Data processing method and device, equipment and storage medium

Abstract

The invention provides a data processing method and device, equipment and a storage medium, which can better process an instruction of a user. Specifically, first reply information sent by a target model can be received; wherein the first reply information is determined by the target model according to the first prompt information, and the first prompt information is information sent to the target model. Then, determining first display content according to the first reply information; the first display content comprises structured content. Then, user input information associated with the first display content may be obtained. And after the information input by the user is obtained, second prompt information can be sent to the target model. The second prompt information comprises user input information and description information. The description information is natural language description information of the structured content in the first display content. Therefore, the target model can accurately determine the intention of the user, so that the reply matched with the intention of the user is given.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a data processing method, device, equipment and storage medium. Background Art

[0002] With the development of computer technology, it is possible to respond to user instructions through models. For example, users can issue instructions multiple times. The model can obtain the user's instructions and perform corresponding actions according to the user's instructions.

[0003] However, in some possible implementations, the model may be unable to parse the instructions issued by the user and unable to respond to the user's instructions. Summary of the invention

[0004] In order to solve the problems of the prior art, the present application provides a data processing method, apparatus, device and storage medium.

[0005] In a first aspect, the present application provides a data processing method, the method comprising:

[0006] Receiving first reply information sent by the target model, where the first reply information is determined by the target model according to the first prompt information;

[0007] Determining first display content according to the first reply information, where the first display content includes structured content;

[0008] In response to receiving user input information associated with the first display content, second prompt information is sent to the target model, the second prompt information includes the user input information and description information, and the description information includes natural language description information of the structured content.

[0009] In some possible implementations, determining the first display content according to the first reply information includes:

[0010] Execute tool calling according to the first reply information;

[0011] The first display content is determined based on the execution result of the tool call.

[0012] In some possible implementations, the method further includes:

[0013] The description information is generated based on a natural language description of the execution result of the tool call.

[0014] In some possible implementations, the generating the description information based on the natural language description of the execution result of the tool call includes:

[0015] In response to the execution result indicating that the first presentation content includes structured content, generating the description information based on pre-configured template information; or,

[0016] In response to the execution result indicating that the first display content includes structured content, third prompt information is generated to be sent to a target model based on the structured content and the example content, and return information of the target model is received as the description information, wherein the example content represents the association relationship between the structural information of the structured content and the natural language description information.

[0017] In some possible implementations, the description information also includes natural language description information generated by user operation information on the first display content.

[0018] In some possible implementations, before receiving the first reply information, the method further includes:

[0019] Obtaining a first instruction message;

[0020] The first prompt information is sent to the target model according to the first instruction message.

[0021] In some possible implementations, the first instruction message is received through an instant messaging (IM) session interface.

[0022] In some possible implementations, the structured content includes a target content card, and the description information is used to describe the target content card.

[0023] In some possible implementations, the structured content includes a first operation control, the first operation control is used to trigger a first operation, and the user input information is obtained according to the first operation.

[0024] In some possible implementations, the description information also includes natural language description information associated with the first operation control, wherein the natural language description information associated with the first operation control is used to describe a function of the first operation control and / or the first operation.

[0025] In some possible implementations, after sending the second prompt information, the method further includes:

[0026] Receiving second reply information sent by the target model;

[0027] Determine and display the second display content according to the second reply information.

[0028] In a second aspect, the present application provides a data processing method, comprising:

[0029] Receive input information;

[0030] In response to the input information including structured content, converting the structured content into natural language description information, wherein the natural language description information is used to describe the structured content in a natural language manner;

[0031] Based on the natural language description information, a data processing request is sent to the target model, and output information returned by the target model is received.

[0032] In some possible implementations, the natural language description information is generated in the following manner:

[0033] Based on pre-configured template information, convert the structured content to obtain the natural language description information; or,

[0034] Based on the structured content and the example content, third prompt information is generated to be sent to the target model, and return information from the target model is received, and natural language description information corresponding to the structured content is determined based on the return information, wherein the example content represents the association relationship between the structure information of the structured content and the natural language description information.

[0035] In some possible implementations, sending a data processing request to a target model based on the natural language description information includes:

[0036] Based on the natural language description information, the structured content included in the input information is updated, and the updated data is sent to the target model.

[0037] In some possible implementations, the structured content includes at least one of the following:

[0038] Display or store target content in a pre-set structure;

[0039] Response or operation information to the target content;

[0040] The target content includes at least one of the following: content information, control information, personnel information, code information, and business entity information.

[0041] In a third aspect, the present application provides a data processing device, the device comprising:

[0042] A receiving unit, configured to receive first reply information sent by a target model, wherein the first reply information is determined by the target model according to the first prompt information;

[0043] a determining unit, configured to determine first display content according to the first reply information, wherein the first display content includes structured content;

[0044] A sending unit is used to send second prompt information to the target model in response to receiving user input information associated with the first display content, wherein the second prompt information includes the user input information and description information, and the description information includes natural language description information of the structured content.

[0045] In some possible implementations, the determining unit is specifically configured to execute a tool call according to the first reply information; and determine the first display content based on an execution result of the tool call.

[0046] In some possible implementations, the determining unit is further configured to generate the description information based on a natural language description of an execution result of the tool call.

[0047] In some possible implementations, the determination unit is specifically used to generate the description information based on pre-configured template information in response to the execution result indicating that the first display content includes structured content; or, in response to the execution result indicating that the first display content includes structured content, generate third prompt information to be sent to the target model based on the structured content and example content, and receive return information from the target model as the description information, wherein the example content represents the association relationship between the structural information of the structured content and the natural language description information.

[0048] In some possible implementations, the description information also includes natural language description information generated by user operation information on the first display content.

[0049] In some possible implementations, the acquisition unit is further used to acquire a first instruction message; and the sending unit is further used to send the first prompt information to the target model according to the first instruction message.

[0050] In some possible implementations, the first instruction message is received through an IM session interface.

[0051] In some possible implementations, the structured content includes a target content card, and the description information is used to describe the target content card.

