WO2026001052A1 - 任务处理方法、装置、设备和计算机可读存储介质 - Google Patents

任务处理方法、装置、设备和计算机可读存储介质

Info

Publication number
WO2026001052A1
WO2026001052A1 PCT/CN2025/079845 CN2025079845W WO2026001052A1 WO 2026001052 A1 WO2026001052 A1 WO 2026001052A1 CN 2025079845 W CN2025079845 W CN 2025079845W WO 2026001052 A1 WO2026001052 A1 WO 2026001052A1
Authority
WO
WIPO (PCT)
Prior art keywords
query
instruction
analysis
data
results
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/CN2025/079845
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 WO2026001052A1 publication Critical patent/WO2026001052A1/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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • 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 relate generally to the field of computers, and more particularly to task 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 task processing requests within the application. Therefore, how to accurately complete these task processing requests is a key concern.
  • a task processing method includes: parsing received user input to obtain a parsing result, wherein the user input indicates a task execution request for a target application; generating a query instruction based on the parsing result; and, in response to determining that a query result obtained using the query instruction does not meet specified conditions, generating an analysis instruction for analyzing the query result obtained using the query instruction.
  • a task processing apparatus comprising: a parsing module configured to parse received user input to obtain a parsing result, wherein the user input indicates a task execution request for a target application; a query instruction generation module configured to generate a query instruction based on the parsing result; and an analysis instruction generation module configured to generate an analysis instruction in response to determining that a query result obtained using the query instruction does not meet specified conditions, wherein the analysis instruction is used to analyze the query result obtained using the query instruction.
  • 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 is provided.
  • the computer program product is tangibly stored in a computer storage medium and includes computer-executable instructions that, when executed by a device, cause the device to perform the method of the first aspect.
  • Figure 1 shows a schematic diagram of an example environment in which embodiments of the present disclosure can be implemented
  • Figure 3 illustrates a schematic diagram of user input according to some embodiments of the present disclosure
  • Figure 4 illustrates a schematic diagram of a task processing procedure according to some embodiments of the present disclosure
  • Figure 5 shows a schematic diagram of a page displaying data results according to some embodiments of the present disclosure
  • Figure 6 shows a schematic structural block diagram of a task processing apparatus according to some embodiments of the present disclosure
  • Figure 7 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.
  • target applications typically receive user input, which constitutes a task execution request for the target application.
  • SQL structured query language
  • SQL is usually used to retrieve the necessary information from large amounts of data.
  • Complex calculations such as year-on-year and month-on-month comparisons, often cannot be completed using only SQL queries. Therefore, it is worthwhile to explore how applications can accurately understand the complex calculation problems posed by users and select appropriate tools to solve the problems raised in the task execution requests.
  • This disclosure provides an improved task processing scheme.
  • received user input is parsed to obtain a parsing result, whereby the user input indicates a task execution request for a target application.
  • a query instruction is generated based on the parsing result.
  • an analysis instruction is generated to analyze the query result obtained using the query instruction.
  • a suitable query instruction is first selected based on the parsing result of the user input, and the user input corresponds to the task execution request sent by the user to the target application.
  • the analysis instruction can be combined.
  • the analysis instruction is used to analyze and perform calculations on the query result obtained from the query instruction, thereby solving the problem posed by the task execution request using multiple methods.
  • 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 end user 145, and the application management platform 110 can receive user input via the terminal device of end user 145.
  • end user 145 is the end 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.
  • FIG. 2 shows a block diagram of a task processing procedure 200 according to some embodiments of the present disclosure. This task processing procedure can be implemented on an 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 parses the received user input and obtains the parsing result.
  • the user input indicates a task execution request for the target application 120.
  • User input can be natural language received during a conversation between the end user and the digital assistant.
  • User input is a task execution request for the target application 120.
  • user input could be "a list of students who got full marks in math on the final exam in this class," or "the monthly sales percentage change of product A last year,” and so on.
  • the conversation between the end user and the digital assistant typically includes explanations or illustrative descriptions of terms that may be involved in the task to be performed.
  • the application management platform 110 can parse the user input. For example, parsing can be performed by the application management platform 110 calling model 155, leveraging the natural language processing capabilities of model 155. Model 155 is also the corresponding target model. Model 155 can also be called a machine learning model. By parsing the user input, the parsing result can be obtained. For example, the parsing may include at least the question posed in the task execution request, and the data required to solve the question.
  • the application management platform 110 can determine that the task execution request asks for the list of people to be retrieved.
  • the required data is the math scores of all students in the class on the final exam.
