Method, apparatus and device for intelligently acquiring information and storage medium

Abstract

The present disclosure provides a method and device for intelligently obtaining information, equipment and storage medium, relates to the field of computer, specifically relates to the technical field of intelligent assistants, large models and the like, and can be applied to scenarios such as obtaining data, opening pages, man-machine question and answer and the like. The specific implementation scheme comprises the following steps: receiving target text from a client through a target plug-in, the target text being used to indicate information queried by a user intention; constructing a prompt word corresponding to the target text according to the target text, a prompt word template corresponding to the target plug-in and a database; inputting the prompt word into a large model to determine query information corresponding to the prompt word through the large model; and sending the query result corresponding to the target text to the client through the target plug-in according to the query information. The present disclosure can enable a user to quickly and accurately obtain desired information.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, specifically to the fields of intelligent assistants and large models, and can be applied to scenarios such as data acquisition, page opening, and human-computer question answering. In particular, it relates to a method, device, equipment, and storage medium for intelligent information acquisition. Background Technology

[0002] The Mini Program Developer Platform is a platform established by the service providers that support Mini Programs for Mini Program developers, providing them with functions such as onboarding, development, and operation.

[0003] Currently, in order to obtain the information they want, users (i.e., mini-program developers) must find the corresponding functional controls on the webpage of the mini-program developer platform and perform interactive operations.

[0004] However, this approach can prevent users who are not familiar with the mini-program developer platform from quickly and accurately obtaining the information they want. Summary of the Invention

[0005] This disclosure provides a method, apparatus, device, and storage medium for intelligent information acquisition, enabling users to quickly and accurately obtain the information they want.

[0006] According to a first aspect of this disclosure, a method for intelligently acquiring information is provided, comprising:

[0007] The system receives target text from the client via a target plugin. This target text indicates the information the user intends to query. Based on the target text, the corresponding prompt word template of the target plugin, and the database, prompt words corresponding to the target text are constructed. These prompt words are then input into a large model, which determines the query information corresponding to the prompt words. Based on the query information, the system sends the query results corresponding to the target text to the client via the target plugin.

[0008] According to a second aspect of this disclosure, a method for intelligently acquiring information is provided, comprising:

[0009] Receive target text input by the user, which indicates the information the user intends to query; send the target text to the target plugin; receive the query results corresponding to the target text from the target plugin; and display the query results corresponding to the target text.

[0010] According to a third aspect of this disclosure, an intelligent information acquisition device is provided, the device comprising:

[0011] The receiving module is used to receive target text from the client via the target plugin. The target text is used to indicate the information that the user intends to query.

[0012] The processing module is used to construct prompts corresponding to the target text based on the target text, the prompt word templates corresponding to the target plugin, and the database; input the prompts into the large model, and determine the query information corresponding to the prompts through the large model.

[0013] The sending module is used to send the query results corresponding to the target text to the client through the target plugin based on the query information.

[0014] According to a fourth aspect of this disclosure, an intelligent information acquisition device is provided, the device comprising:

[0015] The receiving module is used to receive target text input by the user, which indicates the information the user intends to query.

[0016] The sending module is used to send target text to the target plugin.

[0017] The receiving module is also used to receive query results corresponding to the target text from the target plugin.

[0018] The display module is used to show the query results corresponding to the target text.

[0019] According to a fifth aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a method as described in the first or second aspect.

[0020] According to a sixth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing a computer to perform the method according to the first or second aspect.

[0021] According to a seventh aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method according to the first or second aspect.

[0022] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0023] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0024] Figure 1 A schematic diagram illustrating the composition of an intelligent information acquisition system provided in this embodiment of the disclosure;

[0025] Figure 2A flowchart illustrating a method for intelligently acquiring information in a cloud service, provided as an embodiment of this disclosure;

[0026] Figure 3 Another flowchart illustrating a method for intelligently acquiring information applied to a service cloud, as provided in this disclosure embodiment;

[0027] Figure 4 This is yet another flowchart illustrating a method for intelligently acquiring information in a cloud service, as provided in this disclosure embodiment.

[0028] Figure 5 This is yet another flowchart illustrating a method for intelligently acquiring information in a cloud service, as provided in this disclosure embodiment.

[0029] Figure 6 A flowchart illustrating a method for intelligently acquiring information applied to a user terminal, provided in an embodiment of this disclosure;

[0030] Figure 7 Another flowchart illustrating a method for intelligently acquiring information applied to a user terminal, provided in an embodiment of this disclosure;

[0031] Figure 8 This is a schematic diagram illustrating the composition of an intelligent information acquisition device provided in an embodiment of the present disclosure;

[0032] Figure 9 A schematic diagram illustrating the composition of another intelligent information acquisition device provided in an embodiment of this disclosure;

[0033] Figure 10 This is a schematic diagram illustrating the composition of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0034] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0035] It should be understood that in the embodiments of this disclosure, the character " / " generally indicates that the preceding and following objects are in an "or" relationship. The terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated.

[0036] In one or more embodiments of this specification, a large model refers to a deep learning model with a large number of model parameters, typically containing hundreds of millions, tens of billions, hundreds of billions, trillions, or even tens of trillions of model parameters. A large model can also be called a foundation model. It is pre-trained using large-scale unlabeled corpora to produce a pre-trained model with hundreds of millions of parameters. Such models can adapt to a wide range of downstream tasks and have good generalization ability. Examples include Large Language Models (LLMs) and multimodal pre-training models.