[0052] In some possible implementations, the structured content includes a first operation control, the first operation control is used to trigger a first operation, and the user input information is obtained according to the first operation.

[0053] In some possible implementations, the description information also includes natural language description information associated with the first operation control, wherein the natural language description information associated with the first operation control is used to describe a function of the first operation control and / or the first operation.

[0054] In some possible implementations, the receiving unit is further used to receive second reply information sent by the target model; the determining unit is further used to determine and display second display content according to the second reply information.

[0055] In a fourth aspect, the present application provides a data processing device, the device comprising:

[0056] A receiving unit, used for receiving input information;

[0057] a conversion unit, configured to, in response to the input information including structured content, convert the structured content into natural language description information, wherein the natural language description information is used to describe the structured content in a natural language manner;

[0058] A sending unit is used to send a data processing request to a target model based on the natural language description information, and receive output information returned by the target model.

[0059] In some possible implementations, the conversion unit is specifically used to convert the structured content based on pre-configured template information to obtain the natural language description information; or, based on the structured content and the example content, generate third prompt information to be sent to the target model, and receive return information from the target model, and determine the natural language description information corresponding to the structured content based on the return information, wherein the example content represents the association relationship between the structural information of the structured content and the natural language description information.

[0060] In some possible implementations, the conversion unit is specifically used to update the structured content included in the input information based on the natural language description information; and the sending unit is specifically used to send the updated data to the target model.

[0061] In some possible implementations, the structured content includes at least one of the following: displaying or storing the target content in a preset structure; response or operation information to the target content; the target content includes at least one of the following: content information, control information, personnel information, code information, business entity information.

[0062] In a fifth aspect, the present application provides an electronic device, including:

[0063] one or more processors;

[0064] a storage device having one or more programs stored thereon,

[0065] When the one or more programs are executed by the one or more processors, the one or more processors implement any one of the methods described in the first aspect or the second aspect.

[0066] In a sixth aspect, the present application provides a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any one of the first aspect or the second aspect is implemented.

[0067] In a seventh aspect, the present application provides a computer program product, which, when executed on a device, enables the device to execute the method described in the first aspect or the second aspect.

[0068] The present application provides a data processing method, apparatus, device and storage medium, which can better process the user's instructions. Specifically, first, a first reply message sent by a target model can be received. The first reply message is determined by the target model according to the first prompt information, and the first prompt information is information that has been sent to the target model. Then, the first display content can be determined according to the first reply message. The first display content includes structured content, so the computer device can parse the first display content and display the first display content to the user. After viewing the first display content, the user can trigger an operation for the first display content. Accordingly, user input information associated with the first display content can be obtained. After obtaining the user input information, a second prompt information can be sent to the target model. The second prompt information includes not only the user input information but also the description information. The description information includes natural language description information of the structured content. That is, the description information is the structured content described in natural language. Since the description information is natural language description information, the target model has the ability to parse the description information. That is, if the user triggers further input to the content fed back by the target model, the prompt information sent to the target model includes not only the user input information but also the natural language description information of the structured content. In this way, if the target model can determine the first display content according to the description, and thus parse the user input information associated with the first display content, the target model can accurately determine the user's intention, and thus give a response that matches the user's intention. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0070] Figure 1 A flowchart of a data processing method provided in an embodiment of the present application;

[0071] Figure 2 An interactive schematic diagram of a data processing method provided in an embodiment of the present application;

[0072] Figure 3 A flowchart of a data processing method provided in an embodiment of the present application;

[0073] Figure 4 A schematic diagram of the structure of a data processing device provided in an embodiment of the present application;

[0074] Figure 5 A structural example diagram of a data processing device provided in an embodiment of the present application;

[0075] Figure 6 A schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0076] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0077] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0078] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0079] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different objects, devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0080] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0081] With the development of computer technology, some models are endowed with the ability to parse instructions issued by users. Accordingly, in some application scenarios, this ability can be used to improve user experience. Specifically, users can input instruction information through voice or text on the client. The instruction information can be sent to the model through the network. The model can parse the instruction information input by the user, determine the user's intention, and perform corresponding actions according to the user's intention.

[0082] In order to further improve the user experience, some models also have the ability to learn context. That is, the model can determine the corresponding result based on the historical instructions issued by the user. For example, models with natural language processing (NLP) capabilities, such as language models (LM), can not only understand the messages sent by users, but also learn context based on the user's historical interactive messages, so as to better reply to the messages sent by users.

[0083] In some possible implementations, a model with context learning capabilities may be used to parse the user's instructions. Specifically, the user may input instruction information in the form of natural language text through the client. The instruction information is forwarded to the model. The model may parse the instruction information and determine the operation information corresponding to the user's intention, so as to respond according to the user's intention.

[0084] If a model with natural language processing is used to process the user's instructions, the model can learn the context of the instructions issued by the user. In this way, the user can issue instructions multiple times. For the first instruction issued by the user, the model can process the instruction and feedback the processing result to the user. If the user is not satisfied with the processing result or needs other processing, the user can issue a new instruction again. After obtaining the new instruction, the model can learn the context of the processed instructions, so as to better process the new instruction information.

[0085] However, in some possible implementations, the model may not be able to respond to the user's operations. Specifically, assume that the model parses the instructions triggered by the user and determines the corresponding data processing instructions based on the user's intent. Based on the data processing instructions, the server can perform corresponding data processing according to the user's intent and display the processed data to the user. If the user triggers other operations on the processed data. Then when the model performs subsequent processing, it not only needs to process the user's operations, but also needs to perform context learning on the above-mentioned processed data. However, if the model does not have the ability to process the processed data, it will not be able to respond to the new operations triggered by the user.

[0086] For example, suppose the user first sends a natural language description A. The model can parse the natural language description A to obtain the operation information corresponding to the user's intention. Based on the operation information, the server can create a content card A1 and display the content card A1 on the client used by the user. If the user triggers further operations on the content card A1, since the content card A1 is structured content, the model can neither determine the relevant operations performed by the user on the content card A1 nor perform context learning on the content card A1. Therefore, it is impossible to process the operation triggered by the user.