  • the application management platform 110 can determine that the task execution request asks for a sales comparison. It needs to obtain monthly sales data for product A from January to December last year, as well as monthly sales data for all products from January to December last year.
  • Page 300 includes a query content input area 301, a terminology set area 302, a dialogue record area 303, and a reference prompt area 304, etc.
  • the query content input area 301 can be used to receive user input from the terminal user 145.
  • the terminology set area 302 can be used to receive terms. Terminology includes explanations or examples of terms used by the target application's user group.
  • the dialogue record area 303 can be used to record the dialogue process between the terminal user 145 and the digital assistant, including query instructions and/or analysis instructions generated by the digital assistant 122, etc.
  • the reference prompt area 304 can be used to receive reference prompt content.
  • the content of the reference prompt area 304 is the role description that the model 155 can receive from the terminal user 145 to the digital assistant 122, thereby completing the guidance for the digital assistant 122.
  • the application management platform 110 can generate query instructions based on the parsing results.
  • the application management platform 110 can generate query instructions corresponding to those results. For example, if the parsing result is to retrieve the math scores of all students in the class's final exam, the application management platform 110 can generate a query instruction to retrieve the math scores of all students in the class's final exam based on the known database, the tables in the known database, and the columns in the tables. Similarly, if the parsing result is to retrieve the sales data of product A for each month from January to December of last year, and the sales data of all products for each month from January to December of last year, the application management platform 110 can generate a query instruction to retrieve the sales volume of product A and the sales volume of all products based on the known database, the tables in the known database, and the columns in the tables.
  • the management platform 110 can use the code compilation instructions of model 155 to combine user input as tokens for the generated query instructions. Furthermore, the management platform 110 can input parsing results and user interaction content as context information into model 155. Finally, the management platform 110 can also input database information, such as database name and database access address, into model 155. This allows the generation of query instructions using model 155.
  • the query instruction can be an SQL query statement.
  • the application management platform 110 in box 203, the application management platform 110 generates an analysis command in response to determining that the query result obtained by the query command does not meet the specified conditions.
  • the analysis command is used to analyze the query result obtained by the query command.
  • the application management platform 110 can send the query command to the target application 120, causing the target application 120 to execute the query command and obtain the query results.
  • the application management platform 110 can determine whether the query results can resolve the problem raised in the task execution request. The determination of the query results includes whether it can resolve the problem or not.
  • the application management platform 110 can determine that the obtained data represents the math scores of all students in the class. Therefore, the application management platform 110's query result is unresolved. However, if the application management platform 110 determines that the query result can be optimized by combining it with other query statements to obtain a list of students who achieved a perfect score in the math final exam, then the query result can be considered resolvable. For example, a filtering query can be used to filter out students who achieved a perfect score in math, thus obtaining a list of all students who achieved a perfect score in math.
  • the application management platform 110 Based on the example of the user input being "the change in the monthly sales percentage of product A last year," if the application management platform 110 obtains the monthly sales figures for product A over the past 12 months, as well as the monthly sales figures for all products over the past 12 months, it can determine that sales data alone cannot resolve the issue of determining the change in sales percentage. For example, if the application management platform 110 determines that there are no other query commands that can yield the result of the change in sales percentage, then it can be determined that the query result is insufficient to resolve the issue.
  • the application management platform 110 determines that the query results obtained using the query command do not meet the specified conditions, it can indicate that the query command alone cannot resolve the problem raised in the task execution request. In response, the application management platform 110 can introduce an analysis command. While the advantage of the query command is its ability to quickly retrieve data, the advantage of the analysis command lies in its capacity to perform complex calculations on the data. For example, issues such as percentage, year-on-year comparison, month-on-month comparison, and data analysis can all involve complex calculations.
  • the code compilation for both the query command and the analysis command can be completed by the application management platform 110 calling model 155. Utilizing the code compilation capabilities of model 155 saves the time and effort previously spent on manual code compilation.
  • the analysis command can be configured by the application management platform 110 to calculate the monthly sales percentage of product A. Furthermore, the analysis command can also be configured by the application management platform 110 to derive the monthly change in the monthly sales percentage of product A based on the sales percentage over the past 12 months.
  • the system By parsing the user input, the system first matches the query command corresponding to the parsed result. Then, if the query command does not meet the specified conditions, an analysis command is introduced. Combining the analysis and query commands leverages the efficiency of query commands in data retrieval and the ability of analysis commands to handle complex data analysis problems. This allows for the use of different approaches to complete the task execution request response. In this way, the advantages of both query and analysis commands are fully utilized to accomplish the task processing.
