WO2026001060A1 - 信息处理方法、装置、设备和存储介质 - Google Patents

信息处理方法、装置、设备和存储介质

Info

Publication number
WO2026001060A1
WO2026001060A1 PCT/CN2025/080095 CN2025080095W WO2026001060A1 WO 2026001060 A1 WO2026001060 A1 WO 2026001060A1 CN 2025080095 W CN2025080095 W CN 2025080095W WO 2026001060 A1 WO2026001060 A1 WO 2026001060A1
Authority
WO
WIPO (PCT)
Prior art keywords
target data
processing
target
data
processing instruction
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/CN2025/080095
Other languages
English (en)
French (fr)
Inventor
黄巍
马进
楼宇泽
黄树炫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Zitiao Network Technology Co Ltd
Original Assignee
Beijing Zitiao Network Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Zitiao Network Technology Co Ltd filed Critical Beijing Zitiao Network Technology Co Ltd
Publication of WO2026001060A1 publication Critical patent/WO2026001060A1/zh
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models
    • G06N5/041—Abduction
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning

Definitions

  • the exemplary embodiments disclosed herein generally relate to the field of computers, and particularly to information processing methods, apparatus, devices, and computer-readable storage media.
  • terminal devices can provide people with a variety of services in work and life. Applications providing these services can be deployed on these terminal devices.
  • the terminal devices present relevant content and interact with users through the application's user interface to meet various user needs. In some cases, users may initiate information processing requests within the application. Therefore, improving the efficiency of information processing is a key concern.
  • an information processing method includes: in response to receiving user input, acquiring target data indicated by the user input; generating at least one processing instruction for the target data using a target model based on source information relating to the source of the target data; acquiring at least one processing result by executing the at least one processing instruction on the target data; and generating an analysis result for the target data as a response to the user input using the target model based on the at least one processing result.
  • an apparatus for information processing comprising: a target data acquisition module configured to acquire target data indicated by the user input in response to receiving user input; a processing instruction generation module configured to generate at least one processing instruction for the target data based on source information relating to the source of the target data and using a target model; a processing result acquisition module configured to acquire at least one processing result by executing at least one processing instruction on the target data; and an analysis result generation module configured to generate an analysis result for the target data as a response to the user input based on at least one processing result and using the target model.
  • an electronic device in a third aspect of this disclosure, includes at least one processor; and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor. When executed by the at least one processor, the instructions cause the electronic device to perform the method of the first aspect.
  • a computer-readable storage medium stores computer-executable instructions that, when executed by a processor, implement the method of the first aspect.
  • a computer program product which is tangibly stored in a computer storage medium and includes computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, implement the method according to a first aspect of this disclosure.
  • Figure 1 shows a schematic diagram of an example environment in which embodiments of the present disclosure can be implemented
  • FIG. 2 shows a flowchart of an information processing procedure according to some embodiments of the present disclosure
  • Figure 3 illustrates a schematic diagram of an example of information processing according to some embodiments of the present disclosure
  • Figure 4 shows a schematic structural block diagram of an apparatus for information processing according to some embodiments of the present disclosure
  • Figure 5 shows a block diagram of an electronic device that can implement one or more embodiments of the present disclosure.
  • performing a step in response to A does not mean that the step is performed immediately after A, but may include one or more intermediate steps.
  • relevant users should be informed of the type, scope of use, and usage scenarios of the information involved in this disclosure through appropriate means in accordance with relevant laws and regulations, and authorization should be obtained from the relevant users.
  • relevant users may include any type of rights holder, such as individuals, enterprises, and groups.
  • a prompt message is sent to the relevant user to clearly inform the user that the requested operation will require obtaining and using the user's information, thereby enabling the relevant user to choose whether to provide information to the software or hardware such as the electronic device, application, server, or storage medium that performs the operation of the technical solution disclosed herein based on the prompt message.
  • a prompt message in response to a user's active request, can be sent to the user, such as a pop-up window, where the prompt message can be presented in text format.
  • the pop-up window can also include a selection control allowing the user to choose "agree” or "disagree” to provide information to the electronic device.
  • model refers to a model that learns the relationship between inputs and outputs from training data, enabling it to generate corresponding outputs for a given input after training. Model generation can be based on machine learning techniques. Deep learning is a machine learning algorithm that processes inputs and provides corresponding outputs using multiple layers of processing units. A neural network model is an example of a deep learning-based model. In this paper, “model” may also be referred to as a “machine learning model,” “learning model,” “machine learning network,” or “learning network,” and these terms are used interchangeably.
  • Figure 1 illustrates a schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented.
  • Environment 100 relates to an application management platform 110, which can support application creation and/or application execution.
  • the portion of the application management platform 110 used to support application creation may also be referred to as an application creation portion.
  • the portion of the application management platform 110 used to support application execution may also be referred to as an application execution portion.
  • the application creation section provides an environment for user 105 to create and publish applications.
  • User 105 can be referred to as the application creation user or creator.
  • the application creation section can be a low-code platform that provides a collection of tools for application creation.
  • the application creation section can support visual development of various types of applications, allowing developers to skip the manual coding process and accelerate the application development cycle and reduce costs.
  • the application creation section can support any suitable platform for users to develop one or more types of applications, such as an application platform as a service (aPaaS) based platform.
  • aPaaS application platform as a service
  • Such a platform enables users to efficiently develop applications, enabling operations such as application creation and application function adjustment.
  • the application creation component can be deployed locally on user 105's terminal device and/or supported by a server-side device.
  • user 105's terminal device can run a client with the application creation component, which can support interaction between the user and the application creation component provided by the server.
  • the application creation component runs locally on the user's terminal device
  • user 105 can directly interact with the local application creation component using the terminal device.
  • the server-side device can provide services to the client running on the terminal device based on the communication connection with the terminal device.
  • the application creation component can present a corresponding page 130 to user 105 based on user 105's actions, to output and/or receive application creation-related information from user 105.
  • the application creation section may be associated with a corresponding database, which stores the data or information required for the application creation process supported by the application creation section.
  • the database may store the code and description information corresponding to the various functional modules that make up the application.
  • the application creation section can also perform operations such as calling, adding, deleting, and updating the functional modules in the database.
  • the database may also store operations that can be performed on different functional blocks. For example, in a scenario where an application needs to be created, the application creation section can call the corresponding functional blocks from the database to build the application.
  • user 105 can create and publish target application 120 as needed in the application creation section.
  • Target application 120 can be published to any suitable application runtime section, as long as the application runtime section can support the operation of target application 120.
  • target application 120 can be operated by one or more end users 145.
  • End user 145 can operate target application 120 through an associated terminal device 146 and thereby interact with application management platform 110.
  • End user 145 can be referred to as the end user of target application 120.
  • target application 120 may include or be implemented as digital assistant 122.
  • Digital assistant 122 can be configured to have intelligent conversational capabilities.
  • digital assistant 122 can be integrated into target application 120, serving as part of target application 120 to assist in task processing within target application 120.
  • digital assistant 122 can be configured as a standalone application, such as a web application or other type of application.
  • digital assistant 122 and target application 120 can be considered as the same application.
  • Digital assistant 122 is provided to assist users with various task processing needs in different applications and scenarios.
  • the user inputs interactive messages, and digital assistant 122 responds to the user's input by providing reply messages.
  • digital assistant 122 can support users inputting questions in natural language and performs tasks and provides replies based on its understanding of natural language input and logical reasoning capabilities.
  • the digital assistant 122 can interact with the end user 145 as a contact.
  • the digital assistant 122 can be implemented in an instant messaging (IM) application.
  • IM instant messaging
  • the digital assistant 122 can interact with the end user 145 in a one-on-one chat session.
  • the digital assistant 122 can interact with multiple users in a group chat session that includes multiple users.
  • the client of the application runtime portion can present an interaction window 142 of the target application 120 or digital assistant 122 in the client interface, such as a conversation window with the digital assistant 122.
  • the end user 145 can enter conversation messages in the conversation window, and the target application 120 can determine the response message from the digital assistant 122 based on the created configuration information and present it to the user in the interaction window 142.
  • the interaction messages with the target application 120 can include multimodal messages, such as text messages (e.g., natural language text), voice messages, image messages, video messages, and so on.
  • the application runtime component can be deployed locally on each end user's (145's) terminal device and/or supported by a server device.
  • the end user's (145's) terminal device can run a client with the application runtime component, which can support interaction between the user and the application runtime component provided by the server.
  • the application runtime component runs locally on the user's terminal device
  • the end user (145) can directly interact with the local application runtime component using the terminal device.
  • the server device can provide services to the client running on the terminal device based on the communication connection with the terminal device.
  • the application runtime component can present corresponding application pages to the end user (145) based on the user's (145's) actions, outputting and/or receiving application-related information from the user (145).
  • the implementation of at least some functions of the target application 120, and/or the implementation of at least some functions of the digital assistant 122 within the target application 120 may be based on models.
  • one or more models 155 may be invoked, such as the capabilities of model 155.
  • the digital assistant 122 may utilize model 155 to understand user input and provide responses to the user based on the output of model 155.
  • the application management platform 110 needs to use model 155 to test the target application 120 to determine whether the running results of the target application 120 meet expectations.
  • the application operation part may need to use model 155 to determine the response results to users.
  • model 155 may run on the application management platform 110 or other remote servers.
  • model 155 may be a machine learning model, a deep learning model, a learning model, a neural network, etc.
  • the model may be based on a language model (LM).
  • LM language model
  • Model 155 may also be based on other suitable models.
  • the application management platform 110 can run on suitable electronic devices. These electronic devices can be any type of computing-capable device, including terminal devices or server devices. Terminal devices can be any type of mobile terminal, fixed terminal, or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, personal communication system (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio/video players, digital cameras/camcorders, positioning devices, television receivers, radio receivers, e-book devices, gaming devices, or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof.
  • Server devices can include, for example, computing systems/servers, such as mainframes, edge computing nodes, computing devices in cloud environments, and so on. In some embodiments, the management platform 110 can be implemented based on cloud services.
  • an improved information processing scheme in this embodiment, in response to receiving user input, target data indicated by the user input is acquired. Based on source information related to the source of the target data, at least one processing instruction for the target data is generated using a target model. By executing the at least one processing instruction on the target data, at least one processing result is obtained respectively. Based on the at least one processing result, an analysis result for the target data is generated using the target model as a response to the user input.
  • models can be used to generate data processing instructions, and the results of these instructions can be used to determine the analytical outcomes.
  • the actual data processing can be handled by the instruction execution engine, rather than executed by the model itself. This allows data processing that the model may not be well-suited for to be removed from the model, thus improving the efficiency and accuracy of information processing.
  • the task management process described in the embodiments of this disclosure can be implemented on an application management platform, a terminal device with the application management platform installed, and/or a server corresponding to the application management platform.
  • the description is from the perspective of the application management platform, such as the application management platform 110 shown in FIG1.
  • the user interface presented by the application management platform 110 can be presented via the terminal device of user 145, and the application management platform 110 can receive user input via the terminal device of user 145.
  • user 145 is the terminal user of the target application 120.
  • the user interface presented by the application management platform 110 can also be presented via the terminal device of user 105, and the application management platform 110 can also receive user input via the terminal device of user 105.