[0087] That is to say, if a user triggers an operation on structured content, the model cannot parse the operation triggered by the user on the structured content because it does not have the ability to parse structured content.

[0088] In order to solve the problems of the prior art, an embodiment of the present application provides a data processing method, which is described in detail below in conjunction with the drawings in the specification.

[0089] See also Figure 1 , Figure 1 A flow chart of a data processing method provided in an embodiment of the present application, the embodiment of the present application may be applicable to an application scenario in which an action is to be performed according to a user's instruction. Among them, the data processing method may be executed by a data processing device. The data processing device may be implemented by software. Optionally, the device may include a model for determining operation information according to user instructions, or the device may be associated with a model for determining operation information according to user instructions. Optionally, the above-mentioned model may include, for example, an LM model. In an embodiment of the present application, the model may be referred to as a target model.

[0090] Optionally, the data processing device can be integrated into the user's client. The client can be an application client or a web client of the IM software. If the client is a software client of the IM software, the client can be integrated into a personal computer (PC) or a mobile terminal. If the data processing device is integrated into the client, the data processing device can communicate with the target model of the server to send information such as user input and description information to the server. Alternatively, the data processing device can also be integrated into the server. The server can be the server of the IM software, which is used to provide corresponding services to the client of the IM software. Optionally, the data processing device can be integrated with a module on the server for implementing an action execution function. For ease of introduction, the following is an example of the data processing device being integrated into the server.

[0091] like Figure 1 As shown, the method specifically comprises the following steps:

[0092] S101: Receive the first reply information sent by the target model.

[0093] In an embodiment of the present application, a data processing device may receive a first reply message sent by a target model. The first reply message is used to reply to the user's historical input information. The user's historical input information may be information sent by the user to the target model before obtaining the first reply message, and may also be referred to as a first instruction message. In an embodiment of the present application, the target model may be a model with NLP capabilities, and accordingly, the historical input information may be information described in natural language.

[0094] That is, before obtaining the first reply information, the user may send natural language description information to the target model through the data processing device. The target model may process the natural language description information sent by the user to determine the reply information corresponding to the natural language description information.

[0095] In an embodiment of the present application, the data processing device may send relevant information to the target model through prompt information. Accordingly, after obtaining the historical user input information, the data processing device may generate first prompt information according to the historical user input information, and send the first prompt information to the target model. The target model may parse the first prompt information to determine the first reply information.

[0096] The following first introduces in detail the process of determining the first prompt information.

[0097] As described above, the model can perform corresponding actions according to the user's instructions. Accordingly, the data processing device can generate a first prompt message according to the information input by the user and send it to the target model. In the embodiment of the present application, the information input by the user represents the instruction triggered by the user and reflects the user's intention. By parsing the information input by the user, it can be determined which actions the user needs the data processing device to perform.

[0098] In the embodiment of the present application, the user can operate the client so that the data processing device obtains the historical user input information. For the convenience of introduction, the above operation is referred to as the first operation below.

[0099] In some possible implementations, the first operation may include an operation of text information input by a user. That is, the user may issue an instruction to the data processing device by inputting text information to instruct the data processing device to perform a corresponding action. Optionally, the first operation may also include a user triggering an operation of a send control. For example, the user may first input text information in an input box (or other input controls) and then trigger a send control. Then, the client may send the text information input by the user to the server, so that the data processing device obtains the text information input by the user.

[0100] Alternatively, the first operation may also include a preset operation triggered by a user on a preset control. That is, the user may instruct the data processing device to perform a corresponding action by triggering a preset operation (such as a single-click operation) on the preset control. Alternatively, the first operation may also include an operation in which the user issues a preset voice.

[0101] After acquiring the first operation, the client may generate a first instruction message. The first instruction message includes historical user input information. The client may send the first instruction message to the server. The server may determine the historical user input information according to the first instruction message, thereby generating the first prompt information.

[0102] The following is an introduction using historical user input information including text information input by the user as an example.

[0103] Optionally, the user may input text information through an input control. The input control may be an input box on the client. For example, if the client is an IM client, the input control may be an input box in an IM session, and the user may input text information in the input box and send the text information to the server through an IM message. The IM message includes the text information. The historical user input information may include the text information.

[0104] Accordingly, the first operation may be an operation of sending an IM message. Accordingly, the first instruction message may be an IM message. The client may receive historical user input information and the first instruction message sent by the user through the IM session interface. That is, the user may input text information in the IM session and trigger a sending operation. The client may generate an IM message based on the location information input by the user and send it to the server. The data processing device may obtain the IM message sent by the client and determine the historical user input information based on the IM message.

[0105] In some possible implementations, services may be provided to users in the form of a fixed IM session. The fixed IM session may be referred to as a target IM session, and is used to receive IM messages sent by users. That is, if the user wants to trigger the first operation, the user may initiate a target IM session and send an IM message through the target IM session. The data processing device may obtain the target IM message sent by the user, and use the text information in the target IM message as historical user input information.

[0106] Optionally, the target IM session may be initiated by the user. For example, the user may initiate the target IM session through the contact list or service number list in the IM client. Specifically, after the user turns on the corresponding function, the target IM contact may be added to the user's contact list, and the user may initiate the target IM session through the target IM contact. Alternatively, after the user turns on the corresponding function, the target IM session may be added to the user's IM session list. Accordingly, the user may find the target IM session by sliding browsing or searching to send the target IM message.

[0107] Optionally, the text information in the above-mentioned target IM message may be text information in text format input by the user through a keyboard or a virtual keyboard, or may include text information obtained by converting audio input by the user. Specifically, the user may input text information through a text box, for example, through an IM message input box corresponding to the IM software. Alternatively, the user may also input voice information and send it to the data processing device in the form of a voice IM message. The data processing device may parse the voice IM message and convert the voice input by the user into text information to obtain historical user input information.

[0108] For example, user A can send an IM message to the target IM dialogue corresponding to the intelligent robot through the IM client. The content of the IM message includes "Please help me schedule a meeting at 10:00 am tomorrow." The IM message is sent to the IM server through the network and then forwarded to the corresponding language model. Accordingly, the historical user input information includes the text "Please help me schedule a meeting at 10:00 am tomorrow."