[0037] In practical applications, large models only require a small number of samples to fine-tune the pre-trained model before they can be applied to different tasks. Large models can be widely used in fields such as Natural Language Processing (NLP) and Computer Vision. Specifically, they can be applied to computer vision tasks such as Visual Question Answering (VQA), Image Captioning (IC), and Image Generation, as well as natural language processing tasks such as text-based sentiment classification, text summarization, and machine translation. The main application scenarios for large models include digital assistants, intelligent robots, search, online education, office software, e-commerce, and intelligent design.

[0038] The Mini Program Developer Platform is a platform established by the service providers that support Mini Programs for Mini Program developers, providing them with functions such as onboarding, development, and operation.

[0039] Currently, in order to obtain the information they want, users (i.e., mini-program developers) must find the corresponding functional controls on the webpage of the mini-program developer platform and perform interactive operations.

[0040] For example, if a user wants to obtain the operational data of a mini-program, such as the number of returning users in the last 7 days, they must find the corresponding functional control on the webpage and click it to get the data.

[0041] However, this approach can prevent users who are not familiar with the mini-program developer platform from quickly and accurately obtaining the information they want.

[0042] For example, when the function control for retrieving certain data is located in multiple subpages under the main page of the mini-program developer platform, users who are not familiar with the mini-program developer platform need to search for and click on the function control multiple times to obtain the corresponding data.

[0043] Against this background, this disclosure provides a method for intelligently acquiring information, enabling users to quickly and accurately obtain the information they want.

[0044] The intelligent information acquisition method provided in this disclosure can be applied to intelligent information acquisition systems (such as super assistants), wherein, Figure 1 This is a schematic diagram illustrating the composition of an intelligent information acquisition system provided in an embodiment of this disclosure. For example... Figure 1 As shown, the intelligent information acquisition system may include a service cloud 101, a user terminal 102, and a large model cloud 103 that are interconnected.

[0045] The executing entity of this method can be the service cloud or user terminal of the aforementioned data transmission system. Both the service cloud and the user terminal can be computers or servers, or other electronic devices with data processing capabilities; alternatively, it can be a processor (e.g., a central processing unit (CPU)) within the aforementioned electronic device; furthermore, it can be an application (APP) installed on the aforementioned electronic device that can implement the function of this method; or, the executing entity of this method can be a functional module or unit within the aforementioned electronic device that performs the function of this method, etc. No restrictions are placed on the executing entity of this method.

[0046] In some embodiments, the server can be a single server, or it can be a server cluster consisting of multiple servers. In some embodiments, the server cluster can also be a distributed cluster. This disclosure does not limit the specific implementation of the server.

[0047] The following description, with reference to the accompanying drawings, uses the aforementioned intelligent information acquisition system as an example to illustrate the method for intelligent information acquisition in the service cloud that can be applied to the aforementioned intelligent information acquisition system.

[0048] Figure 2 This is a flowchart illustrating a method for intelligently acquiring information in a cloud service, as provided in an embodiment of this disclosure. Figure 2 As shown, the method may include:

[0049] S201. Receive target text from the client through the target plugin. The target text is used to indicate the information the user intends to query.

[0050] For example, a plugin is a program written according to a certain application programming interface (API). In practical applications, plugins can be called application programming interfaces (APIs) or tools.

[0051] For example, the client (i.e., the user terminal) can directly receive the target text input by the user through a text input device (such as a keyboard or touch screen) set on the client, or it can receive the user's voice through an audio input device (such as a microphone) set on the client, and then recognize the user's voice through a pre-trained speech recognition model to obtain the target text corresponding to the user's voice.

[0052] For example, after receiving the target text, the client can send the target text to the target plugin.

[0053] It should be noted that the target plugin can be determined by the client in response to the user's selection. The target plugin can be set directly on the service cloud or on another server, and forwarded to the service cloud after receiving the target text.

[0054] For example, the information a user intends to query may be information stored on the service cloud or on other servers connected to the service cloud. It is understood that the information a user intends to query may also be other information not stored on the service cloud or other servers.

[0055] For example, the information that a user intends to query can be one or more of the following types: data (such as mini-program data stored in the mini-program developer platform), pages (such as pages on the mini-program developer platform), and other related knowledge (such as mini-program documentation knowledge in the mini-program developer platform). There are no restrictions on the type of information that a user intends to query.

[0056] S202. Based on the target text, the prompt word template corresponding to the target plugin, and the database, construct the prompt words corresponding to the target text.

[0057] For example, different target plugins can have different functions, such as calling functions, opening pages, etc.

[0058] For example, the prompt word templates for different target plugins can be the same or different. Similarly, the databases for different target plugins can be the same or different; there is no restriction on this.

[0059] For example, multiple different target plugins can correspond to the same database, or one or more different target plugins can correspond to one or more databases.

[0060] For example, the service cloud can determine multiple data matching the target text in the database corresponding to the target plugin, and construct the prompt words corresponding to the target text by combining them with the prompt word template corresponding to the target plugin.

[0061] For example, the prompt word corresponding to the target text can be natural language text that includes the target text.

[0062] S203. Input the prompt words into the large model, and use the large model to determine the query information corresponding to the prompt words.

[0063] For example, a large model can be a pre-trained deep learning model of natural language.

[0064] For example, a large model can identify one or more pieces of data most relevant to the target text from multiple data points in the prompt words, i.e., the query information corresponding to the prompt words, without restricting the content or form of the query information.

[0065] For example, taking the information that the user intends to query as indicated by the target text as an example, the query information can be the page identifier (such as title, name, etc.).

[0066] S204. Based on the query information, send the query results corresponding to the target text to the client through the target plugin.

[0067] For example, the service cloud can use a binary evaluation confidence level to assess the confidence of the large model in response to the prompt words. Only query information whose confidence level meets a preset threshold is used to generate query results, which are then used as the query results corresponding to the target text. Here, the confidence level indicates the credibility of the query information determined by the large model; if the confidence level does not meet the preset threshold, the large model's answer can be considered fabricated.