  • the application management platform 110 can utilize a target model to determine, based on user input, that a task execution request requires multiple execution steps. In response to a request to split the task execution request into multiple subtasks based on the requirement of multiple execution steps, this is presented as a parsing result.
  • the application management platform 110 can call model 155 and utilize its natural language processing capabilities to parse user input. Taking the example of user input as "the change in the monthly sales percentage of product A last year," the parsing result from the application management platform 110 is that it needs to obtain the monthly sales figures for product A from January to December of last year, as well as the monthly sales figures for all products from January to December of last year. Therefore, the application management platform 110 can determine that the task execution request requires multiple execution steps. For example, the first step is to obtain the monthly sales figures for product A from January to December of last year. The second step is to obtain the monthly sales figures for all products from January to December of last year.
  • the application management platform 110 can further divide the first step into 12 sub-steps, meaning each sub-step corresponds to obtaining the sales figures for product A in one month of last year. Similarly, the application management platform 110 can further subdivide the second step into 12 sub-steps, meaning each sub-step corresponds to obtaining the sales figures for all products in one month of last year.
  • the application management platform 110 can generate 12 query commands, each configured to retrieve one month's sales data for product A.
  • the application management platform 110 can also generate 12 query commands, each configured to retrieve one month's sales data for all products.
  • the application management platform 110 may determine the data description of the data to be acquired based on the parsing results.
  • the data description is used to indicate at least one of the terms and examples of the data to be acquired, and to generate query instructions based on the data description using the target model.
  • parsing can be completed by the application management platform 110 calling model 155, leveraging the natural language processing capabilities of model 155 to obtain the parsing results.
  • the parsing results are based on user input and the natural language processing capabilities of model 155, therefore they are typically unstructured.
  • query instructions such as SQL queries, are structured, which may lead to discrepancies between the parsing results and the query instructions. Therefore, after obtaining the parsing results, the application management platform 110 can match the parsing results with the terminology examples received in the terminology set display area 302, thereby using the successfully matched terms as data descriptions for the data to be acquired.
  • the application management platform 110 can declare examples of the data to be acquired based on those results. These examples can indicate the data format, content, etc., of the data to be acquired. They can also serve as a data description of the data to be acquired.
  • Model 155 can generate a query instruction based on the data description of the data to be acquired, so that the query instruction can better match the data acquisition needs of the parsing results.
  • the application management platform 110 can also add a data source identifier to the query results obtained through the query instruction.
  • the data source identifier is used to indicate the source of the query results in the source data.
  • the application management platform 110 can also add a data source identifier to the query results.
  • the data source identifier is used to indicate the origin of the query results in the source data.
  • the data source identifier can be used to indicate information such as the database name, the table in the database, and the columns in the table from which the query results obtained by the query command originate.
  • the significance of adding data source labels is that it can mark the authenticity of data, thereby avoiding false data without a source.
  • the application management platform 110 determines that the query result obtained using the query instruction does not meet the specified conditions, including: in response to the query result obtained using the query instruction failing to resolve the problem raised in the task execution request, determining the reason, which is the reason why the problem raised in the task execution request cannot be resolved; in response to the reason indicating that the instruction category of the query instruction is correct, adjusting the query instruction to obtain an adjusted query instruction, the adjustment including one of replacing the query instruction and adding a new query instruction; and in response to the adjustment satisfying a preset adjustment process, and the query result obtained using the adjusted query instruction still failing to resolve the problem raised in the task execution request, determining the reason indicating that the instruction category of the query instruction is incorrect.
  • Figure 4 shows a schematic diagram of the task processing method flow 400.
  • the application management platform 110 needs to determine whether the query result obtained from the query command can resolve the problem raised in the task execution request; that is, to determine whether the problem is resolved.
  • the query result can include whether it can be resolved or not.
  • the application management platform 110 needs to further determine the reason for the problem's inability to be resolved in box 404, that is, to determine whether the reason is an error in the query instruction category.
  • the reason for the problem being unresolved can be categorized as the query instruction category being correct, but the query instruction needs adjustment; or the reason for the problem being unresolved can be categorized as the query instruction category being incorrect. If it is determined that the reason for the unresolved problem is that the query instruction category is correct, then it can be indicated that the current query instruction needs adjustment, that is, returning to box 401. Adjustments can include replacing or adding new query instructions.
  • the application management platform 110 can generate a new query command based on the actual situation.
  • the new query command replaces the original query command for data retrieval.
  • the so-called actual situation can be the result of analysis based on model 155, or it can be the instruction obtained after interaction with the end user 145. The specific process of obtaining the actual situation will not be elaborated here.