  • user 105 is the creator, manager, or maintainer of the target application 120.
  • Figure 2 shows a flowchart of an information processing procedure 200 according to some embodiments of the present disclosure.
  • Procedure 200 can be implemented in the application management platform 110, for example, by the application runtime portion of the application management platform 110.
  • the task processing procedure shown in Figure 2 will be described below with reference to Figure 1.
  • the application management platform 110 responds to receiving user input by acquiring the target data indicated by the user input.
  • User input can be from any appropriate user, for example, it could be from user 145.
  • User input can be of any appropriate type, such as text, voice, gesture, etc.
  • Application management platform 110 can receive user input via any appropriate means, for example, receiving text input via an input box, receiving audio input via a microphone, etc.
  • User input can be presented, for example, in an interactive window (e.g., interactive window 142).
  • application management platform 110 can process the user input to determine the corresponding text. For example, if the user input is audio, application management platform 110 can convert the corresponding audio into text.
  • Target data may include raw data from a data object.
  • a data object can be, for example, a structured object.
  • a structured object can be any suitable type of object capable of storing or representing information in a structured way, and may include, but is not limited to, data tables, databases, etc.
  • target data may include raw data from a data table.
  • Target data may also include data determined based on the raw data.
  • target data may include data calculated from the raw data.
  • the application management platform 110 can determine the user's input intent and obtain corresponding target data based on that intent. For example, if the user inputs the text "In which month did product A achieve its highest sales volume in the past year?", the application management platform 110 can determine that the user's input intent is to analyze the sales data of product A over the past year, and thus obtain the sales data of product A over the past year from the data table. Based on the sales data of the past year, the sales volume for each month of the past year can be calculated. In this example, the sales volume for each month and the sales data for the past year can be used as target data.
  • obtaining target data from user input can be achieved using a model.
  • a model can generate data query instructions based on user input to query sales figures for each month over the past year.
  • the application management platform 110 generates at least one processing instruction for the target data based on source information related to the source of the target data and using the target model.
  • the processing instruction can be used to perform any type of data processing on the target data, such as, but not limited to, data transformation, data cleaning, sorting, various numerical operations, etc.
  • Source information may include, for example, one or more data query instructions used to retrieve the target data. These data query instructions could be, for example, Structured Query Language (SQL), which tells the target model how the target data was retrieved, helping the target model better understand the data.
  • Source information may also include, for example, metadata about one or more data objects from which the target data originates, such as the structure of a data table. Meta-information about a data object may include, for example, the fields included in the data object.
  • Source information may also include, for example, examples of data records from one or more data objects. Taking a data table as an example, examples of data records may include, for example, the first few rows of the table.
  • the target model can be a model deployed locally on the application management platform 110 or a model deployed on other electronic devices.
  • the target model can be based on any suitable model architecture, including but not limited to Transformer models, Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Deep Neural Networks (DNNs), and any suitable model.
  • the target model can be a language model (LM).
  • the application management platform 110 can obtain address information for the target data.
  • the address information can indicate the storage location of the target data.
  • the application management platform 110 can store the target data in an instruction execution environment before generating at least one processing instruction.
  • the address information obtained by the application management platform 110 can indicate the storage location of the target data in the instruction execution environment.
  • the instruction execution environment can be a restricted runtime environment that allows users to execute data processing instructions, such as code in a predefined language, within an isolated and secure space. Such an instruction execution environment provides a secure, isolated, and controllable environment.
  • a sandbox is an example of an instruction execution environment.
  • Application management platform 110 can, for example, generate prompt input for a target model based on address information, source information, and user input. This prompt input can also be referred to as prompt information.
  • Application management platform 110 can, for example, obtain a prompt template for the target model and determine the prompt information for the target model by filling the prompt template with address information, source information, and user input.
  • application management platform 110 can generate only one prompt and obtain at least one processing instruction by providing this prompt information to the target model.
  • the application management platform 110 can generate a first prompt message for the target model based on address information, source information, and user input. The application management platform 110 can then provide this first prompt message to the target model. Upon receiving the first prompt message, the target model can generate a corresponding model output based on it. This model output can, for example, indicate a first processing instruction among at least one processing instruction. The application management platform 110 can obtain the model output from the target model and determine the first processing instruction among at least one processing instruction based on the model output.
  • prompts can be generated based on the address and source information of the data, and these prompts can then be provided to the target model. This reduces the amount of data the target model needs to process and improves the accuracy of information processing. Furthermore, providing the model with the address of the target data, rather than providing all the target data, can minimize the input load on the model.
  • the application management platform 110 may also generate at least one prompt message and obtain at least one processing instruction by providing the at least one prompt message to the target model, each processing instruction corresponding to one prompt message. Specifically, the application management platform 110 may, for example, utilize the target model to determine at least one task indicated by the user input. If the at least one task includes multiple tasks, the application management platform 110 may determine the corresponding prompt message for each task. The application management platform 110 may then determine the processing instruction corresponding to each task based on the prompt message corresponding to each task.
  • the application management platform 110 executes at least one processing instruction on the target data and obtains at least one processing result.
  • the at least one processing instruction generated by the application management platform 110 is executed in a preset instruction execution environment.
  • the application management platform 110 can send at least one processing instruction to the preset instruction execution environment so that the at least one processing instruction can be executed in that instruction execution environment.
  • the application management platform 110 can obtain at least one processing result of the at least one processing instruction from the instruction execution environment.
  • At least one processing instruction comprises multiple processing instructions, these multiple processing instructions may be independent of each other or may affect each other.
  • application management platform 110 can determine another processing instruction based on the processing result of one processing instruction.
  • Application management platform 110 can, for example, use a target model to determine that the user input indicates at least a first task and a second task. For example, if the user input includes the text "In which month did the highest sales of product A occur in the past year? What is the average monthly sales?", then application management platform 110 can use a target model to determine that the user input indicates task A "determine the month in which the highest sales of product A occurred in the past year” and task B "determine the average monthly sales of product A in the past year”.
  • the application management platform 110 can generate a first prompt message associated with the first task based on address information, source information, and user input.
  • the application management platform 110 can provide the first prompt message to the target model to obtain a first processing instruction from at least one processing instruction.
  • the first processing instruction is used for the first task.
  • the application management platform 110 can obtain the first processing result of the first processing instruction.
  • the first processing instruction could be an instruction to find the maximum value among 12 sales figures, and the first processing result would be the maximum value and its corresponding month.
  • the application management platform 110 can generate a second prompt message associated with the second task based on address information, source information, and the first processing result.
  • the application management platform 110 can provide the second prompt message to the target model to obtain a second processing instruction from at least one processing instruction.
  • the second processing instruction is used for the second task.
  • the second processing instruction could be an instruction to calculate the average of 12 sales figures, and the second processing result would be the average.
  • the application management platform 110 can also execute in response to a given processing instruction in at least one processing instruction, update the target data based on the processing result of the given processing instruction, and generate subsequent processing instructions based on the updated target data. For example, if the processing result of processing instruction A indicates that new data has been generated, the new data will be added to the target data. Subsequent processing instructions after processing instruction A will be generated based on the target data with the added new data. If the processing result of processing instruction B indicates that the target data has been cleaned, the application management platform 110 will remove the cleaned data from the target data. Subsequent processing instructions after processing instruction B will be generated based on the target data with the cleaned data removed.
  • the application management platform 110 based on at least one processing result and utilizing the target model, generates analytical results for the target data as a response to user input.
  • These analytical results may, for example, include various insights related to the user input.
  • the analytical results may, for example, include a summary of the patterns in sales revenue changes for product A over the past year.
  • the analysis results may include, for example, citation information indicating the source of one or more data points in the analysis results.
  • the prompts provided to the model may instruct the model to provide the source of the data upon which the analysis results depend, thereby enhancing the credibility of the insights and providing users with reliable analysis results.
  • the application management platform 110 may receive user input via an interactive window (e.g., interactive window 142) and provide analysis results via the same window.
  • the analysis results may be provided to the user, for example, in the form of a conversational message from a digital assistant.
  • Application management platform 110 may, for example, provide at least one processing result to a target model to generate analysis results for target data using the target model.
  • application management platform 110 may generate prompt word input for the target model based on at least one processing result and provide the generated prompt word input to the target model.
  • Application management platform 110 may, for example, generate a prompt word input based on at least one processing result, which can instruct the target model to analyze at least one processing result.
  • Application management platform 110 may also, for example, generate at least one prompt word input based on at least one processing result, each prompt word input instructing analysis of the corresponding processing result.
  • the application management platform 110 can also determine the input consumption of source information and at least one processing result, and the input capacity of the target model.
  • Input consumption can represent the amount of data input to the target model using source information and at least one processing result.
  • An example of input consumption can be the number of tokens consumed as input to the model; correspondingly, input capacity can be the total number of tokens that can be provided to the model in a single input.
  • the application management platform 110 can compare the input consumption with the input capacity of the target model to determine whether the input consumption exceeds the input capacity of the target model. If the input consumption does not exceed the input capacity, the application management platform 110 can directly use the target model to process at least one processing result to determine the analysis result for the target data. If the input consumption exceeds the input capacity, the application management platform 110 can generate a third prompt based on at least one processing result to instruct the target model to summarize at least one processing result. The application management platform 110 can provide the third prompt to the target model to obtain the analysis result.
  • the application management platform 110 can determine that the input consumption does not exceed the input capacity, and then directly use the target model to process at least one processing result to determine the analysis result for the target data. If the input consumption is 60, the application management platform 110 can determine that the input consumption exceeds the input capacity, and then can generate a third prompt message based on at least one processing result to instruct the target model to summarize at least one processing result. The application management platform 110 can provide the third prompt message to the target model to obtain the analysis result.
  • 50 e.g., the number of tokens
  • the application management platform 110 can prompt the target model to summarize based on the current existing results, and can provide users with at least some analysis results, thereby improving the user experience.
  • Figure 3 illustrates a schematic diagram of an example 300 of information processing according to some embodiments of the present disclosure.
  • the application management platform 110 can obtain target data 304 indicated by user input 340 in response to receiving user input 340.
  • target data 304 can be the sales data of product A over the past year.
  • the application management platform 110 can generate at least one processing instruction for target data 304 based on source information 306 related to the source of target data 301, using a target model (e.g., model 330).
  • a target model e.g., model 330
  • the processing instruction can be used to perform any type of data processing on the target data, such as, but not limited to, data transformation, data cleaning, sorting, various numerical operations, etc.
  • the processing instruction can be used to calculate the sales revenue and/or sales volume of product A for each month of the past 12 months, etc.
  • the application management platform 110 can generate prompt information for the model 330 based at least on the source information 306, and provide the prompt information to the model 330.
