodiments of this application.

[0207] Example 2

[0208] This application provides a knowledge acquisition method in which an electronic device 100 generates answer content based on question information, knowledge information, and a user profile on the device side.

[0209] For example, Figure 6 A flowchart of a knowledge acquisition method is shown according to an embodiment of this application. It can be understood that... Figure 6 The process shown is executed by electronic device 100. For simplicity, the following description... Figure 6 The executing entity for each step in the illustrated process will not be described again. Furthermore, the electronic device 100 has a first application installed and is associated with a first user.

[0210] like Figure 6 As shown, this process includes, but is not limited to:

[0211] S601: Receives question input for acquiring knowledge.

[0212] S602: Send a question message to server 200 based on the question input.

[0213] It is understandable that the question information can be the question input in the intelligent voice assistant, and / or the feature information of the question input.

[0214] Specifically, please refer to the above. Figure 5 The relevant descriptions in S503 are not repeated here.

[0215] S603: Receive knowledge information from server 200, where the knowledge information is generated based on the question information.

[0216] S604: Generate answer content based on knowledge information and user profiles on the client side.

[0217] In some embodiments, the electronic device 100 generates answer content based on knowledge information and user profiles on the device side using an AIGC model.

[0218] In other embodiments, the electronic device 100 filters and / or sorts multiple items in the knowledge information based on the user profile on the device side, thereby generating answer content.

[0219] Specifically, please refer to the above. Figure 5 The relevant descriptions in S507 are not repeated here.

[0220] Understandable, through Figure 6 The knowledge acquisition method shown indicates that the answer content obtained by the electronic device 100 is generated based on the question information, knowledge information, and the user profile on the device side. Compared with the standardized results, the answer content is more relevant to the content that the user is interested in. Therefore, the user does not need to perform other operations such as re-acquiring or flipping pages to obtain the content of interest, thus improving the user experience.

[0221] The following example uses electronic device 100 as the first terminal device, server 200 as the server, intelligent voice assistant as the first application, terminal-side user profile as the first user profile, and cloud-side user profile as the second user profile, combined with the above. Figures 3A to 3E The schematic diagram showing the changes in the user interface is used to provide a detailed description of the relevant technical solutions in the embodiments of this application.

[0222] Example 3

[0223] This application provides a knowledge acquisition method in which server 200 generates knowledge information based on question information and cloud-side user profile, and electronic device 100 generates answer content based on question information, knowledge information and terminal-side user profile.

[0224] For example, Figure 7 An embodiment of this application illustrates an interactive flow diagram of a knowledge acquisition method that generates knowledge information based on question information and cloud-side user profiles, and then generates answer content based on question information, knowledge information, and terminal-side user profiles.

[0225] like Figure 7 As shown, this process includes, but is not limited to:

[0226] S701: Electronic device 100 acquires and stores the first historical record of the first user within a first time interval.

[0227] Specifically, please refer to the above. Figure 5 The relevant description of S501 in this application will not be repeated here.

[0228] S702: Electronic device 100 determines and stores the end-side user profile based on the first historical record.

[0229] Specifically, please refer to the above. Figure 5 The relevant description of S502 in this application will not be repeated here.

[0230] S703: Server 200 acquires and stores the second historical record of the first user within the second time interval.

[0231] In some embodiments, the history includes, but is not limited to, the history of the first user in electronic device 100, the history of other electronic devices associated with the first user, and the history of other electronic devices associated with electronic device 100.

[0232] For example, other electronic devices associated with the first user can be electronic devices logged into the first user's account, or electronic devices logged into associated accounts of the first user. Other electronic devices associated with electronic device 100 can be electronic devices that establish a wireless communication connection with electronic device 100 via wireless communication methods such as WLAN or Bluetooth, or electronic devices that establish a communication connection with electronic device 100 via a wired connection. This application does not impose specific limitations on other electronic devices associated with the first user or other electronic devices associated with electronic device 100.

[0233] In other words, the second historical record can be the historical record of electronic device 100, or it can be the historical record of other electronic devices associated with the first user. The second historical record can also consist of at least two of the aforementioned types of historical records. It is understood that this application does not limit the source or specific type of the historical record.

[0234] In some embodiments, the second historical record may be different from the first historical record.

[0235] For example, assuming that the user profile on the device side is determined based on the historical records within a recording interval of the electronic device 100, the user profile on the cloud side can be determined based on the historical records collected and accumulated from multiple recording intervals of the electronic device 100.

[0236] For example, assuming that the user profile on the device side is determined based on the knowledge acquisition records of the first application within the electronic device 100, the user profile on the cloud side can be determined based on the knowledge acquisition records of the first application in the electronic device 100 and other electronic devices associated with the first user.

[0237] In some embodiments, the content of the historical record includes, but is not limited to: stored documents, images, audio, video, and other files; historical knowledge acquisition records of the user in at least one application (e.g., historical dialogues in a smart voice assistant); and content corresponding to the user's interactive operations. The functions of the interactive operations include, but are not limited to: viewing, screenshotting, saving, forwarding, commenting, liking, downloading, and disliking. For example, an interactive operation could be a user clicking on the aforementioned... Figure 3B The operation of the "like" control 301 and the "dislike" control 302 in the user interface 30b shown.

[0238] In some embodiments, if the server 200 is the system server corresponding to the electronic device 100, the owner of the user profile on the terminal side and the owner of the user profile on the cloud side are the same user.

[0239] For example, the second user associated with server 200 and the first user associated with electronic device 100 may correspond to the same user account.

[0240] For example, the second user associated with server 200 and the first user associated with electronic device 100 may correspond to the same user account. For instance, the user account logged into electronic device 100 and the user account logged into server 200 are users bound to the same authentication information.

[0241] It can be understood that the cloud-side user profile can be determined based on the first user's historical records within a second time interval (as an instance of the second historical record) independently acquired and stored by server 200 within the second time interval. The second time interval includes the first time interval.

[0242] In other embodiments, if server 200 is the application server for the first application, the owner of the client-side user profile and the owner of the cloud-side user profile are nominally different users. This situation generally occurs when the knowledge acquisition service is a third-party service relative to the terminal system of electronic device 100. The cloud-side user profile is determined based on the application account of the third-party knowledge acquisition service. However, the application account of the knowledge acquisition service is associated with the user of electronic device 100, and this association can be determined based on the login and usage records of the application account on electronic device 100.