[0068] For example, if the query information includes information A and information B, and the confidence level of information A does not meet the preset threshold, the service cloud will only generate query results based on information B, and use the query results generated based on information B as the query results corresponding to the target text.

[0069] For example, the service cloud can obtain user-readable query results based on the query information.

[0070] For example, if the target text indicates the user's intent to query information as a page, and the query information is the page identifier of the page, the query result can be the page address of the page corresponding to the page identifier.

[0071] For example, after receiving the query results, the service cloud can send the query results to the target plugin. The target plugin can be set directly on the service cloud or on another server, and forwards the query results to the client after receiving them.

[0072] For example, after receiving the query results, the client can display the query results to the user through a display device (such as a monitor, projection device, AR / VR device, etc.) or an audio output device (such as a speaker, smart speaker, etc.) set on the client, so that the user receives the expected information.

[0073] This embodiment of the disclosure constructs a prompt word corresponding to the target text, which includes data matching the target text, based on the target text indicating the user's intent to query, the prompt word template corresponding to the target plugin, and a database. The large model can determine the query information corresponding to the prompt word, obtain the query result corresponding to the target text based on the query information, and send it to the client through the target plugin. The user's intent to query can be determined from the database corresponding to the target plugin, so that the user can quickly and accurately obtain the information they want.

[0074] In some possible implementations, the target text is used to indicate the page the user expects to open; the target plugin includes a page redirection plugin; the database corresponding to the target plugin includes vectorized information corresponding to the header information of the page, and the prompt includes the header information of at least two pages; the query information includes the page identifier of the target page; the target page is one of at least two pages; and the query result includes the page address of the target page.

[0075] For example, the vectorized data in the database corresponding to the page redirection plugin can be obtained by periodically or irregularly crawling the page, obtaining the header information of the page, deduplicating it, and then vectorizing it through a preset vectorization model.

[0076] For example, taking the target text as "Open Smart Classroom" as an example, the pages corresponding to the header information of at least two pages included in the prompt can be the "Smart Classroom" page and the "Smart Learning" page, the query information can be the page identifier of the "Smart Classroom" page, and the query result can be the page address of the "Smart Classroom" page.

[0077] For example, when the client displays the page address of the target page, it can also display part of the content of the target page or display the target page directly, so that the user does not need to open the target page according to the page address, thus improving the user experience.

[0078] This embodiment can determine the page identifier of the page that the user expects to open based on the header information of the page, thereby obtaining the page address of the page that the user expects to open, so that the user can quickly and accurately open the page they expect to open.

[0079] Figure 3 This is another schematic flowchart illustrating a method for intelligently acquiring information applied to a cloud service, as provided in an embodiment of this disclosure. Figure 3 As shown, based on the target text, the prompt word template corresponding to the target plugin, and the database, the prompt words corresponding to the target text are constructed, which may include:

[0080] S301. Perform vectorization processing on the target text to obtain the vectorized information corresponding to the target text.

[0081] For example, the target text can be vectorized using a preset vectorization model to obtain the vectorized information corresponding to the target text.

[0082] S302. Based on the vectorized information corresponding to the target text, determine the first vectorized information from the database corresponding to the page jump plugin.

[0083] Among them, the first vectorized information is the vectorized information in the database corresponding to the page jump plugin whose distance from the vectorized information corresponding to the target text is greater than a preset threshold.

[0084] For example, the distance between vectorized information can be Euclidean distance, Manhattan distance, etc., without limitation.

[0085] For example, there is no restriction on the size of the preset threshold.

[0086] For example, the vectorized information in the database corresponding to the page redirection plugin is the same as the vectorized information corresponding to the page header information.

[0087] For example, by calculating the similarity distance between the vectorized information corresponding to the header information of each page in the database corresponding to the page redirection plugin and the vectorized information corresponding to the target text, the vectorized information corresponding to the header information of the page with a similarity distance greater than a preset threshold (i.e., the first vectorized information) can be determined. The page corresponding to the first vectorized information is more likely to be the page that the user expects to open as indicated by the target text.

[0088] S303. Based on the target text, the prompt word template corresponding to the page jump plugin, and the header information of the page corresponding to the first vectorized information, construct the prompt word corresponding to the target text.

[0089] For example, the prompt template corresponding to the page redirection plugin can be "The user input is 'target text'. Please determine the page most relevant to the user input from the 'header information of the page corresponding to the first vectorized information' and output the page identifier of that page. Your response is:".

[0090] For example, if the target text is "Open Smart Classroom", and the header information of the page corresponding to the first vectorized information is header information A of the "Smart Classroom" page and header information B of the "Smart Learning" page, then the prompt would be: "The user input is 'Open Smart Classroom'. Please determine the page most relevant to the user input from 'Header information A of the "Smart Classroom" page and header information B of the "Smart Learning" page', and output the page identifier of that page. Your response is: ."

[0091] This embodiment vectorizes the target text and determines the header information of the page whose similarity distance to the vectorized information corresponding to the target text is greater than a preset threshold from the database corresponding to the page jump plugin. This allows us to obtain the header information of the page corresponding to the first vectorized information that is more likely to be the page the user expects to open, as indicated by the target text. By constructing the prompt words corresponding to the target text using the target text, the prompt word template corresponding to the page jump plugin, and the header information of the page corresponding to the first vectorized information, we can more accurately obtain the page the user expects to open, as indicated by the target text.