  • the model 330 can output at least one processing instruction for the target data 304.
  • the model 330 can be a local model of the application management platform 110 or a model deployed on a remote device. If the model 330 is deployed locally on the application management platform 110, the application management platform 110 can directly use the model 330 to determine at least one processing instruction. If the model 330 is deployed on a remote device, the application management platform 110 can invoke the model 330 deployed on the remote device via a communication connection with the remote device to determine at least one processing instruction.
  • Source information 306 may include one or more data query instructions for obtaining target data, metadata 302 of one or more data objects from which the target data originates, and/or, data record examples 303 in one or more data objects.
  • source information 306 may include one or more query instructions for obtaining sales data of product A over the past year.
  • Application management platform 110 may, for example, upload target data 304 to instruction execution environment 320 in advance before generating at least one processing instruction, so as to store target data 304 in instruction execution environment 320.
  • Application management platform 110 may obtain address information for target data 304, indicating the storage location of target data 304 in instruction execution environment 320.
  • Application management platform 110 may generate prompt information 310 (e.g., first prompt information) for the target model based on address information, source information 306, and user input 340.
  • prompt information 310 e.g., first prompt information
  • the application management platform 110 may, for example, execute the multiple steps included in block 350 to generate at least one processing instruction.
  • the application management platform 110 may determine multiple tasks indicated by user input 340.
  • the application management platform 110 may provide prompts generated based on user input 340 to a target model, which will then parse these tasks.
  • the application management platform 110 may generate multiple prompts based on multiple tasks.
  • the application management platform 110 may provide multiple prompts to the target model to determine multiple processing instructions corresponding to the multiple tasks. It should be noted that at least one of the multiple steps included in block 350 may be determined by the application management platform 110 using model 330.
  • the application management platform 110 may directly use model 330 or invoke model 330 from a remote device to execute this at least one step.
  • the application management platform 110 can provide the prompt information generated based on user input 340 to the target model, which then parses out these tasks.
  • the application management platform 110 can generate corresponding prompt information for each of these tasks.
  • the generated prompt information can be provided to the target model, which then generates processing instructions corresponding to the tasks.
  • the application management platform 110 can obtain at least one processing result by executing at least one processing instruction on the target data 304.
  • the processing result could be the calculated monthly sales volume and/or sales revenue of product A over the past year.
  • the application management platform 110 can execute the processing instructions on the target data 304 multiple times to obtain multiple processing results.
  • the application management platform 110 can provide at least one processing instruction to the instruction execution environment 320 and obtain at least one processing result corresponding to at least one processing instruction from the instruction execution environment 320.
  • the application management platform 110 can provide at least one processing result to the target model (similarly, for example, model 330) to generate analysis results for the target data 304 as a response to user input 340.
  • the monthly sales volume and/or sales revenue of product A can be provided to the target model.
  • the target model can output analysis results on sales volume and/or sales revenue, such as trends in sales volume and/or sales revenue, months with the highest sales volume and/or sales revenue, months with the lowest sales volume and/or sales revenue, etc.
  • the application management platform 110 can determine whether the input consumption 301 of the source information 306 and at least one processing result exceeds the input capacity 305 of the target model.
  • the application management platform 110 can generate a prompt message based on at least one processing result instructing the target model to summarize the at least one processing result.
  • the application management platform 110 can provide this prompt message to the target model to obtain the analysis results.
  • the application management platform 110 can directly provide at least one processing result to the target model in response to determining that the input consumption 301 does not exceed the input capacity 305, so as to obtain the analysis result.
  • a model can be used to generate data processing instructions, and the model can be used to determine the data analysis results based on the processing results of the data instructions. This helps to improve the efficiency and accuracy of information processing. Furthermore, it enhances the model's ability to process complex and large-volume information.
  • FIG. 4 shows a schematic structural block diagram of an apparatus 400 for information processing according to some embodiments of this disclosure.
  • the apparatus 400 may be implemented in or included in an application management platform 110, for example.
  • the various modules/components in the apparatus 400 may be implemented by hardware, software, firmware, or any combination thereof.
  • device 400 includes a target data acquisition module 410, configured to acquire target data indicated by user input in response to receiving user input.
  • Device 400 also includes a processing instruction generation module 420, configured to generate at least one processing instruction for the target data based on source information related to the source of the target data and using a target model.
  • Device 400 further includes a processing result acquisition module 430, configured to acquire at least one processing result by executing at least one processing instruction on the target data.
  • Device 400 also includes an analysis result generation module 440, configured to generate an analysis result for the target data as a response to user input based on at least one processing result and using a target model.
  • the processing instruction generation module 420 includes: an address information acquisition module configured to acquire address information for target data, the address information indicating the storage location of the target data; a first prompt information generation module configured to generate a first prompt information for the target model based on the address information, source information, and user input; and a first processing instruction acquisition module configured to provide the first prompt information to the target model to obtain a first processing instruction from at least one processing instruction.
  • At least one processing instruction is executed in a preset instruction execution environment
  • the apparatus 400 further includes a target data storage module configured to store target data in the instruction execution environment before generating at least one processing instruction, wherein address information indicates the storage location of the target data in the instruction execution environment.
  • the apparatus 400 further includes: a task determination module configured to determine, using a target model, that user input at least indicates a first task and a second task, wherein a first processing instruction is associated with the first task, and the processing instruction generation module 420 further includes: a second prompt information generation module configured to generate a second prompt information for the target model based on address information, source information, and a first processing result of the first processing instruction; and a second processing instruction acquisition module configured to provide the second prompt information to the target model to obtain a second processing instruction from at least one processing instruction for use in the second task.
  • a task determination module configured to determine, using a target model, that user input at least indicates a first task and a second task, wherein a first processing instruction is associated with the first task
  • the processing instruction generation module 420 further includes: a second prompt information generation module configured to generate a second prompt information for the target model based on address information, source information, and a first processing result of the first processing instruction; and a second processing instruction acquisition module configured to provide the second prompt information
  • the source information includes at least one of the following: one or more data query instructions for obtaining target data, metadata of one or more data objects from which the target data originates, or examples of data records in one or more data objects.
  • the analysis result generation module 440 includes: a capacity determination module configured to determine whether the input consumption of source information and at least one processing result exceeds the input capacity of the target model, wherein the input consumption represents the amount of data that takes source information and at least one processing result as input to the target model; a third prompt information generation module configured to generate a third prompt information based on at least one processing result in response to determining that the input consumption exceeds the input capacity, so as to instruct the target model to summarize at least one processing result; and an analysis result acquisition module configured to provide the third prompt information to the target model to obtain the analysis result.
  • a capacity determination module configured to determine whether the input consumption of source information and at least one processing result exceeds the input capacity of the target model, wherein the input consumption represents the amount of data that takes source information and at least one processing result as input to the target model
  • a third prompt information generation module configured to generate a third prompt information based on at least one processing result in response to determining that the input consumption exceeds the input capacity, so as to instruct the target model to summarize at least one
  • the apparatus 400 further includes a target data update module, configured to complete execution in response to a given processing instruction in at least one processing instruction, update target data based on the processing result of the given processing instruction, and generate processing instructions after the given processing instruction based on the updated target data.
  • a target data update module configured to complete execution in response to a given processing instruction in at least one processing instruction, update target data based on the processing result of the given processing instruction, and generate processing instructions after the given processing instruction based on the updated target data.
  • the analysis results include citation information that indicates the source of one or more data points in the analysis results.
  • the units and/or modules included in device 400 can be implemented in various ways, including software, hardware, firmware, or any combination thereof.
  • one or more units and/or modules can be implemented using software and/or firmware, such as machine-executable instructions stored on a storage medium.
  • some or all of the units and/or modules in device 400 can be implemented at least partially by one or more hardware logic components.
  • exemplary types of hardware logic components 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 so on.
  • Figure 5 shows a block diagram of an electronic device 500 in which one or more embodiments of the present disclosure may be implemented. It should be understood that the electronic device 500 shown in Figure 5 is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein.
  • the electronic device 500 shown in Figure 5 may include or be implemented as the application management platform 110 of Figure 1, or the device 400 of Figure 4.
  • electronic device 500 is in the form of a general-purpose electronic device.
  • Components of electronic device 500 may include, but are not limited to, one or more processors 510 or processing units, memory 520, storage device 530, one or more communication units 540, one or more input devices 550, and one or more output devices 560.
  • Processor 510 may be a physical or virtual processor and is capable of performing various processes according to programs stored in memory 520. In a multiprocessor system, multiple processors execute computer-executable instructions in parallel to improve the parallel processing capability of electronic device 500.
  • Electronic device 500 typically includes multiple computer storage media. Such media can be any accessible media that is accessible to electronic device 500, including but not limited to volatile and non-volatile media, removable and non-removable media.
  • Memory 520 can be volatile memory (e.g., registers, cache, random access memory (RAM)), non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof.
  • Storage device 530 can be removable or non-removable media and can include machine-readable media, such as flash drives, disks, or any other media that can be used to store information and/or data and can be accessed within electronic device 500.
  • Electronic device 500 may further include additional removable/non-removable, volatile/non-volatile storage media.
  • disk drives for reading from or writing to removable, non-volatile disks (e.g., "floppy disks") and optical disk drives for reading from or writing to removable, non-volatile optical disks may be provided.
  • each drive may be connected to a bus (not shown) via one or more data media interfaces.
  • Memory 520 may include computer program product 525 having one or more program modules configured to perform various methods or actions of various embodiments of the present disclosure.
  • Communication unit 540 enables communication with other electronic devices via a communication medium. Additionally, the functionality of components of electronic device 500 can be implemented using a single computing cluster or multiple computing machines capable of communicating via communication connections. Therefore, electronic device 500 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or another network node.
  • PCs network personal computers
  • Input device 550 can be one or more input devices, such as a mouse, keyboard, trackball, etc.
  • Output device 560 can be one or more output devices, such as a monitor, speaker, printer, etc.
  • Electronic device 500 can also communicate with one or more external devices (not shown) via communication unit 540 as needed. These external devices include storage devices, display devices, etc., and can communicate with one or more devices that enable user interaction with electronic device 500, or with any device that enables electronic device 500 to communicate with one or more other electronic devices (e.g., network card, modem, etc.). Such communication can be performed via input/output (I/O) interface (not shown).
  • I/O input/output
  • a computer-readable storage medium that stores computer-executable instructions thereon, wherein the computer-executable instructions are executed by a processor to implement the methods described above.
  • a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, which are executed by a processor to implement the methods described above.
  • These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions/actions specified in one or more blocks of the flowchart and/or block diagram.
  • These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and/or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions/actions specified in one or more blocks of the flowchart and/or block diagram.
  • Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions/actions specified in one or more boxes of a flowchart and/or block diagram.
  • each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing the specified logical function.
  • the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved.
  • each block in the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computational Linguistics (AREA)
  • Stored Programmes (AREA)