[0243] It is understandable that a user of electronic device 100 may have one or more application accounts associated with them.

[0244] Therefore, server 200 can obtain the application account logged in during the third time interval in the electronic device 100 (i.e., the electronic device currently used by the user) that sent the question information, and use the knowledge acquisition record corresponding to the user account logged in during the third time interval in the electronic device 100 as the second historical record.

[0245] It is understandable that, since there may be other electronic devices logged into the same user account as electronic device 100, server 200 can also obtain the application accounts logged in during the third time interval from the electronic device 100 (i.e., the user's currently used electronic device) that sent the query information, and / or the application accounts logged in during the third time interval from other electronic devices logged into the same user account as electronic device 100. The knowledge acquisition records of the obtained application accounts will be used as a second historical record.

[0246] It is understandable that, since there may be electronic devices with user accounts associated with electronic device 100, server 200 can also obtain the application accounts logged in during the third time interval from the first terminal device that sent the query information (i.e., the electronic device currently used by the user), and / or the application accounts logged in during the third time interval from the electronic devices with user accounts associated with electronic device 100. The knowledge acquisition records of the obtained application accounts will be used as a second historical record.

[0247] Specifically, the relevant description of the second historical record can be found in the above description of the second historical record, and will not be repeated here.

[0248] S704: Server 200 determines and stores cloud-side user profiles based on the second historical record.

[0249] In some embodiments, the server 200 can identify and extract features from the content of the second historical record, and obtain the usage preferences of the first user based on the identification results and / or the extracted features, as a cloud-side user profile.

[0250] For example, suppose the second history includes stored documents titled "The Impact of Event Y on the US Dollar" and "Current Situation of International Commodities," screenshots of Q&A about "What economic impact will Event X have?", historical knowledge acquisition records from applications such as video apps, encyclopedia apps, and social media apps, and historical dialogues about topics such as commodities, currencies, and the impact of Event Y on currencies.

[0251] Server 200 can determine the cloud-side user profile based on the second historical record: "Interests: Browsing financial and economic websites and articles; Conversation history: Approximately 80% of the content is about commodities and currency-related topics, and approximately 10% of the content is about the impact of event Y on currency...; Frequently used applications: Video applications, encyclopedia applications, self-media applications...".

[0252] It is understood that in other embodiments, server 200 may also determine cloud-side user profiles based on the user information of the first user and the second historical record.

[0253] For example, suppose the second historical record includes documents titled "The Impact of Event Y on the US Dollar" and "Current Status of International Commodities," screenshots of Q&A about "What economic impact will Event X have?", historical knowledge acquisition records from video applications, encyclopedia applications, and social media applications, and historical conversations about commodities, currencies, and the impact of Event Y on currencies. The first user's information is "a 30-year-old male working in the financial industry." Server 200 can determine the cloud-side user profile based on the first user's information and the second historical record: "Gender: Male; Age: 25-35 years old; Occupation: Finance and economics related; Hobbies: Browsing finance and economics related websites and articles; Conversation history: Approximately 80% of the content is about commodities and currencies, approximately 10% of the content is about the impact of Event Y on currencies...; Frequently used applications: Video applications, encyclopedia applications, social media applications...".

[0254] S705: Electronic device 100 receives a question input for acquiring knowledge.

[0255] S706: Electronic device 100 sends a question to server 200 based on the question input.

[0256] It is understandable that the question information can be the question input in the intelligent voice assistant, and / or the feature information of the question input.

[0257] Specifically, please refer to the above. Figure 5 The relevant descriptions in S503 are not repeated here.

[0258] S707: Server 200 determines knowledge information related to the question information based on the question information and the cloud-side user profile.

[0259] In some embodiments, server 200 generates knowledge information based on query information and cloud-side user profiles using an AIGC model. For example, it determines a cloud-side knowledge acquisition plan (or cloud-side knowledge acquisition strategy) based on query information and cloud-side user profiles, and then inputs the cloud-side knowledge acquisition plan into the AIGC model, which generates knowledge information.

[0260] For example, server 200 determines the cloud-side knowledge acquisition plan based on the cloud-side user profile and question information, and then inputs the cloud-side knowledge acquisition plan into the AIGC model, which generates preliminary knowledge acquisition results.

[0261] For example, suppose the question is “Event X”; the cloud-side user profile is as follows: “Gender: Male; Age: 25-35 years old; Occupation: Finance and economics related; Hobbies: Browsing finance and economics related websites and articles; Conversation history: Approximately 80% of the content is about commodities and currency-related topics, and approximately 10% of the content is about the impact of Event Y on currency…; Commonly used applications: Video applications, encyclopedia applications, social media applications…”.

[0262] For example, a cloud-based knowledge acquisition plan could be: "1. Background introduction of event X; 2. Impact of event X on the world economy; 3. Timeline of event X; 4. Economic situation of the parties involved in event X; 5. Impact of event X," or "You are an expert in writing encyclopedia content, and your reader is a male aged 25-30, whose profession is related to finance and economics, who enjoys topics related to international financial and economic development, and frequently reads and browses text. Please generate an encyclopedia entry about event X based on the following question information: Question information: 1. Background introduction of event X; 2. Impact of event X on the world economy; 3. Timeline of event X; 4. Economic situation of the parties involved in event X; 5. Impact of event X."

[0263] The knowledge information generated by the AIGC model could be, for example, "Event X is an event that started with xxx. The background of Event X is... The timeline of Event X is... The impact of Event X on the world economy... In finance, the parties involved in Event X... Other impacts of Event X...".

[0264] In other embodiments, server 200 can determine default knowledge in the database related to the question information based on the question information. If the default knowledge includes multiple content modules, each content module corresponding to a semantic, then server 200 can filter and / or sort the multiple content modules in the default knowledge based on the cloud-side user profile to obtain knowledge information.

[0265] S708: Server 200 sends knowledge information to electronic device 100.

[0266] S709: Electronic device 100 generates response content based on knowledge information and user profiles on the device side.