[0092] In some possible implementations, the target text is used to indicate the knowledge that the user expects to obtain; the target plugin includes a document knowledge plugin; the database corresponding to the target plugin includes vectorized information corresponding to the document knowledge text, and the prompt words include at least two document knowledge texts; the query information includes the target document knowledge text; the target document knowledge text includes at least one of at least two document knowledge texts; and the query results include the target document knowledge text.

[0093] For example, the vectorized information in the database corresponding to the document knowledge plugin can be obtained by semantically segmenting unstructured data related to document knowledge using a semantic segmentation model and then vectorizing it using a preset vectorization model.

[0094] For example, taking the target text as "how to increase traffic in a mini-program", the prompt words could include at least two document knowledge texts such as "to increase traffic", "need to meet basic indicators", and "basic indicators include: no layout style problems on the landing page, no image quality problems on the landing page". The query information could be "need to meet basic indicators, which include: no layout style problems on the landing page, no image quality problems on the landing page". The query result could be "need to meet basic indicators, which include: no layout style problems on the landing page, no image quality problems on the landing page".

[0095] This embodiment can determine the knowledge that the user expects to obtain based on the document knowledge text, thereby obtaining the target document knowledge text that the user expects to obtain, enabling the user to quickly and accurately obtain the knowledge content they expect.

[0096] Figure 4 This is yet another flowchart illustrating a method for intelligently acquiring information in a cloud service, provided as an embodiment of this disclosure. Figure 4 As shown, based on the target text, the prompt word template corresponding to the target plugin, and the database, the prompt words corresponding to the target text are constructed, which may include:

[0097] S401. Perform vectorization processing on the target text to obtain the vectorized information corresponding to the target text.

[0098] For example, the target text can be vectorized using a preset vectorization model to obtain the vectorized information corresponding to the target text.

[0099] S402. Based on the vectorized information corresponding to the target text, determine the second vectorized information from the database corresponding to the document knowledge plugin.

[0100] The second vectorized information is the vectorized information in the database corresponding to the document knowledge plugin whose distance from the vectorized information corresponding to the target text is greater than a preset threshold.

[0101] For example, the distance between vectorized information can be Euclidean distance, Manhattan distance, etc., without limitation.

[0102] For example, there is no restriction on the size of the preset threshold.

[0103] For example, the vectorized information in the database corresponding to the document knowledge plugin is the vectorized information corresponding to the document knowledge text.

[0104] For example, by calculating the similarity distance between the vectorized information corresponding to each document knowledge text in the database corresponding to the document knowledge plugin and the vectorized information corresponding to the target text, the vectorized information corresponding to the document knowledge text with a similarity distance greater than a preset threshold (i.e., the second vectorized information) can be determined. The document knowledge text corresponding to the second vectorized information is more likely to be the knowledge that the user expects to obtain as indicated by the target text.

[0105] S403. Based on the target text, the prompt word template corresponding to the document knowledge plugin, and the document knowledge text corresponding to the second vectorized information, construct the prompt words corresponding to the target text.

[0106] For example, the prompt template corresponding to the document knowledge plugin can be "The user input is 'target text'. Please determine the document knowledge text related to the user input from 'the document knowledge text corresponding to the second vectorized information', and reply to the user directly in a plain and professional tone based on the document knowledge text. Your reply is: ".

[0107] For example, if the target text is "How to increase traffic in a mini-program", the corresponding document knowledge text for the second vectorized information could be "To increase traffic", "It is necessary to meet basic indicators", and "Basic indicators include: the landing page has no layout style problems and the landing page has no image quality problems". Then the prompt would be: "The user input is 'Open Smart Classroom'. Please reply to the user directly with the relevant document knowledge text based on the document knowledge text, using a simple and professional tone. Your reply is: "

[0108] This embodiment vectorizes the target text and determines the vectorized information corresponding to the document knowledge text whose similarity distance to the vectorized information corresponding to the target text is greater than a preset threshold from the database corresponding to the document knowledge plugin. This allows for the acquisition of the document knowledge text corresponding to the second vectorized information, which is more likely to be the knowledge that the user expects to obtain as indicated by the target text. By constructing the prompt words corresponding to the target text using the target text, the prompt word template corresponding to the document knowledge plugin, and the document knowledge text corresponding to the second vectorized information, the prompt words corresponding to the target text can be obtained more accurately.

[0109] In some possible implementations, after receiving the desired knowledge, users can provide feedback on the accuracy of the received knowledge, or directly seek answers from human customer service if dissatisfied with the response. The human customer service representative's response to the user's target text can also undergo semantic segmentation and vectorization, and the vectorized information can be stored in the database corresponding to the document knowledge plugin, enriching the knowledge content.

[0110] In some possible implementations, the target text is used to indicate the application data that the user expects to obtain; the target plugin includes a data analysis plugin; the database corresponding to the target plugin includes application data, identifiers of at least two callable functions, functional description information corresponding to at least two callable functions, and call parameters corresponding to at least two callable functions; the query information includes the identifier of the target callable function and the parameter values ​​of the call parameters corresponding to the target callable function; the target callable function includes at least one of at least two callable functions; and the query results include the application data call results of the target callable function.

[0111] For example, the application data in the database corresponding to the data analysis plugin can be obtained by periodically or irregularly crawling application data; the identifiers of at least two callable functions can be the names of the callable functions, the functional description information corresponding to at least two callable functions, and the calling parameters corresponding to at least two callable functions are pre-configured.

[0112] For example, taking the target text as "users who have revisited the Reading Mini Program in the last 14 days", the prompt words include at least two callable functions, which can be function name A and function name B. The query information can be function name B "smartapp_statistics_query", and the parameter values ​​of the call parameters corresponding to function "smartapp_statistics_query" are "name: Reading" and "time_range: 14" respectively. The query result can be the call result of function "smartapp_statistics_query" with call parameters "name: Reading" and "time_range: 14", and the call result can be "[{'type':3,'count':2,'relativeRatio':-0.5,'display':True}]".