Abstract

提供了信息处理方法、装置、设备和存储介质。方法包括：响应于接收到用户输入，获取用户输入所指示的目标数据（210）；基于与目标数据的来源有关的来源信息，利用目标模型，生成针对目标数据的至少一个处理指令（220）；通过对目标数据执行至少一个处理指令，分别获取至少一个处理结果（230）；以及基于至少一个处理结果，利用目标模型，生成针对目标数据的分析结果作为针对用户输入的响应（240）。

Description

信息处理方法、装置、设备和存储介质
本申请要求2024年6月27日递交的、标题为“信息处理方法、装置、设备和存储介质”、申请号为202410851923.1的中国发明专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本公开的示例实施例总体涉及计算机领域，特别地涉及信息处理方法、装置、设备和计算机可读存储介质。
背景技术
随着信息技术的发展，各种终端设备可以在工作和生活等方面向人们提供各种服务。终端设备中可以部署有提供服务的应用。终端设备通过应用的用户界面呈现相应的内容并实现与用户之间的交互，满足用户的各类需求。在一些情况下，用户可能在应用内发起信息处理请求。因此，如何提高信息处理的效率是所关注的问题。
发明内容
在本公开的第一方面，提供了一种信息处理方法。该方法包括：响应于接收到用户输入，获取用户输入所指示的目标数据；基于与目标数据的来源有关的来源信息，利用目标模型，生成针对目标数据的至少一个处理指令；通过对目标数据执行至少一个处理指令，分别获取至少一个处理结果；以及基于至少一个处理结果，利用目标模型，生成针对目标数据的分析结果作为针对用户输入的响应。
在本公开的第二方面，提供了一种用于信息处理的装置，包括：目标数据获取模块，被配置为响应于接收到用户输入，获取用户输入所指示的目标数据；处理指令生成模块，被配置为基于与目标数据的来源有关的来源信息，利用目标模型，生成针对目标数据的至少一个处理指令；处理结果获取模块，被配置为通过对目标数据执行至少一个处理指令，分别获取至少一个处理结果；以及分析结果生成模块，被配置为基于至少一个处理结果，利用目标模型，生成针对目标数据的分析结果作为针对用户输入的响应。
在本公开的第三方面，提供了一种电子设备。该设备包括至少一个处理器；以及至少一个存储器，至少一个存储器被耦合到至少一个处理器并且存储用于由至少一个处理器执行的指令。指令在由至少一个处理器执行时使电子设备执行第一方面的方法。
在本公开的第四方面，提供了一种计算机可读存储介质。介质上存储有计算机可执行指令，计算机可执行指令被处理器执行时实现第一方面的方法。
根据本公开的第五方面，提供了一种计算机程序产品，计算机程序产品被有形地存储在计算机存储介质中并且包括计算机可执行指令，其中计算机可执行指令被处理器执行时实现根据本公开的第一方面的方法。
应当理解，该部分中所描述的内容并非旨在限定本公开的实施例的关键特征或重要特征，也不用于限制本公开的范围。本公开的其他特征将通过以下的描述而变得容易理解。
附图说明
结合附图并参考以下详细说明，本公开各实施例的上述和其他特征、优点及方面将变得更加明显。在附图中，相同或相似的附图标记表示相同或相似的元素，其中：
图1示出了能够在其中实现本公开的实施例的示例环境的示意图；
图2示出了根据本公开的一些实施例的信息处理的过程的流程图；
图3示出了根据本公开的一些实施例的信息处理的示例的示意图；
图4示出了根据本公开的一些实施例的用于信息处理的装置的示意性结构框图；
图5示出了可以实现本公开的一个或多个实施例的电子设备的框图。
具体实施方式
下面将参照附图更详细地描述本公开的实施例。虽然附图中示出了本公开的某些实施例，然而应当理解的是，本公开可以通过各种形式来实现，而且不应该被解释为限于这里阐述的实施例，相反，提供这些实施例是为了更加透彻和完整地理解本公开。应当理解的是，本公开的附图及实施例仅用于示例性作用，并非用于限制本公开的保护范围。
在本公开的实施例的描述中，术语“包括”及其类似用语应当理解为开放性包含，即“包括但不限于”。术语“基于”应当理解为“至少部分地基于”。术语“一个实施例”或“该实施例”应当理解为“至少一个实施例”。术语“一些实施例”应当理解为“至少一些实施例”。下文还可能包括其他明确的和隐含的定义。
在本文中，除非明确说明，“响应于A”执行一个步骤并不意味着在“A”之后立即执行该步骤，而是可以包括一个或多个中间步骤。
可以理解的是，本技术方案所涉及的数据(包括但不限于数据本身、数据的获得、使用、存储或删除)应当遵循相应法律法规及相关规定的要求。
可以理解的是，在使用本公开各实施例公开的技术方案之前，均应当根据相关法律法规通过适当的方式对本公开所涉及信息的类型、使用范围、使用场景等告知相关用户并获得相关用户的授权，其中，相关用户可以包括任何类型的权利主体，例如个人、企业、团体。
例如，在响应于接收到用户的主动请求时，向相关用户发送提示信息，以明确地提示相关用户，其请求执行的操作将需要获得和使用到相关用户的信息，从而使得相关用户可以根据提示信息来自主地选择是否向执行本公开技术方案的操作的电子设备、应用程序、服务器或存储介质等软件或硬件提供信息。
作为一种可选的但非限制性的实现方式，响应于接收到相关用户的主动请求，向相关用户发送提示信息的方式，例如可以是弹窗的方式，弹窗中可以以文字的方式呈现提示信息。此外，弹窗中还可以承载供用户选择“同意”或“不同意”向电子设备提供信息的选择控件。
可以理解的是，上述通知和获得用户授权过程仅是示意性的，不对本公开的实现方式构成限定，其他满足相关法律法规的方式也可应用于本公开的实现方式中。本公开实施例数字助手相关功能的启用、获取的数据、数据的处理和存储方式等，均应获得所述用户、以及与所述用户关联的其他权利主体的提前授权，并应符合相关法律法规、权利主体间协议规则的约定。
如本文中所使用的，术语“模型”可以从训练数据中学习到相应的输入与输出之间的关联关系，从而在训练完成后可以针对给定的输入，生成对应的输出。模型的生成可以基于机器学习技术。深度学习是一种机器学习算法，通过使用多层处理单元来处理输入和提供相应输出。神经网络模型是基于深度学习的模型的一个示例。在本文中，“模型”也可以被称为“机器学习模型”、“学习模型”、“机器学习网络”或“学习网络”，这些术语在本文中可互换地使用。
图1示出了本公开的实施例能够在其中实现的示例环境100的示意图。环境100涉及应用管理平台110，其可以支持应用的创建和/或应用的运行。在一些实施例中，应用管理平台110中用于支持应用创建的部分也可以称为应用创建部分。在一些实施例中，应用管理平台110中用于支持应用运行的部分也可以称为应用运行部分。
如图所示，应用创建部分可以为用户105提供应用的创建和发布环境。用户105可以被称为应用创建用户、创建者。在一些实施例中，应用创建部分可以是低代码平台，其提供应用创建的工具合集。应用创建部分可以支持对各类应用的可视化开发，从而使开发人员可以跳过手工编码的过程，加快应用的开发周期和成本。应用创建部分可以支持用户开发一个或多个类型的应用的任何适当平台，例如可以包括基于应用平台即服务(aPaaS)的平台。这样的平台能够支持用户对应用进行高效开发，实现应用创建、应用功能调整等操作。
应用创建部分可以被部署在用户105的终端设备本地，和/或可以由服务端设备来支持。例如，用户105的终端设备可以运行有应用创建部分的客户端，该客户端可以支持用户与服务端所提供的应用创建部分的交互。在应用创建部分运行在用户的终端设备本地的情况下，用户105可以直接利用终端设备与本地的应用创建部分进行交互。在应用创建部分运行在服务端设备的情况下，服务端设备可以基于与终端设备之间的通信连接，实现对终端设备中运行的客户端的服务供应。应用创建部分可以基于用户105的操作而向用户105呈现相应的页面130，以向用户105输出和/或从用户105接收与应用创建相关的信息。
在一些实施例中，应用创建部分可以关联到相应的数据库，其中存储应用创建部分所支持的应用创建过程所需的数据或信息。例如，数据库可以存储用于组成应用的各个功能模块对应的代码和描述信息等。应用创建部分还可以对数据库中的功能模块执行调用、增加、删除、更新等操作。数据库还可以存储对不同功能块可执行的操作。示例性的，在要创建应用的场景，应用创建部分可以从数据库中调用相应的功能块来搭建应用。
在本公开的实施例中，用户105可以在应用创建部分上根据需要创建目标应用120，并发布目标应用120。目标应用120可被发布到任一适当的应用运行部分，只要应用运行部分能够支持目标应用120的运行。在发布后，目标应用120可以用于由一个或多个终端用户145进行操作。终端用户145可以通过相关联的终端设备146来操作目标应用120，并进而与应用管理平台110进行交互。终端用户145可以被称目标应用120的终端用户。在一些实施例中，目标应用120可以包括或者被实现为数字助手122。
数字助手122可以被配置为具有智能对话的能力。在图所示的示例中，数字助手122可以被集成在目标应用120内，作为目标应用120的一部分来协助执行目标应用120内的任务处理。在其他示例中，数字助手122可以被配置为独立运行的应用，例如网页应用或其他类型的应用。在这样的示例中，数字助手122与目标应用120可以被视为同一应用。数字助手122被提供用于协助用户在不同应用、场景下的各类任务处理需求。在与数字助手122的交互过程中，用户输入交互消息，并且数字助手122响应于用户输入来提供回复消息。通常，数字助手122能够支持用户以自然语言方式输入提问，并基于对自然语言输入的理解和逻辑推理能力来执行任务和提供回复。
在一些实施例中，数字助手122可以作为终端用户145的联系人与之交互。例如，数字助手122可以被实现在即时消息(IM)应用中。数字助手122可以在与终端用户145的单聊会话中与终端用户145交互。在一些实施例中，数字助手122可以在包括多个用户的群聊会话中与多个用户交互。