[0267] In some embodiments, the electronic device 100 generates answer content based on knowledge information and user profiles on the device side through an AIGC model.

[0268] In other embodiments, the electronic device 100 filters and / or sorts multiple items in the knowledge information based on the user profile on the device side, thereby generating answer content.

[0269] For example, suppose the information is: "Event X is an event that started with xxx. The background of Event X is... The timeline of Event X is... The impact of Event X on the world economy is... In finance, the parties involved in Event X are... Other impacts of Event X are..." Then the answer could be: "Event X is an event that started with xxx. The impact of Event X on the world economy is... In finance, the parties involved in Event X are... The timeline of Event X is... Other impacts of Event X are..."

[0270] Specifically, please refer to the above. Figure 5 The relevant descriptions of S507 are not repeated here.

[0271] Understandable, through Figure 7 The knowledge acquisition method shown involves answers generated by electronic device 100 based on the question information, server 200 based on the question information and cloud-side user profiles, and terminal-side user profiles. Compared to default results obtained directly based on the question information, the answers are more relevant to the content of interest to the first user. Thus, the first user does not need to perform additional operations such as re-acquiring information or flipping through pages to obtain content of interest, thereby improving the user experience.

[0272] Furthermore, if the second time interval includes the first time interval, that is, the duration of the second time interval is longer than the duration of the first time interval, it indicates that the user profile corresponding to the cloud-side user profile is more stable than the client-side user profile. Thus, compared to the answer content obtained through the knowledge acquisition method in Embodiment 1 above, the knowledge information generated by the electronic device 100 based on the question information, the knowledge information generated by the server 200 based on the question information and the cloud-side user profile, and the answer content generated by the client-side user profile are more relevant to the content of interest to the first user, which can further improve the user experience.

[0273] In other embodiments, after receiving the answer, the electronic device 100 may display the answer in the user interface corresponding to the intelligent voice assistant, or broadcast the answer in the form of voice. It is understood that the electronic device 100 may also present the answer to the user in other forms. For example, a hyperlink corresponding to the answer may be displayed in the user interface corresponding to the intelligent voice assistant, and the hyperlink may link to another display interface corresponding to the answer, such as the user interface of an encyclopedia application. This application does not limit this.

[0274] The following example uses electronic device 100 as the first terminal device, server 200 as the server, intelligent voice assistant as the first application, and cloud-based user profile as the first user profile, combined with the above. Figures 3A to 3E The schematic diagram showing the changes in the user interface is used to provide a detailed description of the relevant technical solutions in the embodiments of this application.

[0275] Example 4

[0276] This application provides a knowledge acquisition method in which server 200 generates answer content based on question information and cloud-based user profile, and sends the answer content to electronic device 100.

[0277] For example, Figure 8 An embodiment of this application illustrates an interactive flow diagram of a knowledge acquisition method for determining answer content based on question information and cloud-based user profiles.

[0278] S801: Server 200 retrieves and stores the first historical record of the first user within the first time interval.

[0279] It is understandable that the duration of the first time interval includes, but is not limited to, 24 hours, 48 ​​hours, 3 days, 7 days, 30 days, 60 days, 90 days, etc.

[0280] In some embodiments, the content of the historical record includes, but is not limited to: documents, pictures, audio, video and other files stored in the electronic device 100, historical knowledge acquisition records of the user in at least one application in the electronic device 100 (e.g., historical dialogues in a smart voice assistant), and content corresponding to the user's interactive operations in the electronic device 100.

[0281] The interactive functions include, but are not limited to: viewing, screenshotting, saving, forwarding, commenting, liking, downloading, and disliking. For example, an interactive function could be a user clicking on one of the above. Figure 3B The operation of the "like" control 301 and the "dislike" control 302 in the user interface 30b shown.

[0282] In some embodiments, the history includes, but is not limited to, the history of the first user in electronic device 100, the history of other electronic devices associated with the first user, and the history of other electronic devices associated with electronic device 100.

[0283] In other words, the first historical record can be the historical record of electronic device 100, or it can be the historical record of other electronic devices associated with the first user. The first historical record can also consist of at least two of the aforementioned types of historical records. This application does not impose specific limitations on the source and type of the first historical record.

[0284] Specifically, the first historical record can be referenced above. Figure 7 The relevant description of the second historical record in S703 is not repeated here.

[0285] S802: Server 200 determines and stores cloud-side user profiles based on the first historical record.

[0286] In some embodiments, the server 200 can identify and extract features from the content of the first historical record, and obtain the usage preferences of the first user based on the identification results and / or the extracted features, as a cloud-side user profile.

[0287] In other embodiments, server 200 may also determine a cloud-side user profile based on the user information of the first user and the first historical record.

[0288] Specifically, please refer to the above. Figure 7 The relevant description of S704 in this application will not be repeated here.

[0289] S803: Electronic device 100 receives a question input for acquiring knowledge.

[0290] S804: Electronic device 100 sends a question to server 200 based on the question input.

[0291] It is understandable that the question information can be the question input in the intelligent voice assistant, and / or the feature information of the question input.

[0292] Specifically, please refer to the above. Figure 5 The relevant descriptions in S503 are not repeated here.

[0293] S805: Server 200 generates answer content related to the question information based on the question information and the cloud-side user profile.

[0294] In some embodiments, server 200 generates answer content based on question information and cloud-side user profiles using an AIGC model. For example, it determines a cloud-side knowledge acquisition plan (or cloud-side knowledge acquisition strategy) based on question information and cloud-side user profiles, then inputs the cloud-side knowledge acquisition plan into the AIGC model, which generates the answer content.

[0295] In other embodiments, server 200 retrieves default knowledge from the database that has a similarity greater than a similarity threshold to the question information, based on the question information. If the default knowledge includes multiple content modules, each corresponding to a semantic meaning, server 200 can filter and / or sort the multiple content modules in the default knowledge based on the cloud-side user profile to obtain the answer content.

[0296] Specifically, please refer to the above. Figure 7 The relevant description of S707 in this application will not be repeated here.

[0297] S806: Server 200 sends the response to electronic device 100.