  • the application management platform 110 can retrieve the math scores of all students in the class based on the obtained data. If the application management platform 110 determines that the query results obtained using the query command cannot resolve the problem raised in the task execution request, and the reason indicates that the command category of the query command is correct, then it can choose to add a new query command.
  • the new query command could be to search for students who achieved a perfect score in the math final exam of all students in the class.
  • the pre-defined adjustment process can be used to indicate the adjustment capabilities of the application management platform 110.
  • the adjustment capability can indicate the application management platform 110's ability to generate new query commands. If the capability to generate new query commands is to generate m (m is a positive integer) new query commands, but the query results obtained from the newly generated m query commands still cannot resolve the problem raised in the task execution request, the cause can be determined to be an incorrect command category of the query commands.
  • the adjustment capability can indicate the number of adjustments. If, after n adjustments (n is a positive integer), the query results obtained from the query commands still cannot resolve the problem raised in the task execution request, the cause of the inability to resolve the problem raised in the task execution request can also be determined to be an incorrect command category of the query commands.
  • the query commands can be adjusted to select a more suitable one. Conversely, if adjusting the query commands still fails to resolve the issue, it can be determined that query commands alone are insufficient. This can trigger the introduction of analysis commands, thereby better fulfilling the task execution request.
  • the application management platform 110 further determines that the query result obtained by the query instruction does not meet the specified conditions by: in response to the reason indicating that the instruction category of the query instruction is incorrect, determining that the query result obtained by the query instruction does not meet the specified conditions.
  • the application management platform 110 can determine the instruction category error using the analytical capabilities of model 155. For example, the parsing results and the query results obtained from the query instruction can be input into model 155, and model 155 can determine that the reason for the inability to resolve the problem raised in the task execution request is an incorrect instruction category.
  • the application management platform 110 generating analysis instructions may include: obtaining context information, which includes at least one of user input, parsing results, query instructions, query results obtained using the query instructions, and reasons why the query results do not meet specified conditions; and generating analysis instructions based on the context information.
  • the application management platform 110 can use model 155 to generate analysis instructions based on context information.
  • the application management platform 110 can obtain context information, which includes at least one of the following: user input, parsing results of the user input, all previously selected query instructions, query results corresponding to each query instruction, and reasons why the query results failed to meet specified conditions.
  • model 155 can be used to determine the analysis instructions based on the context. Taking the example of the user input being "the monthly sales percentage change of product A last year," after obtaining the user input, parsing the results, the query instructions, and understanding that the query instructions cannot resolve the problem raised by the task execution request because the query instructions have an incorrect instruction category, the application management platform 110 can use model 155 to perform analysis, thereby determining the necessary analysis instructions to solve the problem. Furthermore, it can be determined that the analysis instructions can be configured to perform percentage calculations and to analyze the results of the percentage calculations.
  • the application management platform 110 can use model 155 for analysis to adjust the analysis instructions.
  • the goal of the adjustment is for the adjusted analysis instructions to resolve the problem posed by the task execution request.
  • the first round of analysis instructions can determine the percentage data, but the percentage data is not equivalent to the percentage change. Therefore, a second round of analysis instructions can be generated, which is to derive the final analysis result of the percentage change based on the percentage data from multiple statistical periods.
  • the analysis instructions can be compiled using any programming language capable of performing the analysis task.
  • the basic statements of programming languages include arithmetic and logical operators, thus easily handling complex problems such as same-year comparisons, year-on-year comparisons, and percentage comparisons.
  • the code compilation process of the programming language can be completed using the code compilation capabilities of Model 155.
  • Figure 5 shows an example of a page 500 displaying data results according to some embodiments of this disclosure. User input can be displayed in the query content display area 511.
  • the data result display area 512 the data sources involved in obtaining the final data result, the query instructions generated or used in obtaining the final data result (query instruction generation), the analysis instructions generated or used in obtaining the final data result (analysis instruction generation), and the final analysis result can be displayed.
  • analysis commands can be introduced based on communication with the user and the factual conclusions drawn during the communication process.
  • the user's problem is resolved by combining analysis commands with query commands.
  • a prompt message is determined based on the context information, and the analysis instruction is generated based on the prompt message using the target model.
  • prompts can be generated based on it.
  • structured processing can be performed on the context information to obtain prompts. Structured processing might begin by identifying key fields, such as user input or parsing results. Then, corresponding content is added to different key fields to obtain the prompts. The process of identifying key fields and/or adding corresponding content can also include text deduplication and simplification.