对于每个终端用户145，应用运行部分的客户端可以在客户端界面中呈现目标应用120或数字助手122的交互窗口142，例如与数字助手122的会话窗口。终端用户145可以在会话窗口中输入会话消息，目标应用120可以基于所创建的配置信息，来确定数字助手122的回复消息并在交互窗口142中呈现给用户。在一些实施例中，取决于目标应用120的配置，与目标应用120的交互消息可以包括多模态形式的消息，诸如文本消息(例如，自然语言文本)、语音消息、图像消息、视频消息，等等。
与应用创建部分类似，应用运行部分可以被部署在每个终端用户145的终端设备本地，和/或可以由服务端设备来支持。例如，终端用户145的终端设备可以运行有应用运行部分的客户端，该客户端可以支持用户与服务端所提供的应用运行部分的交互。在应用运行部分运行在用户的终端设备本地的情况下，终端用户145可以直接利用终端设备与本地的应用运行部分进行交互。在应用运行部分运行在服务端设备的情况下，服务端设备可以基于与终端设备之间的通信连接，实现对终端设备中运行的客户端的服务供应。应用运行部分可以基于终端用户145的操作而向终端用户145呈现相应的应用页面，以向终端用户145输出和/或从终端用户145接收与应用使用相关的信息。
在一些实施例中，目标应用120的至少部分功能的实现，和/或目标应用120中的数字助手122的至少部分功能的实现可以基于模型来实现。在目标应用120的创建或运行过程中，可以调用一个或多个模型155，例如模型155的能力。在目标应用120中，数字助手122可以利用模型155来理解用户输入，并基于模型155的输出来提供对用户的回复。
在创建过程中，应用管理平台110对目标应用120的测试需要利用模型155来确定目标应用120的运行结果符合预期。在运行过程中，响应于目标应用120的用户的不同操作请求，应用运行部分可能需要利用模型155来确定对用户的响应结果。
虽然被示出为独立于应用管理平台110，但一个或多个模型155可以运行在应用管理平台110上，或者其他远端服务器。在一些实施例中，模型155可以是机器学习模型、深度学习模型、学习模型、神经网络等。在一些实施例中，模型可以基于语言模型(LM)。语言模型通过从大量语料中学习，能够具备问答能力。模型155也可以基于其他适当的模型。
应用管理平台110可以运行在适当的电子设备。这里的电子设备可以是任何类型的具有计算能力的设备，包括终端设备或服务端设备。终端设备可以是任何类型的移动终端、固定终端或便携式终端，包括移动手机、台式计算机、膝上型计算机、笔记本计算机、上网本计算机、平板计算机、媒体计算机、多媒体平板、个人通信系统(PCS)设备、个人导航设备、个人数字助理(PDA)、音频/视频播放器、数码相机/摄像机、定位设备、电视接收器、无线电广播接收器、电子书设备、游戏设备或者前述各项的任何组合，包括这些设备的配件和外设或者其任何组合。服务端设备例如可以包括计算系统/服务器，诸如大型机、边缘计算节点、云环境中的计算设备，等等。在一些实施例中，管理平台110可以基于云服务来实现。
应当理解，仅出于示例性的目的描述环境100的结构和功能，而不暗示对于本公开的范围的任何限制。例如，虽然图示出了与应用创建部分交互的单个用户和与应用运行部分交互的单个用户，但实际上可以多个用户访问应用管理平台110来各自创建数字助手，并且每个数字助手可以被用于与多个用户进行交互。
如前文所提及的，用户可以在应用内发起任务处理请求以在应用内处理信息。传统上，应用可以借助分析框架来实现数据分析，例如提供数据洞察。然而，在数据量较大时，应用难以方便快捷的确定较为准确的数据分析结果。为提高数据分析的准确性，往往还可以借助人工来进行数据分析，这会增大数据分析的人力成本。人工进行数据分析还会进一步降低数据分析效率。
有鉴于此，在本公开实施例中，提供了一种信息处理的改进方案。在该方案中，响应于接收到用户输入，获取用户输入所指示的目标数据。基于与目标数据的来源有关的来源信息，利用目标模型，生成针对目标数据的至少一个处理指令。通过对目标数据执行至少一个处理指令，分别获取至少一个处理结果。基于至少一个处理结果，利用目标模型，生成针对目标数据的分析结果作为针对用户输入的响应。
通过此种方式，可以借助模型来生成对数据的处理指令，并可以借助模型来基于数据指令的处理结果来确定数据的分析结果。具体的数据处理可以不由模型来执行，而是可以由指令执行引擎来完成。由此，可以将模型可能不擅长的数据处理从模型剥离。这有助于提高信息处理的效率和准确性。
以下将参考附图的示例来详细描述本公开的一些示例实施例。
本公开的实施例所描述的任务管理过程可以被实现在应用管理平台，安装有应用管理平台的终端设备和/或应用管理平台对应的服务端。在下文的示例中，出于讨论的需要，从应用管理平台的角度来描述，例如图1所示的应用管理平台110。应用管理平台110所呈现的用户界面，可以经由用户145的终端设备被呈现，并且应用管理平台110可以经由用户145的终端设备来接收用户输入。在本公开的一些实施例中，用户145为目标应用120的终端用户。应当理解，应用管理平台110所呈现的用户界面，也可以经由用户105的终端设备被呈现，并且应用管理平台110也可以经由用户105的终端设备来接收用户输入。在本公开的一些实施例中，用户105为目标应用120的创建者、管理者或维护者。
图2示出了根据本公开的一些实施例的信息处理的过程200的流程图。过程200可以在应用管理平台110被实施，例如，可以由应用管理平台110的应用运行部分实施。以下结合图1来描述图2示出的任务处理过程。
在框210，应用管理平台110响应于接收到用户输入，获取用户输入所指示的目标数据。
用户输入可以是来自任意适当用户的输入，例如，其可以是来自用户145的用户输入。用户输入可以是任意适当类型的用户输入，例如可以为文本类型、语音类型、姿态类型等等。应用管理平台110可以经由任意适当方式接收用户输入，例如，可以经由输入框接收文本类型的用户输入、经由麦克风接收音频类型的用户输入等等。用户输入例如可以被呈现在交互窗口(例如交互窗口142)中。在一些实施例中，在用户输入为非文本类型的用户输入的情况下，应用管理平台110可以对用户输入进行处理以确定该用户输入对应的文本。例如，若用户输入为音频类型的用户输入，应用管理平台110可以将用户输入对应的音频转换为文本。
目标数据可以包括数据对象中的原始数据。数据对象例如可以为结构化对象。结构化对象可以是任意适当类型的能够结构化地存储或表示信息的对象，其可以包括但不限于数据表、数据库等等。例如，目标数据可以包括数据表中的原始数据。目标数据还可以包括基于原始数据而确定的数据。例如，目标数据可以包括由原始数据计算得到的数据。
在一些实施例中，应用管理平台110可以确定用户输入的意图，并基于该意图来获取对应的目标数据。例如，若用户输入为文本“商品A过去一年的最高销售额出现在哪个月份？”，则应用管理平台110可以确定用户输入的意图为对商品A过去一年的销售数据进行分析，进而可以从数据表中获取商品A过去一年的销售数据。基于过去一年的销售数据，可以计算过去一年每个月的销售额。在该示例中，每个月的销售额和过去一年的销售数据可以作为目标数据。
在一些实施例中，从用户输入得到目标数据可以利用模型来实现。例如，模型可以根据用户输入生成数据查询指令，以查询过去一年每个月的销售额。
在框220，应用管理平台110基于与目标数据的来源有关的来源信息，利用目标模型，生成针对目标数据的至少一个处理指令。处理指令可以用于对目标数据执行任何类型的数据处理，例如但不限于数据转换、数据清洗、排序、各种数值运算等等。
来源信息例如可以包括用于获取目标数据的一个或多个数据查询指令。这里的数据查询指令例如可以为结构化查询语言(SQL)，其可以告知目标模型目标数据是如何查询到的，可以帮助目标模型更好地理解数据。来源信息例如还可以包括目标数据所源自的一个或多个数据对象的元信息，例如，数据表的结构信息。数据对象的元信息例如可以包括数据对象所包括的字段。来源信息例如还可以包括一个或多个数据对象中的数据记录示例。以数据对象为数据表为例，数据记录示例例如可以包括数据表的前几行。
目标模型可以是部署在应用管理平台110本地的模型，也可以是部署在其他电子设备处的模型。目标模型可以基于任何适当的模型结构，包括但不限于Transformer模型、卷积神经网络(CNN)、循环神经网络(RNN)、深度神经网络(DNN)等任意适当的模型。在一些实施例中，目标模型可以为语言模型(LM)。
在一些实施例中，应用管理平台110可以获取针对目标数据的地址信息。地址信息可以指示目标数据的存储位置。在一些实施例中，应用管理平台110可以在生成至少一个处理指令之前，将目标数据存储到指令执行环境。在这种情况下，应用管理平台110获取到的地址信息可以指示目标数据在指令执行环境中的存储位置。作为示例，指令执行环境可以是一种受限的运行环境，它允许用户在隔离的安全空间内执行数据处理指令，例如预定语言的代码。这种指令执行环境提供了一个安全、隔离且可控制的环境。沙箱是指令执行环境的一个示例。
应用管理平台110例如可以基于地址信息、来源信息和用户输入，生成针对目标模型的提示词(prompt)输入，提示词输入也可以被称之为提示信息。应用管理平台110例如可以获取针对目标模型的提示词模板，并通过将地址信息、来源信息和用户输入填充至提示词模板来确定针对目标模型的提示信息。在一些实施例中，应用管理平台110可以仅生成一个提示信息，并通过将该提示信息提供给目标模型来获取至少一个处理指令。
例如，应用管理平台110可以基于地址信息、来源信息和用户输入，生成针对目标模型的第一提示信息。应用管理平台110可以将第一提示信息提供给目标模型。目标模型接收到第一提示信息后，可以基于第一提示信息，生成对应的模型输出。模型输出例如可以指示至少一个处理指令中的第一处理指令。应用管理平台110可以从目标模型处获取模型输出，并基于模型输出来确定至少一个处理指令中的第一处理指令。