[0298] Understandable, through Figure 8 The knowledge acquisition method shown provides answers generated by server 100 based on the question information and the user profile on the cloud side. Compared to the default results obtained directly based on the question information, the answers are more relevant to the content that the user is interested in. Therefore, users do not need to perform additional operations such as re-acquiring information or flipping through pages to obtain content of interest, thus improving the user experience.

[0299] In other embodiments, after receiving the answer, the electronic device 100 may display the answer in the user interface corresponding to the intelligent voice assistant, or broadcast the answer in the form of voice. It is understood that the electronic device 100 may also present the answer to the user in other forms. For example, a hyperlink corresponding to the answer may be displayed in the user interface corresponding to the intelligent voice assistant, and the hyperlink may link to another display interface corresponding to the answer, such as the user interface of an encyclopedia application. This application does not limit this.

[0300] The following section provides a detailed description of the relevant technical solutions in the embodiments of this application, taking the first terminal device as an electronic device 100, the cloud-side device as a server 200, and the first user profile as a cloud-side user profile as an example.

[0301] Example 5

[0302] For example, Figure 9 According to an embodiment of this application, a flowchart of a knowledge acquisition method is shown, in which a server 200 determines the content of the answer based on question information and a cloud-side user profile. It can be understood that... Figure 9 The execution entity of the process shown is server 200. For simplicity, the following description... Figure 9 The execution entity of each step in the process shown will not be described again.

[0303] like Figure 9 As shown, this process includes, but is not limited to:

[0304] S901: Receive query information from electronic device 100.

[0305] In some embodiments, the question information may be the question input from the intelligent voice assistant in the electronic device 100, and / or the feature information of the question input.

[0306] S902: Generate answer content based on question information and cloud-based user profile.

[0307] In some embodiments, server 200 generates answer content based on question information and cloud-side user profiles using an AIGC model.

[0308] For example, server 200 determines the cloud-side knowledge acquisition plan based on the cloud-side user profile and question information, and then inputs the cloud-side knowledge acquisition plan into the AIGC model, which generates the answer content.

[0309] Specifically, please refer to the above. Figure 8 The relevant descriptions in S805 are not repeated here.

[0310] S903: Send the response to electronic device 100.

[0311] Understandable, through Figure 9 The knowledge acquisition method shown provides answers generated by server 200 based on the question information and the cloud-based user profile. Compared to the default results obtained directly based on the question information, the answers are more relevant to the content that the first user is interested in.

[0312] It is understood that the knowledge acquisition method provided in this application embodiment can be applied to a knowledge acquisition system. The knowledge acquisition system includes a first terminal device and a cloud-side device.

[0313] The first terminal device is an electronic device running an application with knowledge acquisition capabilities. It can also be called a terminal, user equipment (UE), mobile station (MS), or mobile terminal (MT). Examples of first terminal devices include mobile phones, smart TVs, wearable devices, tablets, computers with wireless transceiver capabilities, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, self-driving, remote medical surgery, smart grids, transportation safety, smart cities, and smart homes.

[0314] Cloud-side devices can be physical servers or cloud servers, such as Complex Instruction Set Computing (CISC) architecture servers, Reduced Instruction Set Computing (RISC) architecture servers, etc. They can also be virtual machines (VMs) implemented based on general-purpose physical servers combined with Network Functions Virtualization (NFV) technology. A VM refers to a complete computer system simulated by software, possessing full hardware system functionality and running in a completely isolated environment. It can also be a server cluster composed of multiple servers. Servers include, but are not limited to, the system server corresponding to the first terminal device and the application server corresponding to an application with knowledge acquisition capabilities.

[0315] This application does not impose any restrictions on the specific types of the first terminal device and the cloud-side device.

[0316] The knowledge acquisition method provided in this application can also be applied to a client, which may include an electronic device or an application with knowledge acquisition functionality. The electronic device includes, but is not limited to, mobile phones, smart screens, tablets, laptops, netbooks, personal digital assistants (PDAs), and other electronic devices. Exemplary embodiments of the electronic device include, but are not limited to, devices equipped with… Or other electronic devices with different operating systems. Applications include, but are not limited to, encyclopedia applications, browsers, and other applications with knowledge acquisition functions. This application does not impose any restrictions on the specific types of electronic devices and applications.

[0317] The knowledge acquisition method provided in this application can also be applied to servers. Servers include, but are not limited to, system servers corresponding to the first terminal device and application servers corresponding to applications with knowledge acquisition capabilities.

[0318] The structure of the knowledge acquisition system and device in the embodiments of this application is described below with reference to the accompanying drawings.

[0319] Figure 10 A schematic diagram of the structure of a knowledge acquisition system 1000 is shown according to an embodiment of this application. It is understood that the knowledge acquisition system 1000 includes an electronic device 100 and a server 200.

[0320] like Figure 10As shown, the knowledge acquisition system 1000 includes: a multimodal data processing module 1001, a user memory module 1002, a collaborative generation module 1003, a user profiling module 1004, a knowledge acquisition planning module 1005, and a generation module 1006. The multimodal data processing module 1001, user memory module 1002, and collaborative generation module 1003 can be modules of the electronic device 100, while the user profiling module 1004, knowledge acquisition planning module 1005, and generation module 1006 can be modules of the server 200.

[0321] A multimodal data processing module 1001 is used to receive question input for knowledge acquisition and determine question information based on the question input. In some embodiments, the question information may be question input from an intelligent voice assistant and / or feature information of the question input. The form of question input includes, but is not limited to, text, audio, image, and video. For example, the question information may be as follows: Figure 3A The "Tell me about event X" question in the user interface 30a shown can also be based on the characteristic "Event X" determined by "Tell me about event X".

[0322] The multimodal data processing module 1001 is also used to send the question information to the user memory module 1002 and / or the user profile module 1004, and the user memory module 1002 stores the question information.

[0323] User memory module 1002 is used to acquire and store the first user's first historical record within a first time interval. It can be understood that the first historical record includes historical question information.

[0324] In some embodiments, the content of the first historical record includes, but is not limited to: stored files, knowledge acquisition records of at least one application (e.g., historical conversations in a smart voice assistant), content related to interactive operations, etc.