  • Model 155 can be used to determine the analysis instructions based on them. That is, Model 155 performs natural language code compilation based on the prompts to obtain the corresponding analysis instruction code.
  • the application management platform 110 may also perform the following steps: storing data results in a designated storage area, the data results including at least one of query results obtained from query instructions and analysis results obtained from analysis instructions; and representing the data results using the address of the storage area.
  • the application management platform 110 can store the data results in a designated storage area.
  • the designated area can be a cache area within the sandbox where the analysis commands are executed.
  • the application management platform 110 can generate a storage identifier based on the address where the data results are stored.
  • the storage identifier can be used to represent the data. Therefore, when the application management platform 110 generates a new query command or a generated analysis command that requires data access, it can generate a storage identifier read command, thereby using the storage identifier read command to read the data results stored in the designated storage area.
  • the application management platform 110 can also execute analysis instructions to obtain analysis results.
  • Executing analysis instructions includes: sending information related to the analysis instructions and query results to the instruction execution environment, and obtaining the analysis results output by the instruction execution environment, wherein the analysis results are obtained by processing the query results based on the analysis instructions.
  • Analysis instructions can run in an instruction execution environment.
  • the instruction execution environment can be a sandbox environment, a sandboxed environment, etc. That is, the application management platform 110 can send the code corresponding to the analysis instruction and the query results required by the analysis instruction to the instruction execution environment.
  • the address of the storage area corresponding to the query results can be sent to the instruction execution environment.
  • the corresponding query results can be obtained based on the address of the storage area.
  • Analysis instructions are executed in the instruction execution environment to process the query results, thereby obtaining the analysis results.
  • the analysis results can be sent from the instruction execution environment to a designated storage area.
  • the analysis results stored in the designated storage area can be the final analysis results.
  • the analysis results can be directly displayed on the user interface presented by the application management platform 110.
  • Figure 6 shows a schematic structural block diagram of a task processing apparatus 600 according to some embodiments of the present disclosure.
  • the apparatus 600 may be implemented in or included in an application management platform 110, for example.
  • the various modules/components in the apparatus 600 may be implemented by hardware, software, firmware, or any combination thereof.
  • the device 600 includes a parsing module 601, configured to parse received user input to obtain a parsing result, wherein the user input indicates a task execution request for a target application; a query instruction generation module 602, configured to generate a query instruction based on the parsing result; and an analysis instruction generation module 603, configured to generate an analysis instruction in response to determining that the query result obtained using the query instruction does not meet specified conditions, wherein the analysis instruction is used to analyze the query result obtained using the query instruction.
  • a parsing module 601 configured to parse received user input to obtain a parsing result, wherein the user input indicates a task execution request for a target application
  • a query instruction generation module 602 configured to generate a query instruction based on the parsing result
  • an analysis instruction generation module 603 configured to generate an analysis instruction in response to determining that the query result obtained using the query instruction does not meet specified conditions, wherein the analysis instruction is used to analyze the query result obtained using the query instruction.
  • the parsing module 601 is further configured to include: an execution step identification submodule, which is configured to use a target model to determine, based on user input, that a task execution request requires multiple execution steps; and a splitting submodule, which is configured to, in response to a request to split the task execution request into multiple subtasks based on the requirement that the task execution request requires multiple execution steps, as the parsing result.
  • an execution step identification submodule which is configured to use a target model to determine, based on user input, that a task execution request requires multiple execution steps
  • a splitting submodule which is configured to, in response to a request to split the task execution request into multiple subtasks based on the requirement that the task execution request requires multiple execution steps, as the parsing result.
  • the query instruction generation module 602 is further configured to determine, based on the parsing results, a data description of the data to be acquired, the data description being used to indicate at least one of the terms and examples of the data to be acquired; and to generate a query instruction based on the data description using a target model.
  • the apparatus 600 further includes a data source identifier adding module, which is configured to add a data source identifier to the query result obtained by the query instruction.
  • the data source identifier is used to indicate the source of the query result in the source data.
  • the analysis instruction generation module 603 is further configured to include a cause determination submodule, configured to determine a cause in response to the query result obtained using the query instruction failing to resolve the problem raised in the task execution request, the cause being the reason why the problem raised in the task execution request cannot be resolved; a query instruction adjustment submodule, configured to adjust the query instruction in response to the cause indicating that the instruction category of the query instruction is correct, obtaining an adjusted query instruction, the adjustment including one of replacing the query instruction and adding a new query instruction; and an instruction category error determination submodule, configured to determine a cause indicating that the instruction category of the query instruction is incorrect in response to the adjustment meeting a preset adjustment process, and the query result obtained using the adjusted query instruction still failing to resolve the problem raised in the task execution request.