由此，可以基于数据的地址信息和来源信息来生成提示信息，进而可以将这样的提示信息提供给目标模型，可以降低目标模型处理的数据量，提高信息处理的准确性。此外，将目标数据的地址提供给模型，而不是将所有目标数据提供给模型，可以尽可能避免对模型的输入超过输入容量。
在一些实施例中，应用管理平台110还可以生成至少一个提示信息，并通过将该至少一个提示信息提供给目标模型来获取至少一个处理指令，每个处理指令对应于一个提示信息。具体地，应用管理平台110可以例如可以利用目标模型，确定用户输入所指示的至少一个任务。若至少一个任务包括多个任务，应用管理平台110可以针对每个任务，确定对应的提示信息。应用管理平台110进而可以基于每个任务对应的提示信息，确定每个任务对应的处理指令。
在框230，应用管理平台110通过对目标数据执行至少一个处理指令，分别获取至少一个处理结果。在一些实施例中，若目标数据被存储到指令执行环境，应用管理平台110所生成的至少一个处理指令是在预设的指令执行环境中被执行的。应用管理平台110例如可以将至少一个处理指令发送至预设的指令执行环境，以便在该指令执行环境中执行至少一个处理指令。应用管理平台110可以从该指令执行环境处获取至少一个处理指令的至少一个处理结果。
如果至少一个处理指令包括多个处理指令，这多个处理指令之间可以互不影响，也可以互相影响。例如，应用管理平台110可以基于一个处理指令的处理结果来确定另一个处理指令。应用管理平台110例如可以利用目标模型，确定用户输入至少指示第一任务和第二任务。例如，若用户输入包括文本“商品A过去一年的最高销售额出现在哪个月份？月平均销售额是多少？”，则应用管理平台110可以利用目标模型，确定该用户输入指示任务A“确定商品A过去一年的最高销售额出现的月份”和任务B“确定商品A过去一年的月平均销售额”。
应用管理平台110可以基于地址信息、来源信息和用户输入，生成与第一任务相关联的第一提示信息。应用管理平台110可以将第一提示信息提供给目标模型，以得到至少一个处理指令中的第一处理指令。第一处理指令用于第一任务。应用管理平台110可以获取第一处理指令的第一处理结果。例如，第一处理指令可以是在12个销售额中求最大值的指令，第一处理结果就是最大值及其对应的月份。
应用管理平台110可以基于地址信息、来源信息和第一处理结果，生成与第二任务相关联的第二提示信息。应用管理平台110可以将第二提示信息提供给目标模型，以得到至少一个处理指令中的第二处理指令。第二处理指令用于第二任务。例如，第二处理指令可以是计算12个销售额的平均值的指令，第二处理结果就是平均值。
在一些实施例中，应用管理平台110还可以响应于至少一个处理指令中的给定处理指令完成执行，基于给定处理指令的处理结果更新目标数据，并且在给定处理指令之后的处理指令的生成是基于经更新的目标数据。示例性地，若处理指令A的处理结果指示产生了新的数据，该新的数据将被添加至目标数据。处理指令A之后的处理指令将基于添加有新的数据的目标数据来生成。若处理指令B的处理结果指示对目标数据进行了清洗，应用管理平台110将从目标数据中移除被清洗的数据。处理指令B之后的处理指令将基于移除被清洗的数据的目标数据来生成。
在框240，应用管理平台110基于至少一个处理结果，利用目标模型，生成针对目标数据的分析结果作为针对用户输入的响应。分析结果例如可以包括与用户输入有关的各种洞察分析。继续上文的示例，分析结果例如可以包括对商品A过去一年销售额变化规律的总结。
在一些实施例中，分析结果例如可以包括引用信息，引用信息指示分析结果中的一项或多项数据的来源。在这种实施例中，可以在提供给模型的提示中，指示模型给出分析结果所依赖的数据的来源，由此有利于加强洞察的可信度，从而给用户提供可信的分析结果。
应用管理平台110例如可以经由交互窗口(例如交互窗口142)来接收用户输入，并经由该交互窗口来提供分析结果。分析结果例如可以以来自数字助手的会话消息的形式被提供给用户。
应用管理平台110例如可以将至少一个处理结果提供给目标模型，以利用目标模型生成针对目标数据的分析结果。在一些实施例中，应用管理平台110可以基于至少一个处理结果生成针对目标模型的提示词输入，并将生成的提示词输入提供给目标模型。应用管理平台110例如可以基于至少一个处理结果，生成一个提示词输入，该提示词输入可以指示目标模型对至少一个处理结果进行分析。应用管理平台110例如还可以基于至少一个处理结果，生成至少一个提示词输入，每个提示词输入指示对相应的处理结果进行分析。
在一些实施例中，应用管理平台110还可以确定来源信息和至少一个处理结果的输入消耗量和目标模型的输入容量。输入消耗量可以表示将来源信息和至少一个处理结果作为目标模型的输入的数据量。输入消耗量的一个示例可以是对模型的输入所消耗的词元(token)的数目，相应地，输入容量可以是一次输入中能够提供给模型的词元总数目。应用管理平台110可以将输入消耗量和目标模型的输入容量进行比较以确定输入消耗量是否超过目标模型的输入容量。若输入消耗量未超过输入容量，应用管理平台110可以直接利用目标模型处理至少一个处理结果，以确定针对目标数据的分析结果。若输入消耗量超过输入容量，应用管理平台110可以基于至少一个处理结果生成第三提示信息，以指示目标模型总结至少一个处理结果。应用管理平台110可以将第三提示信息提供给目标模型，以得到分析结果。
示例性地，以输入容量为50(例如，单位是词元数目)为例，若输入消耗量为40，应用管理平台110可以确定输入消耗量未超过输入容量，进而直接利用目标模型处理至少一个处理结果，以确定针对目标数据的分析结果。若输入消耗量为60，应用管理平台110可以确定输入消耗量超过输入容量，进而可以基于至少一个处理结果生成第三提示信息，以指示目标模型总结至少一个处理结果。应用管理平台110可以将第三提示信息提供给目标模型，以得到分析结果。
由此，当输入消耗量达到上限(也即达到输入容量)时，应用管理平台110可以提示目标模型根据当前已有结果进行总结，可以向用户提供至少部分分析结果，可以提高用户的用户体验。
参考图3，图3示出了根据本公开的一些实施例的信息处理的示例300的示意图。如图3所示，应用管理平台110可以响应于获取到用户输入340，获取用户输入340所指示的目标数据304。继续上文的示例，若用户输入340为“总结商品A过去一年的销售额”，则目标数据304可以为商品A过去一年的销售数据。应用管理平台110可以基于与目标数据301的来源有关的来源信息306，利用目标模型(例如模型330)，生成针对目标数据304的至少一个处理指令。处理指令可以用于对目标数据执行任何类型的数据处理，例如但不限于数据转换、数据清洗、排序、各种数值运算等等。继续上文的示例，若用户输入340为“总结商品A过去一年的销售额”，处理指令可以用于计算过去12个月每个月商品A的销售额和/或销售量等。
如前文在框220处所描述的，应用管理平台110可以至少基于来源信息306来生成针对模型330的提示信息，并将提示信息提供给模型330。模型330可以输出针对目标数据304的至少一个处理指令。模型330可以是应用管理平台110本地的模型，也可以是被部署在远端设备处的模型。如果模型330被部署在应用管理平台110本地，应用管理平台110可以直接利用该模型330来确定至少一个处理指令。如果模型330被部署在远端设备，应用管理平台110可以经由和远端设备之间的通信连接，调用部署在远端设备处的模型330来确定至少一个处理指令。
来源信息306可以包括用于获取目标数据的一个或多个数据查询指令，目标数据所源自的一个或多个数据对象的元信息302，和/或，一个或多个数据对象中的数据记录示例303。例如，来源信息306可以包括用于获取商品A过去一年的销售数据的一个或多个查询指令。应用管理平台110例如可以在生成至少一个处理指令之前，将目标数据304提前上传至指令执行环境320，以将目标数据304存储到指令执行环境320。应用管理平台110可以获取针对目标数据304的地址信息，地址信息指示目标数据304在指令执行环境320中的存储位置。应用管理平台110可以基于地址信息、来源信息306和用户输入340，生成针对目标模型的提示信息310(例如第一提示信息)。
关于生成至少一个处理指令的具体方式，应用管理平台110例如可以执行框350所包括的多个步骤来生成至少一个处理指令。具体地，在框351，应用管理平台110可以确定用户输入340所指示的多个任务。例如，应用管理平台110可以将基于用户输入340所生成的提示信息提供给目标模型，由目标模型来解析出这些任务。在框352，应用管理平台110可以基于多个任务，生成多个提示信息。在框353，应用管理平台110可以将多个提示信息提供给目标模型，以利用目标模型确定多个任务对应的多个处理指令。需要注意的是，框350所包括的多个步骤中的至少一个步骤可以是应用管理平台110借助模型330来确定的。应用管理平台110可以直接利用模型330或者从远端设备处调用模型330来执行这至少一个步骤。作为示例，在框351，应用管理平台110可以将基于用户输入340所生成的提示信息提供给目标模型，由目标模型来解析出这些任务。在框352，应用管理平台110可以针对这些任务，分别生成对应的提示信息。在框353，所生成的提示信息可以被提供给目标模型，从而由目标模型生成与任务对应的处理指令。
进一步地，在框360，应用管理平台110可以通过对目标数据304执行至少一个处理指令，分别获取至少一个处理结果。例如，处理结果可以是所计算的商品A在过去一年中每个月的销售量和/或销售额。在一些实施例中，若包括多个处理指令，应用管理平台110可以多次对目标数据304执行处理指令，以得到多个处理结果。应用管理平台110例如可以将至少一个处理指令提供给指令执行环境320，并从指令执行环境320处获取至少一个处理指令对应的至少一个处理结果。