[0325] The interactive functions include, but are not limited to: viewing, screenshotting, saving, forwarding, commenting, liking, downloading, and disliking. For example, an interactive function could be a user clicking on one of the above. Figure 3B The operation of the "like" control 301 and the "dislike" control 302 in the user interface 30b shown.

[0326] In some embodiments, the first history record is information recorded by at least one terminal device while the first user is logged into at least one terminal device. The at least one terminal device may include electronic device 100. That is, the first history record includes, but is not limited to, the history record in electronic device 100, the history record in other electronic devices associated with the first user, and the history record in other electronic devices associated with electronic device 100, etc.

[0327] The user memory module 1002 is also used to determine and store the user profile on the client side based on the first historical record.

[0328] In some embodiments, the electronic device 100 may use historical records with a frequency greater than a frequency threshold in the first historical record as a user profile on the device side. In other embodiments, the electronic device 100 may perform identification and feature extraction on the content of historical records with a frequency greater than a frequency threshold in the first historical record, and use the identification results and / or extracted features as a user profile on the device side.

[0329] The collaborative generation module 1003 is used to generate answer content based on knowledge information and user profiles on the client side. The knowledge information can be obtained from the generation module 1006.

[0330] In some embodiments, the electronic device 100 generates answer content through an AIGC model based on knowledge information and user profiles. For example, it determines a knowledge acquisition plan (or knowledge acquisition strategy) based on question information and user profiles, and then inputs the knowledge acquisition plan and knowledge information into the AIGC model, which generates the answer content.

[0331] In other embodiments, the electronic device 100 filters and / or sorts multiple content items in the knowledge information based on the user profile on the device side, thereby generating answer content. For example, if the knowledge information includes multiple content modules, each corresponding to a semantic meaning, the electronic device 100 can reorder the multiple content modules in the knowledge information based on the user profile on the device side to obtain answer content. For example, the electronic device 100 determines the priority of multiple content modules in the knowledge information based on the user profile on the device side, and sorts the multiple content modules in the knowledge information based on the priority (e.g., descending order) to obtain answer content.

[0332] User profile module 1004 is used to acquire and store the second historical records of the first user within the second time interval.

[0333] It is understood that the content of the second historical record includes, but is not limited to: stored files, knowledge acquisition records of at least one application (such as historical dialogues in a smart voice assistant), and content related to interactive operations. The functions of these interactive operations include, but are not limited to: viewing, screenshotting, saving, forwarding, commenting, liking, downloading, and disliking. For example, an interactive operation could be a user clicking on one of the above... Figure 3B The operation of the "like" control 301 and the "dislike" control 302 in the user interface 30b shown.

[0334] In some embodiments, the second historical record is a historical record obtained by the cloud-side device from at least one terminal device, wherein the historical record is information recorded by at least one terminal device while the first user is logged in. The at least one terminal device may include electronic device 100. That is, the second historical record may be the historical record of electronic device 100, or it may be the historical record of other electronic devices associated with the first user (e.g., terminal devices logged in with the same user account as electronic device 100). The second historical record may also consist of at least two of the aforementioned types of historical records.

[0335] The user profile module 1004 is also used to determine and store cloud-side user profiles based on the second historical record.

[0336] In some embodiments, the server 200 can identify and extract features from the content of the second historical record, and obtain the usage preferences of the first user based on the identification results and / or the extracted features, as a cloud-side user profile.

[0337] In other embodiments, server 200 may also determine cloud-side user profiles based on the user information of the first user and the second historical record.

[0338] The knowledge acquisition planning module 1005 is used to generate cloud-side knowledge acquisition plans based on question information and cloud-side user profiles.

[0339] The generation module 1006 is used to determine knowledge information based on cloud-side knowledge acquisition planning and send the knowledge information to the collaborative generation module 1003.

[0340] It is understood that the structures illustrated in the embodiments of this application do not constitute a specific limitation on the knowledge acquisition system 1000. In other embodiments of this application, the knowledge acquisition system 1000 may include more or fewer modules than illustrated, or combine some modules, split some modules, or have different module arrangements. The illustrated modules may be implemented in hardware, software, or a combination of software and hardware.

[0341] It is understood that the knowledge acquisition system 1000 can be used to implement the knowledge acquisition methods provided in Embodiments 1, 3 and 4 above.

[0342] The following is combined Figure 11 and Figure 12 The specific process of the knowledge acquisition system 1000 implementing the knowledge acquisition method provided in Embodiment 3 above is described below. That is, the server 200 generates knowledge information based on the question information and the cloud-side user profile, and the electronic device 100 generates answer content based on the question information, the knowledge information, and the terminal-side user profile.

[0343] For example, Figure 11 An interactive schematic diagram of various modules of a knowledge acquisition system 1000 is shown according to an embodiment of this application.

[0344] like Figure 11 As shown, the multimodal data processing module 1001 acquires the question input from the first user and determines the question information based on the question input. The multimodal data processing module 1001 sends the question information to the user memory module 1002, the collaborative generation module 1003, the user profiling module 1004, and the knowledge acquisition planning module 1005. The question information can be the question input and / or the feature information of the question input.

[0345] It is understood that the form of input for questions includes, but is not limited to, text, audio, images, and video. For example, the question information could be like... Figure 3A The user input "Tell me about event X" in the user interface 30a shown can also be a question based on the characteristic "event X" determined by "Tell me about event X".

[0346] User memory module 1002 acquires and stores the first user's first historical record within a first time interval. Furthermore, user memory module 1002 determines and stores the client-side user profile based on the first historical record, and sends the client-side user profile to collaborative generation module 1003. The content of the historical record includes, but is not limited to: stored documents, images, and other files; historical knowledge acquisition records of at least one application (cross-application historical records); knowledge acquisition records; and dialogue history.

[0347] It is understood that the first historical record includes, but is not limited to, the historical record of the first user in electronic device 100, the historical record of other electronic devices associated with the first user, and the historical record of other electronic devices associated with electronic device 100.

[0348] It is understandable that the knowledge acquisition record can be the question information obtained from the multimodal data processing module 1001.