[0109] After the historical user input information is acquired, the first prompt information may be generated according to the historical user input information. Optionally, if the user inputs other information before inputting the historical user input information, the information may also be carried in the first prompt information.

[0110] The above describes a method for determining the first prompt information. The following describes a method for determining the first reply information based on the first prompt information.

[0111] After obtaining the first prompt information, the data processing device may send the first prompt information to the target model so that the target model determines the first reply information according to the first prompt information. Optionally, if the data processing device runs on the server, the data processing device may send the first prompt information to the target model through the interface of the cloud service corresponding to the target model.

[0112] According to the first prompt information, the target model can perform semantic analysis on the information input by the user to determine the corresponding first reply information. The first reply information is used to reply to the information input by the user. Specifically, the first reply information may include action indication information. The action indication information is used to indicate the action that the user wants the client to perform. Accordingly, after obtaining the first reply information, the target model can parse the first reply information to perform the corresponding action according to the user's instruction.

[0113] S102: Determine first display content according to the first reply information.

[0114] After determining the first reply information, the target model may send the target reply information to the data processing device so as to display the corresponding content to the user. In an embodiment of the present application, the content corresponding to the first reply information and used to display to the user may be referred to as the first display content. The first display content includes structured content.

[0115] If the data processing device runs on the client, the data processing device can determine the first display content according to the first reply information, and display the first display content to the user through the client. Alternatively, if the data processing device runs on the server, the data processing device can first determine the first display content, and then send the first display content to the client through the network, so as to display the first display content to the user.

[0116] The following describes a method for determining the first display content.

[0117] In an embodiment of the present application, the first display content can be determined by calling a tool. Specifically, after obtaining the first reply information, the tool call can be executed according to the first reply information, and then the first display content can be determined according to the execution result of the tool call. The tool call refers to data processing by calling a tool.

[0118] Specifically, the target model can determine the user's intention. Accordingly, the first reply content can include operation information obtained by understanding and reasoning the user's intention. In this way, according to the first reply content, the operation corresponding to the user's intention can be performed according to the understanding and reasoning of the user's intention. Specifically, after obtaining the first reply content, the corresponding tool can be called according to the operation information in the first reply content, and the tool can be controlled to perform corresponding data processing according to the user's intention. The result obtained after the tool completes the data processing is the execution result of the above tool call.

[0119] In the embodiment of the present application, the structure of the tool call may be structured content. Therefore, the first display content determined according to the result of the tool call may also be structured content. Then, after obtaining the execution result of the tool call, the description information may be generated based on the natural language description of the execution result of the tool call. In other words, the execution result of the tool call may be described in natural language to obtain the description information. For an introduction to the description information, please refer to the following, which will not be repeated here.

[0120] The above describes a method for determining the first display content, and the following describes a method for displaying the first display content.

[0121] In the embodiment of the present application, the first display content is determined according to the first reply information, which is in turn determined according to the historical user input information. Accordingly, the first display content can be displayed based on the way of obtaining the historical user input information.

[0122] For example. If the historical user input information is input by the user in the search input box of the search interface, then after the first display content is determined, the first display content can be displayed to the user in the result display area of ​​the search interface. Alternatively, if the historical user input information is input by the user in the input box of the IM message interface, then after the first display content is determined, the first display content can be displayed to the user in the IM message interface. Specifically, a reply IM message can be sent to the user, and the reply IM message includes the first display content.

[0123] The above example is used for explanation. After obtaining the IM message "Please help me schedule a meeting at 10 o'clock tomorrow morning" sent by user A, the data processing device can parse the IM message sent by the user through the target model to determine the first reply information. Then, the data processing device can then call the schedule creation tool to create a meeting schedule for 10 o'clock the next morning. Then, the tool for sending IM messages can be called, for example, calling the interface corresponding to the IM software, to send a schedule notification message to the IM client of user A, so that the IM client of user A displays the meeting schedule to user A. In this embodiment, the first display content can be determined at least in part based on the content in the schedule notification message.

[0124] According to the above description, the first operation may be an operation of sending an IM message by a user through a target IM session. Accordingly, the first display content may also be displayed based on the target IM session. That is, an IM message sent by a user may be received and an IM message may be sent to the user through the same IM session. For example, the data processing device may call an IM tool to generate a reply IM message, and send the reply IM message to the client used by the user. The specific content included in the reply IM message is the above-mentioned first display content.

[0125] In some possible implementations, the first display content may include a target content card. The content card is a form of information display, and the corresponding information can be displayed to the user according to a preset format. That is, after the first reply information is determined, a tool call can be performed according to the first reply information to obtain the corresponding content card, and the content card can be displayed to the user.

[0126] Specifically, the format and related conditions of the content card may be preset. After the first reply information is acquired, the first display content may be generated according to the format of the content card and displayed to the user.

[0127] For example, the format of the content card of the schedule type can be preset, and the relevant information of the schedule reserved by the user can be displayed to the user in the format of the content card. If the first reply information indicates that the user intends to create a schedule, then after the schedule is created, the relevant information of the schedule can be displayed to the user in the form of a content card. The content card related to the schedule is the first display content.

[0128] For ease of introduction, in the embodiment of the present application, the content card corresponding to the first reply information may be referred to as a target content card.

[0129] According to the above description, the first reply information can be displayed to the user via an IM message. Accordingly, the target content card can be sent to the user via an IM message. Specifically, if the first instruction message is obtained based on an IM message sent by the user via a target IM session, the data processing device can send an IM message to the user via the target IM session, and the content of the IM message includes the target content card.

[0130] In some possible implementations, the user can perform further operations on the relevant information displayed by the content card. For example, the user can confirm the information displayed by the content card, or the user can reject or adjust the information displayed by the content card. Optionally, the user can trigger the above operation by sending a new IM message. Alternatively, the user can also trigger the above operation through the content card.