  • a cause determination submodule configured to determine a cause in response to the query result obtained using the query instruction failing to resolve the problem raised in the task execution request, the cause being the reason why the problem raised in the task execution request cannot be resolved
  • a query instruction adjustment submodule configured
  • the instruction category error determination submodule is further configured to determine, in response to an instruction category error indicating a cause of query instruction, that the query result obtained using the query instruction does not meet the specified conditions.
  • the analysis instruction generation module 603 is further configured to obtain context information, which includes at least one of user input, parsing results, query instructions, query results obtained using the query instructions, and reasons why the query results do not meet specified conditions; and to generate analysis instructions based on the context information.
  • the analysis instruction generation module 603 is further configured to determine the prompt information based on context information, and generate analysis instructions based on the prompt information using the target model.
  • the apparatus 600 further includes a data characterization module configured to store data results in a designated storage area, the data results including at least one of query results obtained from query instructions and analysis results obtained from analysis instructions; and to characterize the data results using the stored address.
  • a data characterization module configured to store data results in a designated storage area, the data results including at least one of query results obtained from query instructions and analysis results obtained from analysis instructions; and to characterize the data results using the stored address.
  • the apparatus 600 further includes an analysis instruction execution module, which is configured to: send information related to analysis instructions and query results to an instruction execution environment; and obtain analysis results output by the instruction execution environment, wherein the analysis results are obtained based on the analysis instructions processing the query results.
  • an analysis instruction execution module configured to: send information related to analysis instructions and query results to an instruction execution environment; and obtain analysis results output by the instruction execution environment, wherein the analysis results are obtained based on the analysis instructions processing the query results.
  • Figure 7 shows a block diagram of an electronic device 700 in which one or more embodiments of the present disclosure may be implemented. It should be understood that the electronic device 700 shown in Figure 7 is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein.
  • the electronic device 700 shown in Figure 7 may include or be implemented as the application management platform 110 of Figure 1, or the device 600 of Figure 6.
  • the electronic device 700 is in the form of a general-purpose electronic device.
  • Components of the electronic device 700 may include, but are not limited to, one or more processors 710 or processing units, a memory 720, a storage device 730, one or more communication units 740, one or more input devices 750, and one or more output devices 760.
  • the processor 710 may be a physical or virtual processor and is capable of performing various processes according to programs stored in the memory 720. In a multiprocessor system, multiple processors execute computer-executable instructions in parallel to improve the parallel processing capability of the electronic device 700.
  • Electronic device 700 typically includes multiple computer storage media. Such media can be any accessible media that is accessible to electronic device 700, including but not limited to volatile and non-volatile media, removable and non-removable media.
  • Memory 720 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 730 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 700.
  • Electronic device 700 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 720 may include computer program product 725 having one or more program modules configured to perform various methods or actions of various embodiments of the present disclosure.
  • the communication unit 740 enables communication with other electronic devices via a communication medium. Additionally, the functionality of the components of the electronic device 700 can be implemented using a single computing cluster or multiple computing machines capable of communicating via communication connections. Therefore, the electronic device 700 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 750 can be one or more input devices, such as a mouse, keyboard, trackball, etc.
  • Output device 760 can be one or more output devices, such as a monitor, speaker, printer, etc.
  • Electronic device 700 can also communicate with one or more external devices (not shown) via communication unit 740 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 700, or with any device that enables electronic device 700 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 create 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.