[0349] The user profiling module 1004 acquires and stores the second historical record of the first user within the second time interval. Furthermore, the user profiling module 1004 determines and stores the cloud-side user profile based on the second historical record, and sends the cloud-side user profile to the knowledge acquisition planning module 1005. The cloud-side user profile is based on static information (e.g., the first user's user information such as gender, age, occupation, etc.) and dynamic information (e.g., the first user's behavioral information such as the content corresponding to interactive operations, knowledge acquisition records, etc.).

[0350] It is understandable that the knowledge acquisition record can be the question information obtained from the multimodal data processing module 1001.

[0351] The knowledge acquisition planning module 1005 generates a cloud-side knowledge acquisition plan based on the question information and the cloud-side user profile, and sends the cloud-side knowledge acquisition plan to the generation module 1006.

[0352] The generation module 1006 generates knowledge information based on cloud-side knowledge acquisition and sends the knowledge information to the collaborative generation module 1003.

[0353] The collaborative generation module 1003 generates answer content based on question information, knowledge information, and user profiles on the client side.

[0354] In some embodiments, the collaborative generation module 1003 generates a knowledge acquisition plan based on the question information and the user profile on the client side, and then intelligently generates knowledge information based on the knowledge acquisition plan to obtain the answer content.

[0355] It's understandable that the answer content is generated based on the question information, knowledge information generated from the question information and the cloud-side user profile, and the client-side user profile. Compared to the default result obtained directly based on the question information, the answer content is more relevant to the content that the first user is interested in. Therefore, the first user doesn't need to perform other operations such as re-retrieval or page flipping to find the content they are interested in, thus improving the first user's experience.

[0356] Understandable. Figure 11 This paper describes the interaction process between the various modules of the knowledge acquisition system 1000 in the process of implementing the knowledge acquisition method provided in Embodiment 3 above.

[0357] The following example uses the first user's question input as "Tell me about event X" and combines it with... Figure 12 The specific process of implementing the knowledge acquisition method provided in Embodiment 3 above in the knowledge acquisition system 1000 is described.

[0358] For example, Figure 12 An embodiment of this application illustrates a schematic diagram of a process for generating response content.

[0359] like Figure 12 As shown, the first user's question input received by the multimodal data processing module 1001 is "Tell me about event X". The multimodal data processing module 1001 sends the question information to the user memory module 1002, the collaborative generation module 1003, the user profile module 1004, and the knowledge acquisition planning module 1005. The question information can be the question input "Tell me about event X" and / or the feature information "Event X" of the question input.

[0360] The user memory module 1002 stores the first user's first historical records within a first time interval, including: stored documents (e.g., PDF, PPT, DOC formats), stored images (e.g., screenshots), knowledge acquisition records, dialogue history knowledge acquisition records, knowledge acquisition records from encyclopedia applications, video applications, and self-media applications. Furthermore, the user memory module 1002 determines the client-side user profile based on the first historical records and sends the client-side user profile to the collaborative generation module 1003. The user memory module 1002 also sends the stored historical records to the user profile module 1004.

[0361] The user profile stored in the user profile module 1004 is as follows: "Gender: Male; Age: 25-35; Occupation: Finance and economics related; Hobbies: Browsing finance and economics related websites and articles; Conversation history: Approximately 80% of the content is about commodities and currency-related topics, and approximately 10% of the content is about the impact of event Y on currency...; Frequently used applications: Video applications, encyclopedia applications, social media applications...". The user profile module 1004 sends the cloud-side user profile to the knowledge acquisition planning module 1005.

[0362] The knowledge acquisition planning module 1005 generates a cloud-based knowledge acquisition plan based on the question information and the cloud-based user profile, and sends the cloud-based knowledge acquisition plan to the generation module 1006. The cloud-based knowledge acquisition plan can be: "1. Background introduction of event X; 2. Impact of event X on the world economy; 3. Timeline of event X; 4. Economic situation of the parties involved in event X; 5. Impact of event X."

[0363] The generation module 1006 generates knowledge information based on cloud-side knowledge acquisition planning and sends the knowledge information to the collaborative generation module 1003. For example, inputting the following question into the AIGC model—"You are an expert in writing encyclopedia content. Your reader is a male aged 25-3, whose profession is related to finance and economics, who is interested in topics related to international financial and economic development, and frequently reads and browses text—to generate an encyclopedia entry about event X based on the following questions: 1. Background introduction of event X; 2. Impact of event X on the world economy; 3. Timeline of event X; 4. Economic situation of the parties involved in event X; 5. Impact of event X"—will generate encyclopedia knowledge.

[0364] The collaborative generation module 1003 generates a knowledge acquisition plan for the client side based on the question information and the client-side user profile.

[0365] For example, the knowledge acquisition plan for the edge side could be the book document "The Impact of Event Y on the US Dollar" and the presentation document "The Current Situation of International Commodities", as well as screenshots of Q&A in the album about "What economic impact will Event X bring?"

[0366] The collaborative generation module 1003 intelligently generates or reorders the answer content based on the knowledge acquisition plan and knowledge information on the edge side.

[0367] For example, input the following information into the AIGC model to generate encyclopedic knowledge: "Knowledge information: Event X is an event that started with xxx. The background of Event X is… The timeline of Event X is… The impact of Event X on the world economy… In finance, the parties involved in Event X… Other impacts of Event X…". The process involves: 1. Analyzing the user profile on the client side to determine the first user's interest in "Event X"; 2. Generating or reordering the initial knowledge acquisition results based on the user's interests.

[0368] The answer could be output by a smart voice assistant, such as: "Event X is an event that started with xxx. The impact of Event X on the world economy... In finance, the parties involved in Event X... The timeline of Event X is... Other impacts of Event X...."

[0369] For example, Figure 13 A schematic diagram of the structure of an electronic device 100 is shown according to an embodiment of this application.

[0370] like Figure 13 As shown, the electronic device 100 may include a processor 110, a memory 120, an interface module 130, a power module 140, a wireless communication module 150, a mobile communication module 160, an audio module 170, a sensor module 180, buttons 190, a camera 191, a display screen 192, etc.