在框370，应用管理平台110可以将至少一个处理结果提供给目标模型(同样地，例如可以为模型330)，以利用目标模型，生成针对目标数据304的分析结果作为针对用户输入340的响应。继续上文的示例，可以将商品A每个月的销售量和/或销售额提供给目标模型。这样，目标模型可以输出对销售量和/或销售额的分析结果，例如销售量和/或销售额的变化趋势、最高销售量和/或销售额出现的月份、最低销售量和/或销售额出现的月份等。在一些实施例中，应用管理平台110可以确定来源信息306和至少一个处理结果的输入消耗量301是否超过目标模型的输入容量305。应用管理平台110可以响应于确定输入消耗量301超过输入容量305，基于至少一个处理结果生成指示目标模型总结至少一个处理结果的提示信息。应用管理平台110可以将该提示信息提供给目标模型，以得到分析结果。应用管理平台110可以响应于确定输入消耗量301未超过输入容量305，直接将至少一个处理结果提供给目标模型，以得到分析结果。
综上所述，根据本公开的实施例，可以借助模型来生成对数据的处理指令，并可以借助模型来基于数据指令的处理结果来确定数据的分析结果。这有助于提高信息处理的效率和准确性。此外，还加强了模型对复杂信息和大数据量信息的处理能力。
本公开的实施例还提供了用于实现上述方法或过程的相应装置。图4示出了根据本公开的一些实施例的用于信息处理的装置400的示意性结构框图。装置400例如可以被实现在或被包括在应用管理平台110中。装置400中的各个模块/组件可以由硬件、软件、固件或者它们的任何组合来实现。
如图所示，装置400包括目标数据获取模块410，被配置为响应于接收到用户输入，获取用户输入所指示的目标数据。装置400还包括处理指令生成模块420，被配置为基于与目标数据的来源有关的来源信息，利用目标模型，生成针对目标数据的至少一个处理指令。装置400还包括处理结果获取模块430，被配置为通过对目标数据执行至少一个处理指令，分别获取至少一个处理结果。装置400还包括分析结果生成模块440，被配置为基于至少一个处理结果，利用目标模型，生成针对目标数据的分析结果作为针对用户输入的响应。
在一些实施例中，处理指令生成模块420包括：地址信息获取模块，被配置为获取针对目标数据的地址信息，地址信息指示目标数据的存储位置；第一提示信息生成模块，被配置为基于地址信息、来源信息和用户输入，生成针对目标模型的第一提示信息；以及第一处理指令获取模块，被配置为将第一提示信息提供给目标模型，以得到至少一个处理指令中的第一处理指令。
在一些实施例中，至少一个处理指令是在预设的指令执行环境中被执行的，并且装置400还包括：目标数据存储模块，被配置为在生成至少一个处理指令之前，将目标数据存储到指令执行环境，地址信息指示目标数据在指令执行环境中的存储位置。
在一些实施例中，装置400还包括：任务确定模块，被配置为利用目标模型，确定用户输入至少指示第一任务和第二任务，并且其中第一处理指令与第一任务相关联，并且处理指令生成模块420还包括：第二提示信息生成模块，被配置为基于地址信息、来源信息和第一处理指令的第一处理结果，生成针对目标模型的第二提示信息；以及第二处理指令获取模块，被配置为将第二提示信息提供给目标模型，以得到至少一个处理指令中的第二处理指令，以用于第二任务。
在一些实施例中，来源信息包括以下至少一项：用于获取目标数据的一个或多个数据查询指令，目标数据所源自的一个或多个数据对象的元信息，或一个或多个数据对象中的数据记录示例。
在一些实施例中，分析结果生成模块440包括：容量确定模块，被配置为确定来源信息和至少一个处理结果的输入消耗量是否超过目标模型的输入容量，输入消耗量表示将来源信息和至少一个处理结果作为目标模型的输入的数据量；第三提示信息生成模块，被配置为响应于确定输入消耗量超过输入容量，基于至少一个处理结果生成第三提示信息，以指示目标模型总结至少一个处理结果；以及分析结果获取模块，被配置为将第三提示信息提供给目标模型，以得到分析结果。
在一些实施例中，装置400还包括：目标数据更新模块，被配置为响应于至少一个处理指令中的给定处理指令完成执行，基于给定处理指令的处理结果更新目标数据，并且在给定处理指令之后的处理指令的生成是基于经更新的目标数据。
在一些实施例中，分析结果包括引用信息，引用信息指示分析结果中的一项或多项数据的来源。
装置400中所包括的单元和/或模块可以利用各种方式来实现，包括软件、硬件、固件或其任意组合。在一些实施例中，一个或多个单元和/或模块可以使用软件和/或固件来实现，例如存储在存储介质上的机器可执行指令。除了机器可执行指令之外或者作为替代，装置400中的部分或者全部单元和/或模块可以至少部分地由一个或多个硬件逻辑组件来实现。作为示例而非限制，可以使用的示范类型的硬件逻辑组件包括现场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准品(ASSP)、片上系统(SOC)、复杂可编程逻辑器件(CPLD)，等等。
图5示出了其中可以实现本公开的一个或多个实施例的电子设备500的框图。应当理解，图5所示出的电子设备500仅仅是示例性的，而不应当构成对本文所描述的实施例的功能和范围的任何限制。图5所示出的电子设备500可以包括或被实现为图1的应用管理平台110，或图4的装置400。
如图5所示，电子设备500是通用电子设备的形式。电子设备500的组件可以包括但不限于一个或多个处理器510或处理单元、存储器520、存储设备530、一个或多个通信单元540、一个或多个输入设备550以及一个或多个输出设备560。处理器510可以是实际或虚拟处理器并且能够根据存储器520中存储的程序来执行各种处理。在多处理器系统中，多个处理器并行执行计算机可执行指令，以提高电子设备500的并行处理能力。
电子设备500通常包括多个计算机存储介质。这样的介质可以是电子设备500可访问的任何可以获取的介质，包括但不限于易失性和非易失性介质、可拆卸和不可拆卸介质。存储器520可以是易失性存储器(例如寄存器、高速缓存、随机访问存储器(RAM))、非易失性存储器(例如，只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、闪存)或它们的某种组合。存储设备530可以是可拆卸或不可拆卸的介质，并且可以包括机器可读介质，诸如闪存驱动、磁盘或者任何其他介质，其可以能够用于存储信息和/或数据并且可以在电子设备500内被访问。
电子设备500可以进一步包括另外的可拆卸/不可拆卸、易失性/非易失性存储介质。尽管未在图5中示出，可以提供用于从可拆卸、非易失性磁盘(例如“软盘”)进行读取或写入的磁盘驱动和用于从可拆卸、非易失性光盘进行读取或写入的光盘驱动。在这些情况中，每个驱动可以由一个或多个数据介质接口被连接至总线(未示出)。存储器520可以包括计算机程序产品525，其具有一个或多个程序模块，这些程序模块被配置为执行本公开的各种实施例的各种方法或动作。
通信单元540实现通过通信介质与其他电子设备进行通信。附加地，电子设备500的组件的功能可以以单个计算集群或多个计算机器来实现，这些计算机器能够通过通信连接进行通信。因此，电子设备500可以使用与一个或多个其他服务器、网络个人计算机(PC)或者另一个网络节点的逻辑连接来在联网环境中进行操作。
输入设备550可以是一个或多个输入设备，例如鼠标、键盘、追踪球等。输出设备560可以是一个或多个输出设备，例如显示器、扬声器、打印机等。电子设备500还可以根据需要通过通信单元540与一个或多个外部设备(未示出)进行通信，外部设备诸如存储设备、显示设备等，与一个或多个使得用户与电子设备500交互的设备进行通信，或者与使得电子设备500与一个或多个其他电子设备通信的任何设备(例如，网卡、调制解调器等)进行通信。这样的通信可以经由输入/输出(I/O)接口(未示出)来执行。
根据本公开的示例性实现方式，提供了一种计算机可读存储介质，其上存储有计算机可执行指令，其中计算机可执行指令被处理器执行以实现上文描述的方法。根据本公开的示例性实现方式，还提供了一种计算机程序产品，计算机程序产品被有形地存储在非瞬态计算机可读介质上并且包括计算机可执行指令，而计算机可执行指令被处理器执行以实现上文描述的方法。
这里参照根据本公开实现的方法、装置、设备和计算机程序产品的流程图和/或框图描述了本公开的各个方面。应当理解，流程图和/或框图的每个方框以及流程图和/或框图中各方框的组合，都可以由计算机可读程序指令实现。
这些计算机可读程序指令可以提供给通用计算机、专用计算机或其他可编程数据处理装置的处理器，从而生产出一种机器，使得这些指令在通过计算机或其他可编程数据处理装置的处理器执行时，产生了实现流程图和/或框图中的一个或多个方框中规定的功能/动作的装置。也可以把这些计算机可读程序指令存储在计算机可读存储介质中，这些指令使得计算机、可编程数据处理装置和/或其他设备以特定方式工作，从而，存储有指令的计算机可读介质则包括一个制造品，其包括实现流程图和/或框图中的一个或多个方框中规定的功能/动作的各个方面的指令。
可以把计算机可读程序指令加载到计算机、其他可编程数据处理装置、或其他设备上，使得在计算机、其他可编程数据处理装置或其他设备上执行一系列操作步骤，以产生计算机实现的过程，从而使得在计算机、其他可编程数据处理装置、或其他设备上执行的指令实现流程图和/或框图中的一个或多个方框中规定的功能/动作。
附图中的流程图和框图显示了根据本公开的多个实现的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或指令的一部分，模块、程序段或指令的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
以上已经描述了本公开的各实现，上述说明是示例性的，并非穷尽性的，并且也不限于所公开的各实现。在不偏离所说明的各实现的范围和精神的情况下，对于本技术领域的普通技术人员来说许多修改和变更都是显而易见的。本文中所用术语的选择，旨在最好地解释各实现的原理、实际应用或对市场中的技术的改进，或者使本技术领域的其他普通技术人员能理解本文公开的各个实现方式。