  • 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)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Databases & Information Systems (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种任务处理方法、装置、设备和计算机可读存储介质。方法包括：对接收到的用户输入进行解析，得到解析结果，用户输入指示针对目标应用的任务执行请求；基于解析结果生成查询指令；以及响应于确定利用查询指令得到的查询结果未满足指定条件，生成分析指令，分析指令用于对利用查询指令得到的查询结果进行分析。

Description

任务处理方法、装置、设备和计算机可读存储介质
本申请要求2024年06月27日递交的、标题为“任务处理方法、装置、设备和计算机可读存储介质”、申请号为202410851944.3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，特别地涉及任务处理方法、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任务处理请求。因此，如何准确完成任务处理请求是所关注的问题。
发明内容
在本公开的第一方面，提供了一种任务处理方法。该方法包括：对接收到的用户输入进行解析，得到解析结果，用户输入指示针对目标应用的任务执行请求；基于解析结果生成查询指令；以及响应于确定利用查询指令得到的查询结果未满足指定条件，生成分析指令，分析指令用于对利用查询指令得到的查询结果进行分析。
在本公开的第二方面，提供了一种任务处理装置，包括：解析模块，被配置为对接收到的用户输入进行解析，得到解析结果，用户输入指示针对目标应用的任务执行请求；查询指令生成模块，被配置为基于解析结果生成查询指令；以及分析指令生成模块，被配置为响应于确定利用查询指令得到的查询结果未满足指定条件，生成分析指令，分析指令用于对利用查询指令得到的查询结果进行分析。
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。
在本公开的第五方面，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，计算机可执行指令在由设备执行时使设备执行第一方面的方法。
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任务处理过程的框图；
图3示出了根据本公开的一些实施例的用户输入的示意图；
图4示出了根据本公开的一些实施例的任务处理过程的示意图；
图5示出了根据本公开的一些实施例的展示数据结果的页面的示意图；
图6示出了根据本公开的一些实施例的任务处理装置的示意性结构框图；
图7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。
目前，在目标应用中通常会接收到用户的用户输入，也就是针对目标应用的任务执行请求。以用户输入包含数据分析需求为示例，面对用户输入，通常会利用结构化查询语句(SQL，Structured Query Language)从大体量的数据中检索出所需信息。然而，随着数据分析需求的日益复杂，例如需要对数据进行跨时间段的比较分析、趋势预测、异常检测等任务，单纯依靠SQL查询的局限性逐步显现。诸如同比、环比等复杂计算问题，如果单纯依赖于SQL查询往往无法完成。对于应用而言，如何能够准确理解用户复杂的计算问题，并且选择合适的工具解决任务执行请求所提出问题是值得探讨的。
在本公开实施例中，提供了一种任务处理的改进方案。在该方案中，对接收到的用户输入进行解析，得到解析结果，用户输入指示针对目标应用的任务执行请求。基于解析结果生成查询指令。响应于确定利用查询指令得到的查询结果未满足指定条件，生成分析指令，分析指令用于对利用查询指令得到的查询结果进行分析。通过此种方式，首先根据对用户输入的解析结果选择合适的查询指令，用户的输入也就对应着用户发给目标应用的任务执行请求。在确定出查询指令无法满足指定条件，也就是在确定出仅依赖于查询指令并非是最优解的情况下，还可以结合分析指令。利用分析指令对查询指令得到的查询结果进行分析、运算等，从而从利用多种方式来解决任务执行请求所提出的问题。
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。应当理解，附图中示出的页面仅仅是示例，实际可以存在各种页面设计。页面中的各个图形元素可以具有不同的布置和不同的视觉表示，其中的一个或多个元素可以省略或被替换，并且还可以存在一个或多个其他元素。本公开的实施例在此方面不受限制。
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，可以经由终端用户145的终端设备被呈现，并且应用管理平台110可以经由终端用户145的终端设备来接收用户输入。在本公开的一些实施例中，终端用户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任务处理过程200的框图。该任务处理过程可以在应用管理平台110被实施。以下结合图1来描述图2示出的任务处理过程。
如图2所示，在框201，应用管理平台110对接收到的用户输入进行解析，得到解析结果，用户输入指示针对目标应用120的任务执行请求。