[0113] This embodiment can determine the identifier of the target callable function and the parameter values ​​of the call parameters required to obtain the application data that the user expects to obtain based on the functional description information of the callable function. Thus, the target callable function can be called according to the parameter values ​​of the call parameters, so that the user can quickly and accurately obtain the application data they expect to obtain.

[0114] In some possible implementations, the query results include the response text corresponding to the application data call results of the target callable function. Figure 5 This is yet another flowchart illustrating a method for intelligently acquiring information in a cloud service, provided as an embodiment of this disclosure. Figure 5 As shown, based on the query information, the target plugin sends the query results corresponding to the target text to the client, which may include:

[0115] S501. Based on the parameter values ​​of the call parameters corresponding to the target callable function and the application data, call the target callable function to obtain the application data call result of the target callable function.

[0116] For example, the service cloud can call the target callable function based on the parameter values ​​of the call parameters corresponding to the target callable function, thereby obtaining the application data call result of the target callable function from the application data.

[0117] S502. Obtain the response word based on the target text, the application data call result of the target callable function, and the response word template.

[0118] For example, a response template could be: "Background: Assume you are a mini-program data robot. Please perform data analysis for users and only output conclusions. The data meanings are as follows: count is the total data volume, relativeRatio is the month-on-month data (percentages need to be multiplied by 100), type is the data type (1 for acquiring users, 2 for retained users, 3 for repeat users, 4 for transacting users; do not output other fields); output according to the data type specified by the user; the following is the data 'Application data call result of the target callable function'; the user's instruction is 'target text'; please reply to the user directly in a plain and professional tone, and the field names need to be converted to specific meanings. Your reply is:."

[0119] For example, if the target text is "Repeat users of the Reading Mini Program in the last 14 days", and the application data call result of the target callable function is the call result in the previous example, then the reply would be: "Background: Assuming you are a mini program data robot, please perform data analysis for the user and only output the conclusion. The data meanings are as follows: count is the total data volume, relativeRatio is the month-on-month data, the percentage needs to be multiplied by 100, type is the data type, 1 is the user who acquired traffic, 2 is the user who retained traffic, 3 is the user who returned traffic, and 4 is the user who made a transaction. Do not output other fields; output according to the data type specified by the user; the following is the data '[{'type':3,'count':2,'relativeRatio':-0.5,'display':True}]'; the user's instruction is 'Repeat users of the Reading Mini Program in the last 14 days'; please reply to the user directly in a plain and professional tone, and the field names need to be converted to specific meanings. Your reply is:."

[0120] S503. Input the response words into the large model, and obtain the response text corresponding to the application data call result of the target callable function through the large model.

[0121] For example, the response text corresponding to the application data call result of the target callable function obtained through the large model can be "According to the analysis of the last 14 days, the number of repeat users of the reading mini-program is 2, which is a decrease of 50% compared with the previous period."

[0122] S504. Send the response text corresponding to the application data call result of the target callable function to the client through the data analysis plugin.

[0123] For example, after receiving the response text corresponding to the application data call result of the target callable function, the service cloud can forward the response text to the data analysis plugin, which then sends the response text to the client.

[0124] This embodiment calls the target callable function based on the parameter values ​​of the call parameters and application data, obtaining the application data call result of the target callable function. Based on the target text, the application data call result of the target callable function, and the response word template, a response word is obtained. This response word is input into a large model, which then generates the response text corresponding to the application data call result of the target callable function. The response text is then sent to the client via a data analysis plugin. This ensures that users receive clear and easy-to-read application data call results, allowing them to easily obtain the desired application data and improving the user experience.

[0125] In some possible implementations, the information that the user intent to query, indicated by the target text, can be of various types. That is, the target text may include multiple questions corresponding to different target plugins, so that the above embodiments can be combined with each other for implementation.

[0126] For example, the target text could be "How many repeat users did the reading mini-program have in 14 days? What is the address of the Smart Classroom?", then the target plugins could include a data analysis plugin and a page redirection plugin, and the prompt words could include two, the query information could include two, and the query results could include two.

[0127] The following description, with reference to the accompanying drawings, uses the aforementioned intelligent information acquisition system as an example to illustrate a method for intelligent information acquisition by a user terminal that can be applied to the aforementioned intelligent information acquisition system.

[0128] Figure 6 This is a flowchart illustrating a method for intelligently acquiring information applied to a user terminal, provided as an embodiment of this disclosure. Figure 6 As shown, the method may include:

[0129] S601, Receive the target text input by the user.

[0130] The target text is used to indicate the information the user intends to query.

[0131] S602, Send target text to the target plugin.

[0132] S603: Receive the query results corresponding to the target text from the target plugin.

[0133] S604. Display the query results corresponding to the target text.

[0134] The beneficial effects and specific implementation methods of the intelligent information acquisition method applied to user terminals in the above embodiments can be referred to the intelligent information acquisition method applied to service cloud in the foregoing embodiments, and will not be repeated here.

[0135] Figure 7 This is another schematic flowchart illustrating a method for intelligently acquiring information applied to a user terminal, provided as an embodiment of this disclosure. For example... Figure 7 As shown, before sending the target text to the target plugin, the method may further include:

[0136] S701. Construct intent recognition prompts based on the intent recognition prompt template, target text, and intent list.

[0137] The intent list includes the identifiers of at least two plugins and the functional descriptions of at least two plugins.