[0131] Specifically, the target content card displayed to the user may include at least one operation control. The user may trigger a corresponding operation through the operation control to instruct the target model to perform a corresponding action.

[0132] In actual application scenarios, the target content card may correspond to a display area on the IM client. In addition, the background color of the target content card may be different from that of other display areas, and / or the target content card may be separated from other display areas by a border to distinguish the target content card from non-target content cards. The target content card includes an information display area for displaying information and an operation control for triggering operations. For an introduction to triggering operations through target content cards, please refer to the following text and will not be repeated here.

[0133] After determining the first display content, the data processing device may also determine description information corresponding to the first display content. The description information is used to describe the first display content in natural language. In the embodiment of the present application, the description information may also be referred to as hint information.

[0134] For example, the description information may include a natural language description of the function of a certain control in the first display content. Since the description information is natural language description information, the target model has the ability to parse the description information. Optionally, if the first display content is obtained according to the execution result of the tool call, the description information may also be a natural language description of the execution result.

[0135] According to the foregoing description, in some implementations, the first display content may include structured content. However, the target model does not have the ability to parse structured content. If the user performs further operations on the first display content, the target model cannot parse the structured content in the first display content, and thus cannot parse the user's operations. Therefore, description information based on natural language can be generated based on the first display content. In this way, the target model can understand the first display content based on the description information, and accordingly understand the user's further operations on the first display content, so that the target model can process subsequent operations triggered by the user.

[0136] The previous example is used for explanation. Assume that the first display content includes a content card corresponding to the schedule. Then when determining the description information, the content card can be described in natural language. Optionally, the content card corresponding to the first display content may include a display area for displaying information and an operation control. The operation control is used to trigger related operations. Accordingly, the description information may include description information corresponding to the information displayed by the content card, and description information corresponding to the operation control. The description information corresponding to the operation control can be used to describe the function of the operation control, for example, to describe the action to be performed after the operation control is triggered.

[0137] The description information is introduced above. The following describes the method of determining the description information.

[0138] In an embodiment of the present application, the description information may be a natural language description based on the execution result of the tool call. Accordingly, after obtaining the execution result through the tool call, the execution result may be described in natural language to obtain the corresponding description information. In some possible implementations, template information may be preconfigured. The template information may record the natural language expression of each structure in the structured information. In this way, after obtaining the execution result, it may be determined whether the first display content indicated by the execution result includes structured content. If included, the structured content in the first display content may be converted into corresponding natural language description information based on the template information.

[0139] In a second possible implementation, the description information may be determined by the model. Specifically, after determining that the first display content indicated by the execution result includes structured content, prompt information including the structured content may be sent to the model, and the structured content may be converted into description information by the model.

[0140] For example, the data processing device can generate third prompt information based on the structured content and the sample content, and send the third prompt information to the model. The sample content represents the association relationship between the structured content's structural information and the natural language description information. Based on the sample content, the model can determine the association relationship between the structured content's structural information and the natural language description information, thereby converting the structured content into description information according to the association relationship. Then, the model can send return information to the data processing device. The return information includes the natural language description information corresponding to the structured content. It can be understood that the model here can be a target model or a model other than the target model.

[0141] After the description information is determined, the description information can be stored. Optionally, the description information can be stored in the memory or storage of the server. Alternatively, the description information can be stored through cloud storage or other means. Accordingly, after obtaining further instruction information, the stored description information can be called, so as to generate prompt information according to the description information and send it to the model. For an introduction to this part, please refer to the following, which will not be repeated here.

[0142] S103: In response to receiving user input information associated with the first display content, sending second prompt information to the target model.

[0143] After the first display content is displayed to the user, the user can trigger other operations with respect to the first display content. Alternatively, the user can also input additional text information with respect to the first display content. In other words, the user can further input other information with respect to the first associated content. Accordingly, the data processing device can obtain the user input information associated with the first display content.

[0144] After obtaining the user input information, the data processing device may generate second prompt information and send the second prompt information to the target model. The second prompt information includes the user input information and description information associated with the first display content. Similar to the first prompt information, the second prompt information may also be prompt information.

[0145] The user input information included in the second prompt information is the information newly input by the user after the first display content. Optionally, the user input information can also be called a second instruction message.

[0146] The following first introduces user input information.

[0147] In an embodiment of the present application, the user input information can be obtained based on text information input by the user, or based on an operation triggered by the user.

[0148] In some possible implementations, after viewing the first display content, the user can input new text information for the first display content and send it to the data processing device. Accordingly, the user input information may include text information input by the user. Optionally, the text information input by the user may be a natural language description of the operation that the user wants to perform on the first display content.

[0149] For example, if the user wants to confirm the first displayed content, the user can enter text information such as "confirm". Or, if the user wants to modify the first displayed content, the user can enter text information such as "reject" or enter the specific content to be modified.

[0150] In some other possible implementations, the user input information may be obtained according to an operation triggered by the user. For ease of introduction, the operation that may be used to trigger the user input information is referred to as a second operation. After the user triggers the second operation, the client may generate a second instruction message according to the second operation and send the second instruction message to the data processing device. The data processing device may determine the user input information according to the second instruction message to obtain the second prompt information.

[0151] Optionally, the second operation may be an operation triggered for the first display content. That is, after seeing the first display content, the user may trigger the second operation for the first display content. Then, the data processing device may obtain the second instruction message and generate a second prompt message. Accordingly, the structured content in the first display content may include an operation control for triggering the second operation. In an embodiment of the present application, the operation control for triggering the second operation may be referred to as a first operation control.

[0152] The above example is used for explanation. Assuming that the target content card is a schedule-type content card, the target content card may include a confirmation operation control for triggering a confirmation operation and a rejection operation control for triggering a rejection operation. If the user triggers the confirmation operation control, that is, the second operation is a confirmation operation triggered by the confirmation operation control, then the second instruction message may include the relevant identifier of the confirmation operation control. Alternatively, if the user triggers the rejection operation control, that is, the second operation is a rejection operation triggered by the rejection operation control, then the second instruction message may include the identifier of the rejection operation control.

[0153] The second instruction message is introduced above, and the second prompt message is introduced below.