[0371] It is understood that the structures illustrated in the embodiments of this application do not constitute a specific limitation on the electronic device 100. In other embodiments of this application, the electronic device 100 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0372] Processor 110 may include one or more processing units, such as application processors (APs), modem processors, graphics processing units (GPUs), image signal processors (ISPs), controllers, video codecs, digital signal processors (DSPs), baseband processors, and / or neural network processing units (NPUs). These different processing units may be independent devices or integrated into one or more processors.

[0373] The processor 110 may also include a memory for storing instructions and data. In some embodiments, the memory in the processor 110 is a cache memory. This memory can store instructions or data that the processor 110 has just used or that are used repeatedly. If the processor 110 needs to use the instruction or data again, it can retrieve it directly from the memory. This avoids repeated accesses, reduces the waiting time of the processor 110, and thus improves the efficiency of the system.

[0374] In this embodiment, the processor 110 can control the fetching and execution of instructions from the memory via a controller, so that the electronic device 100 can perform... Figure 5 The implementation process shown above Figure 6 The implementation process shown above Figure 7 The interaction process shown or the above Figure 8 The functions performed by the electronic device 100 in the interactive process shown are used to implement the knowledge acquisition method provided in this application.

[0375] Interface module 130 may include an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a universal asynchronous receiver / transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input / output (GPIO) interface, a subscriber identity module (SIM) interface, and / or a universal serial bus (USB) interface, etc.

[0376] The power module 140 is connected to the processor 110 and provides power to the processor 110, memory 120, camera 191, display screen 192, and mobile communication module 160.

[0377] The wireless communication function of electronic device 100 can be implemented through wireless communication module 150, mobile communication module 160, antenna, modem processor and baseband processor, etc.

[0378] The wireless communication module 150 can provide solutions for wireless communication applications on the electronic device 100, including wireless local area networks (such as Wi-Fi networks), Bluetooth, global navigation satellite system (GNSS), frequency modulation (FM), NFC, infrared technology, etc. In some embodiments, the electronic device 100 can establish a communication connection with the server 200 through the wireless communication module 150 and receive knowledge information sent by the server 200.

[0379] The mobile communication module 160 can provide wireless communication solutions, including 2G / 3G / 4G / 5G, for use on electronic devices 100.

[0380] Audio module 170 is used to convert digital audio information into analog audio signal output, and also to convert analog audio input into digital audio signal. Audio module 170 can also be used for encoding and decoding audio signals.

[0381] The sensor module 180 may include pressure sensors, gyroscope sensors, barometric pressure sensors, magnetic sensors, accelerometers, distance sensors, proximity sensors, fingerprint sensors, temperature sensors, touch sensors, ambient light sensors, bone conduction sensors, etc.

[0382] A pressure sensor is used to sense pressure signals and convert them into electrical signals. In some embodiments, the pressure sensor may be located on the display screen 192. There are many types of pressure sensors, such as resistive pressure sensors, inductive pressure sensors, and capacitive pressure sensors. A capacitive pressure sensor may include at least two parallel plates with conductive material. When force is applied to the pressure sensor, the capacitance between the electrodes changes. The electronic device 100 determines the pressure intensity based on the change in capacitance. When a touch operation is applied to the display screen 192, the electronic device 100 detects the intensity of the touch operation based on the pressure sensor. The electronic device 100 may also calculate the touch location based on the detection signal from the pressure sensor. In some embodiments, touch operations applied to the same touch location but with different touch operation intensities may correspond to different operation commands. For example, in some embodiments, the electronic device 100 may detect user interaction operations based on the pressure sensor, such as clicking a "like" control.

[0383] A touch sensor, also known as a "touch device," can be located on the display screen 192. The touch sensor and the display screen 192 together form a touchscreen, also known as a "touchscreen." The touch sensor detects touch operations applied to or near it. The touch sensor can then transmit the detected touch operation to the application processor to determine the type of touch event. Visual output related to the touch operation can be provided through the display screen 192. For example, in some embodiments, the electronic device 100 can detect user interactions based on pressure sensors, such as clicking a "like" control.

[0384] Electronic device 100 implements display functions through GPU, display screen 192, and application processor.

[0385] The display screen 192 is used to display images, videos, etc. The display screen 192 includes a display panel. The display panel may be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode (AMOLED), a flexible light-emitting diode (FLED), a quantum dot light-emitting diode (QLED), etc. In some embodiments, the electronic device 100 may include one or N display screens 192, where N is a positive integer greater than 1.

[0386] Electronic device 100 can perform shooting functions through ISP, camera 191, video codec, GPU, display 192 and application processor.

[0387] Camera 191 is used to capture still images or videos. In some embodiments, electronic device 100 may include one or N cameras 191, where N is a positive integer greater than 1.

[0388] The memory 120 can be used to store computer executable program code, which includes instructions. The memory 120 may include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback, image playback, etc.), etc. The data storage area may store data created during the use of the electronic device 100 (such as audio data, phonebook, etc.). Furthermore, the memory 120 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, universal flash storage (UFS), etc. The processor 110 executes various functional applications and data processing of the electronic device 100 by running instructions stored in the memory 120 and / or instructions stored in memory disposed in the processor.

[0389] Button 190 includes the power button, volume buttons, etc. Button 190 can be a mechanical button or a touch button.

[0390] For example, Figure 14 A schematic diagram of the hardware structure of a server 200 is shown according to an embodiment of this application.

[0391] like Figure 14As shown, server 200 includes one or more (only one is shown in the figure) processors 210, memory 220, communication interface 230, and bus 240. The processors 210, memory 220, and communication interface 230 are interconnected via bus 240.

[0392] The processor 210 may include one or more processing units, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU), an application-specific integrated circuit, etc., for executing relevant programs to achieve the functions required by the modules in the stylus of this embodiment.

[0393] Memory 220 may include one or more memories for storing data or one or more applications. The memory may be read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM).

[0394] Processor 210 can also be an integrated circuit chip with signal processing capabilities. The aforementioned processor 210 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory 220. Processor 210 reads information from memory 220 and, in conjunction with its hardware, completes the functions required by the units included in the stylus of the embodiments of this application.

[0395] The communication interface 230 is used to enable communication between the server 200 and other devices or communication networks. In some embodiments, the server 200 establishes a communication connection with the electronic device 100 through the communication interface 230 and sends knowledge information to the electronic device 100.