[0138] For example, the intent recognition prompt template could be: "We have a list of intents. Please select the appropriate intent based on the user input and output only the plugin name. If there is no appropriate intent, please output 'No appropriate intent'; the intent list is 'Plugin number.Plugin name:Plugin function description'; the user input is 'target text', and your judgment is: ."

[0139] For example, given the target text as "Repeat users of the Reading Mini Program in the last 14 days", and the intent list as "1. Page jump plugin: Find a specific application and function, and open the specific URL to jump to it. 2. Data analysis plugin: Query the mini program's personal data, such as the mini program's operational data / user data / traffic data, etc.", the intent recognition prompt would be: "We have a list of intents. Please select the appropriate intent based on the user's input and only output the plugin name. If there is no suitable intent, please output 'No suitable intent'." The intent list is "1. Page jump plugin: Find a specific application and function, and open the specific URL to jump to it. 2. Data analysis plugin: Query the mini program's personal data, such as the mini program's operational data / user data / traffic data, etc."; the user's input is "Repeat users of the Reading Mini Program in the last 14 days". Your judgment is:

[0140] S702. Input the intent recognition prompts into the large model, and use the large model to determine the target plugin.

[0141] For example, continuing with the example above, after inputting the intent recognition prompt into the large model, the target plugin determined by the large model is "data analysis plugin".

[0142] This embodiment can quickly and accurately identify target plugins by using a large model to determine the target plugins that match the user intent indicated by the target text from the plugins in the intent list based on the functional description information and target text of the plugins in the intent list.

[0143] The foregoing primarily describes the solutions provided by the embodiments of this disclosure from a methodological perspective. To achieve the aforementioned functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, this disclosure can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0144] In an exemplary embodiment, this disclosure also provides an intelligent information acquisition apparatus, which can be used to implement the intelligent information acquisition method applied to the above-described intelligent information acquisition system in a service cloud.

[0145] Figure 8 This is a schematic diagram illustrating the composition of an intelligent information acquisition device provided in an embodiment of this disclosure. Figure 8 As shown, the device may include: a receiving module 801, a processing module 802, and a sending module 803.

[0146] The receiving module 801 is used to receive target text from the client through the target plugin. The target text is used to indicate the information that the user intends to query.

[0147] The processing module 802 is used to construct prompts corresponding to the target text based on the target text, the prompt template corresponding to the target plugin, and the database; input the prompts into the large model, and determine the query information corresponding to the prompts through the large model.

[0148] The sending module 803 is used to send the query results corresponding to the target text to the client through the target plugin based on the query information.

[0149] In one possible embodiment, the target text is used to indicate the page the user expects to open; the target plugin includes a page redirection plugin; the database corresponding to the target plugin includes vectorized information corresponding to the header information of the page, and the prompt includes the header information of at least two pages; the query information includes the page identifier of the target page; the target page is one of at least two pages; and the query result includes the page address of the target page.

[0150] In one possible embodiment, the processing module 802 is specifically used for:

[0151] The target text is vectorized to obtain the corresponding vectorized information. Based on the vectorized information of the target text, the first vectorized information is determined from the database corresponding to the page jump plugin. The first vectorized information is the vectorized information in the database corresponding to the page jump plugin whose distance from the vectorized information corresponding to the target text is greater than a preset threshold. Based on the target text, the prompt word template corresponding to the page jump plugin, and the header information of the page corresponding to the first vectorized information, the prompt word corresponding to the target text is constructed.

[0152] In one possible embodiment, the target text is used to indicate the knowledge that the user expects to obtain; the target plugin includes a document knowledge plugin; the database corresponding to the target plugin includes vectorized information corresponding to the document knowledge text, and the prompt words include at least two document knowledge texts; the query information includes the target document knowledge text; the target document knowledge text includes at least one of the at least two document knowledge texts; and the query result includes the target document knowledge text.

[0153] In one possible embodiment, the processing module 802 is specifically used for:

[0154] The target text is vectorized to obtain the vectorized information corresponding to the target text. Based on the vectorized information corresponding to the target text, second vectorized information is determined from the database corresponding to the document knowledge plugin. The second vectorized information is the vectorized information in the database corresponding to the document knowledge plugin whose distance from the vectorized information corresponding to the target text is greater than a preset threshold. Based on the target text, the prompt word template corresponding to the document knowledge plugin, and the document knowledge text corresponding to the second vectorized information, prompt words corresponding to the target text are constructed.

[0155] In one possible embodiment, the target text is used to indicate the application data that the user expects to obtain; the target plugin includes a data analysis plugin; the database corresponding to the target plugin includes application data, at least two callable functions, functional description information corresponding to at least two callable functions, and call parameters corresponding to at least two callable functions; the query information includes the target callable function and the parameter values ​​of the call parameters corresponding to the target callable function; the target callable function includes at least one of at least two callable functions; the query result includes the application data call result of the target callable function.

[0156] In one possible embodiment, the query result includes the response text corresponding to the application data call result of the target callable function, and the sending module 803 is specifically used for:

[0157] Based on the parameter values ​​of the call parameters corresponding to the target callable function and the application data, the target callable function is called to obtain the application data call result of the target callable function; based on the target text, the application data call result of the target callable function, and the response word template, the response word is obtained; the response word is input into the large model, and the large model obtains the response text corresponding to the application data call result of the target callable function; the response text corresponding to the application data call result of the target callable function is sent to the client through the data analysis plugin.

[0158] In an exemplary embodiment, this disclosure also provides an intelligent information acquisition apparatus, which can be used to implement the intelligent information acquisition method of a user terminal applied to the aforementioned intelligent information acquisition system as described in the foregoing embodiments.