[0154] In the embodiment of the present application, the second prompt information includes user input content and description information. Accordingly, after obtaining the user input content, it can be determined whether the user input content is associated with the structured content in the first display content. If the user input content is associated with the structured content in the first display content, the data processing device can obtain the description information corresponding to the first display content and generate the second prompt information.

[0155] According to the above introduction, the target model is a model with NLP capabilities. Accordingly, the target model can process and / or learn context of information described in natural language. The first display information includes structured content. Therefore, the target model does not have the ability to process structured content, and the target model cannot process user input information associated with the structured content.

[0156] To solve this problem, description information can be carried in the second prompt information. In this way, since the description information is natural language description information of structured content, the target model can determine the structured content in the first display content according to the description information, thereby processing the user input information.

[0157] The present application provides a data processing method, which can better process the user's instructions. Specifically, first, a first reply message sent by a target model can be received. The first reply message is determined by the target model according to the first prompt information, and the first prompt information is information that has been sent to the target model. Then, the first display content can be determined according to the first reply message. The first display content includes structured content, so the computer device can parse the first display content and display the first display content to the user. After viewing the first display content, the user can trigger an operation for the first display content. Accordingly, user input information associated with the first display content can be obtained. After obtaining the user input information, a second prompt information can be sent to the target model. The second prompt information includes not only the user input information but also description information. The description information is the natural language description information of the structured content in the first display content. That is, the description information is the structured content described in natural language. Since the description information is natural language description information, the target model has the ability to parse the description information. That is, if the user triggers further input to the content fed back by the target model, the prompt information sent to the target model includes not only the user input information but also the natural language description information of the content fed back by the target model. In this way, if the target model can determine the first display content according to the description, and thus parse the user input information associated with the first display content, the target model can accurately determine the operation information corresponding to the user's intention, and thus give a response that matches the user's intention.

[0158] The following is a further introduction to the data processing method provided in the embodiment of the present application in combination with actual application scenarios.

[0159] See also Figure 2 , Figure 2 An interactive schematic diagram of a data processing method provided in an embodiment of the present application, Figure 2 The corresponding embodiment can be applied to the process of processing user input information through the target model. Specifically, it is applicable to the application scenario where the user instructs the server to perform the corresponding operation through the client. Figure 2 In the implementation shown, the data processing device is implemented based on a server, wherein the client may be an IM client running on a user's terminal device, and the server may be an IM software server for providing services to the IM client.

[0160] like Figure 2 As shown, the method specifically comprises the following steps:

[0161] S201: The client sends a first instruction message to the server.

[0162] If the user triggers the first operation on the client, the client may send a first instruction message to the server. In some application scenarios, the first instruction message may include historical user input information. For an introduction to the first instruction message, please refer to the above text and will not be repeated here.

[0163] S202: The server sends a first prompt message to the target model.

[0164] After acquiring the first instruction message, the server may generate first prompt information according to the first instruction message, and send the first prompt information to the target model.

[0165] Optionally, if the first instruction message is the first instruction message triggered by the client, then the first prompt information may include the first instruction message. If the first instruction message is not the first instruction message triggered by the client, then the first prompt information may include the first instruction message, and historical instruction information and / or historical reply information corresponding to the first instruction message. If the first operation is triggered by the user based on the displayed content, then the first prompt information may include the first instruction message and description information corresponding to the displayed content.

[0166] It should be noted that the "first" here is the first in a logical sense. For example, after the user ends using the cloud service corresponding to the target model, if the user reuses the target model, the first instruction information sent when reusing the target model can also be regarded as the first instruction information.

[0167] S203: The target model determines first reply information according to the first prompt information.

[0168] After obtaining the first prompt information, the target model can determine the first reply information corresponding to the first instruction message according to the first prompt information. Optionally, if the first prompt information includes other information (such as historical instruction information, historical reply information or description information, etc.) in addition to the first instruction message, the target model can first perform context learning on the above other information, and process the first instruction message after the context learning to obtain the corresponding first reply information.

[0169] S204: The target model sends a first reply message to the server.

[0170] After determining the first reply information, the target model may send the first reply information to the server.

[0171] S205: The server determines the first display content and description information according to the first reply information.

[0172] After receiving the first reply information sent by the target model, the server can call the tool according to the first reply information, and determine the first display content according to the execution result of the tool call. In addition, if the first display content includes structured content, the server can generate corresponding description information according to the first display content. For the introduction of determining the first display content and description information, please refer to the above, which will not be repeated here.

[0173] S206: The server sends the first display content to the client.

[0174] After determining the first display content, the server may send the first display content to the client.

[0175] S207: The client displays the first display content.

[0176] The client may display the first display content to the user so that the user can view the first display content. For example, the client may display the first display content to the user via an IM message.

[0177] S208: The client sends a second instruction message to the server.

[0178] After receiving the first display content, the user can trigger a second operation on the first display content. For example, the user can trigger an operation such as a click operation on a control on the first display content. Accordingly, the client can obtain the second operation triggered by the user on the first display content.

[0179] After acquiring the second operation, the client may send a second instruction message to the server, wherein the second instruction message includes user input information associated with the first display content.

[0180] S209: The server generates second prompt information according to the second instruction message and the description information.

[0181] After receiving the second instruction message sent by the client, the server can generate the second prompt information according to the second instruction message and the description information. The introduction of generating the second prompt information can be found above and will not be repeated here.

[0182] S210: The server sends a second prompt message to the target model.

[0183] In order to process the second instruction message, the server may send a second prompt message to the target model.

[0184] S211: The target model determines the second reply information according to the second prompt information.

[0185] After acquiring the second prompt information, the target model may perform context learning on the description information. After performing context learning, the target model may process the second prompt information to obtain second reply information corresponding to the second instruction message.

[0186] S212: The target model sends a second reply message to the server.

[0187] After determining the second reply information, the target model may send the second reply information to the server. The server may determine the second display content according to the second reply information, and display the second display content to the user through the client.