[0396] Bus 240 is used to connect processor 210, memory 220, communication interface 230 and other possible modules or circuits.

[0397] It should be understood that Figure 14 The structure of server 200 shown is only an example. In other embodiments, server 200 may include more or fewer modules, which is not limited here.

[0398] In some embodiments, a computer-readable storage medium is also provided, which stores at least one computer program instruction, at least one program segment, code set, or instruction set, which is loaded and executed by a processing circuit to implement the knowledge acquisition method provided in the above-described method embodiments.

[0399] In some embodiments, a computer program product is also provided, which includes computer program instructions that, when executed by a device, enable the device to implement the knowledge acquisition methods provided in the above-described method embodiments.

[0400] The various embodiments of the mechanisms disclosed in this application can be implemented in hardware, software, firmware, or a combination of these implementation methods. Embodiments of this application can be implemented as computer programs or program code executable on a programmable system, the programmable system including at least one processor, a storage system (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device.

[0401] Program code can be applied to input instructions to execute the functions described in this application and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of this application, the processing system includes any system having a processor such as, for example, a digital signal processor (DSP), a microcontroller, an application-specific integrated circuit (ASIC), or a microprocessor.

[0402] The program code can be implemented using a high-level procedural language or an object-oriented programming language to communicate with the processing system. Assembly language or machine language can also be used when needed. In fact, the mechanisms described in this application are not limited to any particular programming language. In either case, the language can be a compiled language or an interpreted language.

[0403] In some cases, the disclosed embodiments may be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried or stored thereon on one or more temporary or non-temporary machine-readable (e.g., computer-readable) storage media, which may be read and executed by one or more processors. For example, the instructions may be distributed via a network or through other computer-readable media. Therefore, machine-readable media may include any mechanism for storing or transmitting information in a machine-readable (e.g., computer-readable) form, including but not limited to floppy disks, optical disks, optical discs, magneto-optical disks, read-only memory (ROM), random access memory (RAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic cards or optical cards, flash memory, or tangible machine-readable storage for transmitting information (e.g., carrier waves, infrared signals, digital signals, etc.) using the Internet in the form of electrical, optical, acoustic, or other forms of propagated signals. Therefore, machine-readable media include any type of machine-readable medium suitable for storing or transmitting electronic instructions or information in a machine-readable (e.g., computer-readable) form.

[0404] In the accompanying drawings, some structural or methodological features may be shown in a specific arrangement and / or order. However, it should be understood that such a specific arrangement and / or order may not be necessary. Rather, in some embodiments, these features may be arranged in a manner and / or order different from that shown in the illustrative drawings. Furthermore, the inclusion of structural or methodological features in a particular figure does not imply that such features are required in all embodiments, and in some embodiments, these features may be omitted or may be combined with other features.

[0405] It should be noted that all units / modules mentioned in the device embodiments of this application are logical units / modules. Physically, a logical unit / module can be a physical unit / module, a part of a physical unit / module, or a combination of multiple physical units / modules. The physical implementation of these logical units / modules themselves is not the most important factor; the combination of functions implemented by these logical units / modules is the key to solving the technical problems proposed in this application. Furthermore, to highlight the innovative aspects of this application, the above-described device embodiments of this application have not introduced units / modules that are not closely related to solving the technical problems proposed in this application. This does not mean that the above-described device embodiments do not contain other units / modules.

[0406] It should be noted that, in the examples and description of this patent, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0407] Although this application has been illustrated and described with reference to certain preferred embodiments thereof, those skilled in the art should understand that various changes in form and detail may be made thereto without departing from the spirit and scope of this application.

Claims

1. A knowledge acquisition method, characterized in that, include: The first terminal device receives a question input for acquiring knowledge; The first terminal device sends a question to the cloud-side device based on the question input; The first terminal device receives knowledge information from the cloud-side device, wherein the knowledge information is generated based on the question information; The first terminal device generates answer content based on the knowledge information and the first user profile, wherein the first user profile is determined based on the first historical record of the user's operation of at least one terminal device.

2. The method according to claim 1, characterized in that, The first terminal device generates answer content based on the knowledge information and the first user profile, including: The first terminal device generates the answer content based on the knowledge information and the first user profile using a generative artificial intelligence model; or, The first terminal device filters and / or sorts multiple items in the knowledge information based on the first user profile, thereby generating the answer content.

3. The method according to claim 1 or 2, characterized in that, The contents of the first historical record include at least one of the following: At least one application's knowledge acquisition records, stored files, and content related to interactive operations; The interactive operation functions include at least one of the following: viewing, sharing, favorite, commenting, and downloading.

4. The method according to any one of claims 1 to 3, characterized in that, The first historical record is the information recorded by the at least one terminal device when the user is logged into the at least one terminal device.

5. The method according to any one of claims 1 to 4, characterized in that, The at least one terminal device includes the first terminal device.

6. The method according to any one of claims 1 to 5, characterized in that, The input format for the question includes at least one of the following: text, image, audio, and video.

7. A terminal device, characterized in that, include: Memory is used to store computer program instructions; A processor for executing the computer program instructions to support the terminal device in implementing the method as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processing circuit, implement the method as described in any one of claims 1-6.

9. A computer program product, characterized in that, The computer program product includes computer program instructions that, when executed on the device, cause the device to perform the method of any one of claims 1 to 6.

10. A knowledge acquisition system, characterized in that, This includes the first terminal device and the cloud-side device; The first terminal device is used to execute the method according to any one of claims 1-6; The cloud-side device is used to obtain knowledge information based on the question information and send the knowledge information to the first terminal device.

11. The knowledge acquisition system according to claim 10, characterized in that, The cloud-side equipment is also used for: Based on the question information and the second user profile, the knowledge information is generated by a generative artificial intelligence model, wherein the second user profile is determined based on the second historical record of the user's operation of at least one terminal device and / or the user's user information, and the second historical record is different from the first historical record.

12. The knowledge acquisition system according to claim 11, characterized in that, The first historical record is used to record the user's history of using at least one terminal device within a first time interval, and the second historical record is used to record the user's history of using at least one terminal device within a second time interval, wherein the second time interval includes the first time interval.

Images