Claims (12)

  1. 一种信息处理方法，包括：
    响应于接收到用户输入，获取所述用户输入所指示的目标数据；
    基于与所述目标数据的来源有关的来源信息，利用目标模型，生成针对所述目标数据的至少一个处理指令；
    通过对所述目标数据执行所述至少一个处理指令，分别获取至少一个处理结果；以及
    基于所述至少一个处理结果，利用所述目标模型，生成针对所述目标数据的分析结果作为针对所述用户输入的响应。
  2. 根据权利要求1所述的方法，其中生成针对所述目标数据的至少一个处理指令包括：
    获取针对所述目标数据的地址信息，所述地址信息指示所述目标数据的存储位置；
    基于所述地址信息、所述来源信息和所述用户输入，生成针对所述目标模型的第一提示信息；以及
    将所述第一提示信息提供给所述目标模型，以得到所述至少一个处理指令中的第一处理指令。
  3. 根据权利要求2所述的方法，其中所述至少一个处理指令是在预设的指令执行环境中被执行的，并且所述方法还包括：
    在生成所述至少一个处理指令之前，将所述目标数据存储到所述指令执行环境，所述地址信息指示所述目标数据在所述指令执行环境中的存储位置。
  4. 根据权利要求2所述的方法，还包括：
    利用所述目标模型，确定所述用户输入至少指示第一任务和第二任务，并且
    其中所述第一处理指令与所述第一任务相关联，并且生成针对所述目标数据的至少一个处理指令还包括：
    基于所述地址信息、所述来源信息和所述第一处理指令的第一处理结果，生成针对所述目标模型的第二提示信息；以及
    将所述第二提示信息提供给所述目标模型，以得到所述至少一个处理指令中的第二处理指令，以用于所述第二任务。
  5. 根据权利要求1所述的方法，其中所述来源信息包括以下至少一项：
    用于获取所述目标数据的一个或多个数据查询指令，
    所述目标数据所源自的一个或多个数据对象的元信息，或
    所述一个或多个数据对象中的数据记录示例。
  6. 根据权利要求1所述的方法，其中生成针对所述目标数据的分析结果作为针对所述用户输入的响应包括：
    确定所述来源信息和所述至少一个处理结果的输入消耗量是否超过所述目标模型的输入容量，所述输入消耗量表示将所述来源信息和所述至少一个处理结果作为所述目标模型的输入的数据量；
    响应于确定所述输入消耗量超过所述输入容量，基于所述至少一个处理结果生成第三提示信息，以指示所述目标模型总结所述至少一个处理结果；以及
    将所述第三提示信息提供给所述目标模型，以得到所述分析结果。
  7. 根据权利要求1所述的方法，还包括：
    响应于所述至少一个处理指令中的给定处理指令完成执行，基于所述给定处理指令的处理结果更新所述目标数据，并且
    在所述给定处理指令之后的处理指令的生成是基于经更新的所述目标数据。
  8. 根据权利要求1所述的方法，其中所述分析结果包括引用信息，所述引用信息指示所述分析结果中的一项或多项数据的来源。
  9. 一种用于信息处理的装置，包括：
    目标数据获取模块，被配置为响应于接收到用户输入，获取所述用户输入所指示的目标数据；
    处理指令生成模块，被配置为基于与所述目标数据的来源有关的来源信息，利用目标模型，生成针对所述目标数据的至少一个处理指令；
    处理结果获取模块，被配置为通过对所述目标数据执行所述至少一个处理指令，分别获取至少一个处理结果；以及
    分析结果生成模块，被配置为基于所述至少一个处理结果，利用所述目标模型，生成针对所述目标数据的分析结果作为针对所述用户输入的响应。
  10. 一种电子设备，包括：
    至少一个处理器；以及
    至少一个存储器，所述至少一个存储器被耦合到所述至少一个处理器并且存储用于由所述至少一个处理器执行的指令，所述指令在由所述至少一个处理器执行时使所述电子设备执行根据权利要求1至8中任一项所述的方法。
  11. 一种计算机可读存储介质，其上存储有计算机可执行指令，所述计算机可执行指令可由处理器执行以实现根据权利要求1至8中任一项所述的方法。
  12. 一种计算机程序产品，所述计算机程序产品被有形地存储在计算机存储介质中并且包括计算机可执行指令，所述计算机可执行指令在由设备执行时使所述设备执行根据权利要求1-8中任一项所述的方法。
PCT/CN2025/080095 2024-06-27 2025-02-28 信息处理方法、装置、设备和存储介质 Pending WO2026001060A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202410851923.1A CN118839768A (zh) 2024-06-27 2024-06-27 信息处理方法、装置、设备和存储介质
CN202410851923.1 2024-06-27

Publications (1)

Publication Number Publication Date
WO2026001060A1 true WO2026001060A1 (zh) 2026-01-02

Family

ID=93141635

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2025/080095 Pending WO2026001060A1 (zh) 2024-06-27 2025-02-28 信息处理方法、装置、设备和存储介质

Country Status (2)

Country Link
CN (1) CN118839768A (zh)
WO (1) WO2026001060A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118839768A (zh) * 2024-06-27 2024-10-25 北京字跳网络技术有限公司 信息处理方法、装置、设备和存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210357592A1 (en) * 2020-05-14 2021-11-18 Oracle International Corporation Method and system for defining an adaptive polymorphic object agnostic conversational interaction model
CN115905284A (zh) * 2022-11-15 2023-04-04 北京火山引擎科技有限公司 一种数据处理方法、装置、设备及存储介质
CN117273017A (zh) * 2023-09-21 2023-12-22 广州趣研网络科技有限公司 信息处理方法、信息处理装置、计算机设备及存储介质
CN117555897A (zh) * 2023-10-10 2024-02-13 北京百度网讯科技有限公司 基于大模型的数据查询方法、装置、设备和存储介质
CN117952212A (zh) * 2024-01-02 2024-04-30 科大讯飞股份有限公司 基于大语言模型和区块链的数据处理方法及装置
CN118069820A (zh) * 2024-04-22 2024-05-24 来未来科技(浙江)有限公司 一种对话式数据分析方法、装置、设备及存储介质
CN118839768A (zh) * 2024-06-27 2024-10-25 北京字跳网络技术有限公司 信息处理方法、装置、设备和存储介质

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114205417B (zh) * 2021-11-16 2024-05-31 上海翀远智能科技有限公司 一种信息处理方法、装置、电子设备及存储介质
CN116821103B (zh) * 2023-08-29 2023-12-19 腾讯科技（深圳）有限公司 数据处理方法、装置、设备及计算机可读存储介质
CN117235222A (zh) * 2023-09-18 2023-12-15 北京奇艺世纪科技有限公司 基于模型的交互方法、装置、电子设备及存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210357592A1 (en) * 2020-05-14 2021-11-18 Oracle International Corporation Method and system for defining an adaptive polymorphic object agnostic conversational interaction model
CN115905284A (zh) * 2022-11-15 2023-04-04 北京火山引擎科技有限公司 一种数据处理方法、装置、设备及存储介质
CN117273017A (zh) * 2023-09-21 2023-12-22 广州趣研网络科技有限公司 信息处理方法、信息处理装置、计算机设备及存储介质
CN117555897A (zh) * 2023-10-10 2024-02-13 北京百度网讯科技有限公司 基于大模型的数据查询方法、装置、设备和存储介质
CN117952212A (zh) * 2024-01-02 2024-04-30 科大讯飞股份有限公司 基于大语言模型和区块链的数据处理方法及装置
CN118069820A (zh) * 2024-04-22 2024-05-24 来未来科技(浙江)有限公司 一种对话式数据分析方法、装置、设备及存储介质
CN118839768A (zh) * 2024-06-27 2024-10-25 北京字跳网络技术有限公司 信息处理方法、装置、设备和存储介质

Also Published As

Publication number Publication date
CN118839768A (zh) 2024-10-25

Similar Documents

Publication Publication Date Title
CN108519967B (zh) 图表可视化方法、装置、终端和存储介质
WO2025223028A1 (zh) 信息处理方法、装置、设备和存储介质
WO2024240105A1 (zh) 用于信息处理的方法、装置、设备和存储介质
WO2026001052A1 (zh) 任务处理方法、装置、设备和计算机可读存储介质
WO2026001060A1 (zh) 信息处理方法、装置、设备和存储介质
WO2025223030A1 (zh) 信息处理方法、装置、设备和存储介质
WO2025223031A1 (zh) 信息处理方法、装置、设备和存储介质
WO2025246492A1 (zh) 信息处理方法、装置、设备和存储介质
WO2026001042A1 (zh) 信息处理方法、装置、设备和存储介质
WO2026040475A1 (zh) 数据处理方法、装置、设备和可读存储介质
WO2026040472A1 (zh) 应用配置方法、信息处理方法、装置、设备和存储介质
WO2026001059A1 (zh) 任务处理方法、装置、设备和计算机可读存储介质
WO2026000958A1 (zh) 用于人机对话的方法、装置、设备、介质和程序产品
WO2025222808A1 (zh) 代码编辑的方法、装置、设备和存储介质
WO2025107890A1 (zh) 用于应用创建的方法、装置、设备和存储介质
CN118433145A (zh) 消息处理方法、装置、设备和存储介质
CN111291026B (zh) 数据接入方法、系统、装置以及计算机可读介质
CN114139091A (zh) 处理延迟的评估方法、装置、介质及电子设备
CN113687881A (zh) 元数据调用方法、装置、电子设备及存储介质
US20260093702A1 (en) Method for data processing, apparatus, device, storage medium and program product
US20260050437A1 (en) Application creation
US20260050494A1 (en) Method, device and storage medium for function invocation
WO2025108115A1 (zh) 数据处理的方法、装置、设备和存储介质
WO2026061216A1 (zh) 用于请求处理的方法、装置、设备、存储介质和程序产品
WO2025200591A1 (zh) 用于创建数字助手的方法、装置、设备和介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 25824455

Country of ref document: EP

Kind code of ref document: A1