[0188] In the implementation described above, if the user input information is triggered based on structured content, prompt information can be generated based on the description information corresponding to the pre-stored structured content so that the model can process the user input information. In some other possible implementations, the input information may include structured content. Accordingly, the embodiment of the present application also provides a data processing method that can convert the structured content in the input information into natural language description information. Figure 2 Make an introduction.

[0189] See also Figure 3 , Figure 3 A flowchart of a data processing method provided in an embodiment of the present application is shown below. Figure 3 The corresponding embodiment can be applied to the process of processing user input information through the target model. Figure 3 In the implementation shown, the data processing device can be implemented based on the server. The client can be an IM client running on the user's terminal device, and the server can be an IM software server for providing services to the IM client. Optionally, Figure 3 The server and Figure 1The server in the embodiment shown can be the same server or different server.

[0190] like Figure 3 As shown, the method specifically comprises the following steps:

[0191] S301: receiving input information.

[0192] exist Figure 3 In a corresponding implementation, the input information may include structured content. The input information may be, for example, information input by a user, or input information corresponding to a user instruction. For example, a user copies a content card, pastes the content card in an input box corresponding to a target IM session, and sends the content card to the model. In this implementation, the input information includes the content card pasted by the user. The content card is the structured content included in the input information.

[0193] In the embodiment of the present application, the structured content may include displaying or storing the target content in a preset structure, and may also include response or operation information to the target content. The response or operation information to the target content may include, for example, user-triggered operations on the target content, such as accepting operations, confirming operations, and the like. The target content may include at least one of content information, control information, personnel information, code information, and business entity information.

[0194] S302: In response to the input information including structured content, convert the structured content into natural language description information.

[0195] After obtaining the input information, the data processing device can determine whether the input information includes structured content. If the input information does not include structured content, the data processing device can generate prompt information based on the input information and send it to the target model. Optionally, if the input information is associated with structured content, the following method can be used: Figure 1 The corresponding implementation method calls back the description information corresponding to the structured content and generates prompt information.

[0196] If the input information includes structured content, in order for the model to process the input information, the structured content in the input information can be converted into natural language description information.

[0197] In some possible implementations, the structured content can be converted based on pre-configured template information to obtain natural language description information corresponding to the input information. Alternatively, in some other possible implementations, prompt information can also be generated based on the structured content and the example content, and the prompt information can be sent to the target model. The target model can learn the association between the structural information of the structured content and the natural language description information based on the example information, and thus convert the structured content into natural language description information based on the association and feed it back to the data processing device through the return information. For the introduction to determining the natural language description information, please refer to the above, which will not be repeated here.

[0198] S303: Based on the natural language description information, a data processing request is sent to the target model, and output information returned by the target model is received.

[0199] After obtaining the natural language description information, the data processing device may send a data processing request to the target model and receive output information returned by the target model. The data processing request is generated according to the natural language description information. The output information returned by the target model may be determined by the target model according to the natural language description information.

[0200] Optionally, after obtaining the natural language description information, the data processing device may update the structured content contained in the input information based on the natural language description information. In this way, the structured content in the updated input information is replaced by the natural language description information. Then, the updated input information is sent to the target model. Since the original structured content in the updated input information is replaced by the natural language description information, the target model can process the updated input information, thereby returning output information to the data processing device.

[0201] It can be seen that even if the input information includes structured content, the structured content in the input information can be converted into natural language description information. In this way, if the input information includes a pre-set structure to display or store the target content, or includes corresponding or operational information on the target content, the input information can be converted into a natural language description so that the target model can process the input information.

[0202] Based on a data processing method provided by the above method embodiment, an embodiment of the present application also provides a data processing device, which will be described below in conjunction with the accompanying drawings.

[0203] See also Figure 4 As shown in FIG. 1 , this figure is a schematic diagram of the structure of a data processing device provided in an embodiment of the present application. Figure 4 As shown, the data processing device 400 includes:

[0204] A receiving unit 410 is configured to receive first reply information sent by a target model, where the first reply information is determined by the target model according to first prompt information;

[0205] A determining unit 420, configured to determine first display content according to the first reply information, where the first display content includes structured content;

[0206] The sending unit 430 is used to send second prompt information to the target model in response to receiving user input information associated with the first display content, wherein the second prompt information includes the user input information and description information, and the description information includes natural language description information of the structured content.

[0207] In some possible implementations, the determining unit 420 is specifically configured to execute a tool call according to the first reply information; and determine the first display content based on an execution result of the tool call.

[0208] In some possible implementations, the determining unit 420 is further configured to generate the description information based on a natural language description of the execution result of the tool call.

[0209] In some possible implementations, the description information also includes natural language description information generated by user operation information on the first display content.

[0210] In some possible implementations, the receiving unit 410 is further configured to obtain a first instruction message; and the sending unit 430 is further configured to send the first prompt information to the target model according to the first instruction message.

[0211] In some possible implementations, the first instruction message is received through an IM session interface.

[0212] In some possible implementations, the structured content includes a target content card, and the description information is used to describe the target content card.

[0213] In some possible implementations, the target content card includes a first operation control, the first operation control is used to trigger a first operation, and the user input information is obtained according to the first operation.

[0214] In some possible implementations, the description information also includes natural language description information associated with the first operation control, wherein the natural language description information associated with the first operation control is used to describe a function of the first operation control and / or the first operation.

[0215] In some possible implementations, the receiving unit 410 is further used to receive second reply information sent by the target model; the determining unit 420 is further used to determine and display second display content according to the second reply information.

[0216] Based on a data processing method provided by the above method embodiment, an embodiment of the present application also provides a data processing device, which will be described below in conjunction with the accompanying drawings.

[0217] See also Figure 5 As shown in FIG. 1 , this figure is a schematic diagram of the structure of a data processing device provided in an embodiment of the present application. Figure 5 As shown, the data processing device 500 includes: a receiving unit 510, used to receive input information; a conversion unit 520, used to convert the structured content into natural language description information in response to the input information including structured content, and the natural language description information is used to describe the structured content in a natural language; a sending unit 530, used to send a data processing request to the target model based on the natural language description information, and receive output information returned by the target model.