[0159] Figure 9 This is a schematic diagram illustrating the composition of another intelligent information acquisition device provided in an embodiment of this disclosure. (See diagram below.) Figure 9 As shown, the device may include: a receiving module 901, a sending module 903, and a display module 904.

[0160] The receiving module 901 is used to receive target text input by the user, the target text being used to indicate the information the user intends to query.

[0161] Sending module 903 is used to send target text to the target plugin;

[0162] The receiving module 901 is also used to receive query results corresponding to the target text from the target plugin;

[0163] Display module 904 is used to display the query results corresponding to the target text.

[0164] In one possible embodiment, the device further includes:

[0165] The determination module 902 is used to construct intent recognition prompts based on the intent recognition prompt template, the target text, and the intent list before sending the target text to the target plugin. The intent list includes the identifiers of at least two plugins and the functional description information of at least two plugins. The intent recognition prompts are input into the large model, and the target plugin is determined through the large model.

[0166] It should be noted that, Figure 8 and Figure 9 The module division described herein is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used. For example, two or more functions can be integrated into a single processing module. This disclosure does not impose any limitations on this. The integrated modules described above can be implemented in hardware or as software functional modules.

[0167] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0168] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0169] In an exemplary embodiment, an electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method described in the above embodiments. The electronic device may be the computer or server described above.

[0170] In an exemplary embodiment, the readable storage medium may be a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method described in the above embodiments.

[0171] In an exemplary embodiment, the computer program product includes a computer program that, when executed by a processor, implements the method described in the above embodiments.

[0172] Figure 10 A schematic block diagram of an example electronic device 1000 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0173] like Figure 10 As shown, the electronic device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. The RAM 1003 may also store various programs and data required for the operation of the electronic device 1000. The computing unit 1001, ROM 1002, and RAM 1003 are interconnected via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0174] Multiple components in electronic device 1000 are connected to I / O interface 1005, including: input unit 1006, such as keyboard, mouse, etc.; output unit 1007, such as various types of displays, speakers, etc.; storage unit 1008, such as disk, optical disk, etc.; and communication unit 1009, such as network card, modem, wireless transceiver, etc. Communication unit 1009 allows electronic device 1000 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0175] The computing unit 1001 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1001 performs the various methods and processes described above, such as the method of intelligently acquiring information. For example, in some embodiments, the method of intelligently acquiring information may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 1000 via ROM 1002 and / or communication unit 1009. When the computer program is loaded into RAM 1003 and executed by the computing unit 1001, one or more steps of the method of intelligently acquiring information described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to perform a method of intelligently acquiring information by any other suitable means (e.g., by means of firmware).

[0176] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0177] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0178] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0179] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0180] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), middleware components (e.g., an application server), or frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0181] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0182] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0183] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for intelligently acquiring information, the method comprising: receiving, by a target plug-in, target text from a client, the target text being used to indicate information that a user intends to query; wherein the target text is used to indicate a page that the user expects to open; the target plug-in comprises a page jump plug-in; constructing, according to the target text, a prompt word corresponding to the target text, a prompt word template corresponding to the target plug-in, and a database corresponding to the target plug-in, the constructing comprising: performing vectorization processing on the target text to obtain vectorization information corresponding to the target text; determining, from the database corresponding to the page jump plug-in, first vectorization information according to the vectorization information corresponding to the target text, the first vectorization information being vectorization information in the database corresponding to the page jump plug-in that has a distance greater than a preset threshold from the vectorization information corresponding to the target text; and constructing the prompt word corresponding to the target text according to the target text, the prompt word template corresponding to the page jump plug-in, and head information of a page corresponding to the first vectorization information; inputting the prompt word into a large model to determine, by the large model, query information corresponding to the prompt word; sending, according to the query information, a query result corresponding to the target text to the client by the target plug-in.

2. The method of claim 1, wherein the database corresponding to the target plug-in comprises vectorization information corresponding to head information of a page, the prompt word comprises head information of at least two pages, the query information comprises a page identifier of a target page, the target page is one of the at least two pages, and the query result comprises a page address of the target page.

3. The method of claim 1, wherein the target text is used to indicate knowledge that a user expects to acquire, the target plug-in comprises a document knowledge plug-in, the database corresponding to the target plug-in comprises vectorization information corresponding to document knowledge text, the prompt word comprises at least two document knowledge texts, the query information comprises target document knowledge text, the target document knowledge text comprises at least one of the at least two document knowledge texts, and the query result comprises the target document knowledge text.

4. The method of claim 3, wherein the constructing, according to the target text, a prompt word template corresponding to the target plug-in, and a database corresponding to the target plug-in, the prompt word corresponding to the target text, further comprises: performing vectorization processing on the target text to obtain vectorization information corresponding to the target text; determining, from the database corresponding to the document knowledge plug-in, second vectorization information according to the vectorization information corresponding to the target text, the second vectorization information being vectorization information in the database corresponding to the document knowledge plug-in that has a distance greater than a preset threshold from the vectorization information corresponding to the target text; and constructing the prompt word corresponding to the target text according to the target text, a prompt word template corresponding to the document knowledge plug-in, and document knowledge text corresponding to the second vectorization information.

5. The method of any one of claims 1-4, wherein the target text is used to indicate application data that the user expects to obtain; the target plug-in comprises a data analysis plug-in; the database corresponding to the target plug-in comprises application data, identification of at least two callable functions, function description information corresponding to the at least two callable functions, and calling parameters corresponding to the at least two callable functions; the query information comprises identification of a target callable function and parameter values of calling parameters corresponding to the target callable function; the target callable function comprises at least one of the at least two callable functions; and the query result comprises an application data calling result of the target callable function.