[0218] In some possible implementations, the conversion unit 520 is specifically used to convert the structured content based on pre-configured template information to obtain the natural language description information; or, based on the structured content and the example content, generate a third prompt information to be sent to the target model, and receive return information from the target model, and determine the natural language description information corresponding to the structured content based on the return information, wherein the example content represents the association between the structural information of the structured content and the natural language description information.

[0219] In some possible implementations, the conversion unit 520 is specifically used to update the structured content included in the input information based on the natural language description information; and the sending unit is specifically used to send the updated data to the target model.

[0220] In some possible implementations, the structured content includes at least one of the following: displaying or storing the target content in a preset structure; response or operation information to the target content; the target content includes at least one of the following: content information, control information, personnel information, code information, business entity information.

[0221] Based on a data processing method provided by the above method embodiment, the present application also provides an electronic device, including: one or more processors; a storage device, on which one or more programs are stored, when the one or more programs are executed by the one or more processors, the one or more processors implement the data processing method as described in any of the above embodiments.

[0222] Reference below Figure 6 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (portable android devices), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs (televisions), desktop computers, etc. Figure 6 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0223] like Figure 6 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0224] Typically, the following devices may be connected to the I / O interface 605: an input device 608 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0225] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

[0226] The electronic device provided in the embodiment of the present application and the data processing method provided in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0227] Based on a data processing method provided in the above method embodiment, an embodiment of the present application provides a computer storage medium on which a computer program is stored, wherein when the program is executed by a processor, the data processing method as described in any of the above embodiments is implemented.

[0228] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0229] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (Hyper Text Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0230] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0231] The computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device executes the data processing method.

[0232] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0233] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0234] The units involved in the embodiments described in this application may be implemented by software or hardware. The name of a unit / module does not limit the unit itself in some cases. For example, a voice data acquisition module may also be described as a "data acquisition module".

[0235] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0236] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0237] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the system or device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0238] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0239] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0240] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A data processing method, characterized in that: The method comprises: Receiving first reply information sent by the target model, where the first reply information is determined by the target model according to the first prompt information; Determining first display content according to the first reply information, where the first display content includes structured content; In response to receiving user input information associated with the first display content, second prompt information is sent to the target model, the second prompt information includes the user input information and description information, and the description information includes natural language description information of the structured content.

2. The method according to claim 1, characterized in that The determining the first display content according to the first reply information includes: Execute tool calling according to the first reply information; The first display content is determined based on the execution result of the tool call.

3. The method according to claim 2, characterized in that The method further comprises: The description information is generated based on a natural language description of the execution result of the tool call.

4. The method according to claim 3, characterized in that The generating the description information based on the natural language description of the execution result of the tool call includes: In response to the execution result indicating that the first presentation content includes structured content, generating the description information based on pre-configured template information; or, In response to the execution result indicating that the first display content includes structured content, third prompt information is generated to be sent to a target model based on the structured content and the example content, and return information of the target model is received as the description information, wherein the example content represents the association relationship between the structural information of the structured content and the natural language description information.

5. The method according to claim 1, characterized in that The description information also includes natural language description information generated by the user's operation information on the first display content.

6. The method according to claim 1, characterized in that Before receiving the first reply information, the method further includes: Obtaining a first instruction message; The first prompt information is sent to the target model according to the first instruction message.

7. The method according to claim 6, characterized in that The first instruction message is received through an instant messaging IM session interface.

8. The method according to claim 1, characterized in that The structured content includes a target content card, and the description information is used to describe the target content card.

9. The method according to claim 1, characterized in that: The structured content includes a first operation control, the first operation control is used to trigger a first operation, and the user input information is obtained according to the first operation.

10. The method according to claim 9, characterized in that The description information also includes natural language description information associated with the first operation control, wherein the natural language description information associated with the first operation control is used to describe a function of the first operation control and / or the first operation.

11. The method according to claim 1, characterized in that: After sending the second prompt information, the method further includes: Receiving second reply information sent by the target model; Determine and display the second display content according to the second reply information.

12. A data processing method, characterized in that: include: Receive input information; In response to the input information including structured content, converting the structured content into natural language description information, wherein the natural language description information is used to describe the structured content in a natural language manner; Based on the natural language description information, a data processing request is sent to the target model, and output information returned by the target model is received.

13. The method according to claim 12, characterized in that The natural language description information is generated in the following manner: Based on pre-configured template information, convert the structured content to obtain the natural language description information; or, Generate prompt information to be sent to a target model based on the structured content and the example content, receive return information from the target model, and determine natural language description information corresponding to the structured content based on the return information, wherein the example content represents the association relationship between the structure information of the structured content and the natural language description information.

14. The method according to claim 12, characterized in that The sending a data processing request to the target model based on the natural language description information includes: Based on the natural language description information, the structured content included in the input information is updated, and the updated data is sent to the target model.

15. The method according to claim 12, characterized in that: The structured content includes at least one of the following: Display or store target content in a pre-set structure; Response or operation information to the target content; The target content includes at least one of the following: content information, control information, personnel information, code information, and business entity information.

16. A data processing device, characterized in that: The device comprises: A receiving unit, configured to receive first reply information sent by a target model, wherein the first reply information is determined by the target model according to the first prompt information; a determining unit, configured to determine first display content according to the first reply information, wherein the first display content includes structured content; A sending unit is used to send second prompt information to the target model in response to receiving user input information associated with the first display content, wherein the second prompt information includes the user input information and description information, and the description information includes the structured content described in a natural language.

17. A data processing device, characterized in that: The device comprises: A receiving unit, used for receiving input information; a conversion unit, configured to, in response to the input information including structured content, convert the structured content into natural language description information, wherein the natural language description information is used to describe the structured content in a natural language manner; A sending unit is used to send a data processing request to a target model based on the natural language description information, and receive output information returned by the target model.

18. An electronic device, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 15.

19. A computer readable medium, characterized in that A computer program is stored thereon, wherein when the program is executed by a processor, the method according to any one of claims 1 to 15 is implemented.

Images