6. The method of claim 5, wherein the query result comprises a reply text corresponding to the application data calling result of the target callable function, and the sending, by the target plug-in, of the query result corresponding to the target text to the client according to the query information comprises: calling the target callable function according to the parameter values of the calling parameters corresponding to the target callable function and the application data, to obtain the application data calling result of the target callable function; obtaining a reply word according to the target text, the application data calling result of the target callable function, and a reply word template; inputting the reply word into a large model to obtain the reply text corresponding to the application data calling result of the target callable function through the large model; and sending, by the data analysis plug-in, the reply text corresponding to the application data calling result of the target callable function to the client.

7. A method for intelligently obtaining information, the method comprising: receiving a target text input by a user, the target text being used to indicate information that the user intends to query; sending the target text to a target plug-in; receiving a query result corresponding to the target text from the target plug-in, wherein the query result is obtained according to the method of claim 1; and displaying the query result corresponding to the target text.

8. The method of claim 7, wherein before the sending of the target text to the target plug-in, the method further comprises: constructing an intent recognition prompt word according to an intent recognition prompt word template, the target text, and an intent list, the intent list comprising identification of at least two plug-ins and function description information of the at least two plug-ins; and inputting the intent recognition prompt word into a large model to determine the target plug-in through the large model.

9. An apparatus for intelligently obtaining information, the apparatus comprising: a receiving module configured to receive a target text from a client through a target plug-in, the target text being used to indicate information that a user intends to query; wherein the target text is used to indicate a page that the user expects to open; and the target plug-in comprises a page jump plug-in. ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ The processing module is configured to construct a prompt word corresponding to the target text according to the target text, a prompt word template corresponding to the target plug-in, and a database; input the prompt word into a large model, and determine query information corresponding to the prompt word by using the large model; and the processing module is specifically configured to: perform vectorization processing on the target text to obtain vectorization information corresponding to the target text; determine first vectorization information from the database corresponding to the page jump plug-in according to the vectorization information corresponding to the target text, the first vectorization information being vectorization information in the database corresponding to the page jump plug-in and having a distance greater than a preset threshold from the vectorization information corresponding to the target text; and construct the prompt word corresponding to the target text according to the target text, the prompt word template corresponding to the page jump plug-in, and head information of a page corresponding to the first vectorization information. The sending module is configured to send, to the client, a query result corresponding to the target text by using the target plug-in according to the query information.

10. The apparatus of claim 9, wherein the database corresponding to the target plug-in includes vectorization information corresponding to head information of a page, and the prompt word includes head information of at least two pages; the query information includes a page identifier of a target page; the target page is one of the at least two pages; and the query result includes a page address of the target page.

11. The apparatus of claim 9, wherein the target text is used to indicate knowledge expected to be obtained by a user; the target plug-in includes a document knowledge plug-in; the database corresponding to the target plug-in includes vectorization information corresponding to a document knowledge text; the prompt word includes at least two document knowledge texts; the query information includes a target document knowledge text; the target document knowledge text includes at least one of the at least two document knowledge texts; and the query result includes the target document knowledge text.

12. The apparatus of claim 11, wherein the processing module is specifically configured to: perform vectorization processing on the target text to obtain vectorization information corresponding to the target text; determine second vectorization information from the database corresponding to the document knowledge plug-in according to the vectorization information corresponding to the target text, the second vectorization information being vectorization information in the database corresponding to the document knowledge plug-in and having a distance greater than a preset threshold from the vectorization information corresponding to the target text; and construct the prompt word corresponding to the target text according to the target text, a prompt word template corresponding to the document knowledge plug-in, and a document knowledge text corresponding to the second vectorization information.

13. The apparatus of any one of claims 9-12, wherein the target text is used to indicate application data expected to be obtained by a user; the target plug-in comprises a data analysis plug-in; a database corresponding to the target plug-in comprises application data, identification of at least two callable functions, function description information corresponding to the at least two callable functions, and calling parameters corresponding to the at least two callable functions; the query information comprises identification of a target callable function and parameter values of calling parameters corresponding to the target callable function; the target callable function comprises at least one of the at least two callable functions; and the query result comprises an application data calling result of the target callable function.

14. The apparatus of claim 13, wherein the query result comprises a reply text corresponding to the application data calling result of the target callable function, and the sending module is specifically configured to: call the target callable function according to the parameter values of the calling parameters corresponding to the target callable function and the application data, to obtain the application data calling result of the target callable function; obtain a reply word according to the target text, the application data calling result of the target callable function, and a reply word template; input the reply word into a large model to obtain the reply text corresponding to the application data calling result of the target callable function through the large model; and send the reply text corresponding to the application data calling result of the target callable function to the client through the data analysis plug-in.

15. An apparatus for intelligently obtaining information, comprising: a receiving module configured to receive a target text input by a user, the target text being used to indicate information intended to be queried by the user; a sending module configured to send the target text to a target plug-in; the receiving module is further configured to receive a query result corresponding to the target text from the target plug-in, wherein the query result is obtained according to the method of claim 1; and a display module configured to display the query result corresponding to the target text.

16. The apparatus of claim 15, further comprising: a determining module configured to, before the sending module sends the target text to the target plug-in, construct an intent recognition prompt word according to an intent recognition prompt word template, the target text, and an intent list, the intent list comprising identification of at least two plug-ins and function description information of the at least two plug-ins; and input the intent recognition prompt word into a large model to determine the target plug-in through the large model. at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6 or 7-8.

18. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1-6 or 7-8. ​ ​ ​ ​ ​ ​ ​ ​ 17. An electronic device comprising: ​ ​ ​ ​ 19. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-6 or claims 7-8.

Images