用户输入可以是终端用户与数字助手的会话中接收到的自然语言。用户输入为针对目标应用120的任务执行请求。例如，用户输入可以是“本班级期末考试数学满分的同学名单”、“A款商品去年每个月的销量占比变化”，等等。此外，终端用户与数字助手的会话中通常会对待执行任务中可能涉及到的术语所进行的解释说明或者示例性说明。
应用管理平台110接收到用户输入后，可以对用户输入进行解析。示例性地，解析可以通过应用管理平台110调用模型155，借助模型155的自然语言处理能力完成。模型155也即对应目标模型。模型155也可以称为机器学习模型。通过对用户输入进行解析，可以得到解析结果。示例性地，解析可以至少包括完成对任务执行请求所提出问题，以及解决该问题所需的数据等。
以用户输入是“本班级期末考试数学满分的同学名单”为示例，则应用管理平台110可以确定出任务执行请求所提出问题是获取人员名单。所需数据是需要获取本班级期末考试所有同学的数学成绩。
以用户输入是“A款商品去年每个月的销量占比变化”为示例，则应用管理平台110可以确定出任务执行请求所提出问题是销量比较。需要获取A款商品去年1至12个月中每个月的销量数据、所有商品去年1至12个月中每个月的销量数据等。
参考图3，图3示出了根据本公开的一些实施例的数据信息的获取页面300的示例。页面300包括查询内容输入区域301、术语集区域302、对话记录区域303和参考提示词区域304等。其中，查询内容输入区域301可以用于接收终端用户145的用户输入。术语集区域302可以用于接收术语。术语包括对目标应用的用户群体所使用的术语的解释或者给出示例。对话记录区域303可以用于记录终端用户145与数字助手之间的对话过程，包括数字助手122生成的查询指令和/或分析指令等。参考提示词区域304可以用于接收参考提示内容。通过参考提示词，可以给数字助手122提供更好的指引，帮助数字助手122更好地完成用户的任务。即，参考提示词区域304的内容是模型155可以接收到终端用户145给数字助手122的角色说明，从而完成对数字助手122的指引。
如图2所示，在框202，应用管理平台110可以基于解析结果生成查询指令。
应用管理平台110根据解析结果，可以生成与解析结果对应的查询指令。例如，解析结果为获取本班级期末考试所有同学的数学成绩，则应用管理平台110可以根据已知数据库、已知数据库中的表以及表中的列等信息，生成获取本班级期末考试所有同学的数学成绩的查询指令。同理，对于解析结果为需要获取A款商品去年1至12个月中每个月的销量数据、所有商品去年1至12个月中每个月的销量数据，应用管理平台110可以根据已知数据库、已知数据库中的表以及表中的列等信息，生成获取A商品销量和所有商品销量的查询指令。
管理平台110可以借助模型155的代码编译指令，将结合用户输入作为待生成的查询指令的词元(token)。此外，管理平台110可以向模型155输入解析结果以及与用户的交互内容等作为上下文信息。最终，管理平台110还可以向模型155输入数据库信息，例如数据库名称、数据库访问地址等。由此可以借助模型155生成查询指令。示例性地，查询指令可以是SQL查询语句。
如图2所示，在框203，应用管理平台110响应于确定利用查询指令得到的查询结果未满足指定条件，生成分析指令，分析指令用于对利用查询指令得到的查询结果进行分析。
在生成查询指令后，应用管理平台110可以将查询指令发给目标应用120，以使得目标应用120执行查询指令，从而得到查询结果。应用管理平台110可以对查询结果是否可以解决对任务执行请求所提出问题进行判断。判断查询结果包括可以解决或者无法解决。
结合前述用户输入是“本班级期末考试数学满分的同学名单”的示例，则应用管理平台110可以根据获取到的数据为本班级期末考试所有同学的数学成绩，那么应用管理平台110的判断查询结果为没有解决。进一步的，应用管理平台110如果判断出可以结合其他查询语句对查询结果进行优化，得出期末考试数学满分的同学名单，即可确定判断查询结果是可以解决。例如，可以利用筛选查询指令，筛选出数学成绩是满分的学员，即可得到所有数学满分的同学名单。
结合前述用户输入是“A款商品去年每个月的销量占比变化”的示例，如果获取到A款商品过往12个月每个月的销量，以及获取到所有商品过往12个月每个月的销量，应用管理平台110通过判断可以确定，只有销量数据无法解决确定销量占比变化的问题。示例性地，如果应用管理平台110判断出没有其他查询指令可以得出销量占比变化的结果，那么，即可确定判断查询结果是无法解决。
应用管理平台110如果确定利用查询指令得到的查询结果未满足指定条件，那么可以表示只利用查询指令是无法解决任务执行请求提出的问题的。对此，应用管理平台110可以引入分析指令。如果说查询指令的优势是快速获取数据，那么分析指令的优势则在于可以对数据进行复杂的计算。示例性地，占比问题、同比问题、环比问题以及数据分析问题等，均可以对应复杂的计算。在当前实施例中，对于查询指令和分析指令的代码编译可以由应用管理平台110调用模型155完成。利用模型155的代码编译能力，从而节省人工进行代码编译的过程。
结合前述用户输入是“A款商品去年每个月的销量占比变化”的示例，则分析指令可以被应用管理平台110配置为计算A商品每个月的销量占比。此外，分析指令还可以被应用管理平台110配置为根据过往12个月的销量占比，得出A商品每个月的销量占比变化。
通过对用户输入的解析，首先匹配出于解析结果对应的查询指令。其次在确定查询指令不满足于指定条件时引入分析指令，将分析指令和查询指令相结合，一方面发挥查询指令在数据查询效率方面的优势，另一方面发挥分析指令在数据分析方面应对复杂问题能力的优势，从而可以结合不同方案完成任务执行请求的响应。以此方式，能够充分发挥查询指令和分析指令各自的优势，从而完成任务处理。
在一些实施例中，应用管理平台110可以利用目标模型，基于用户输入确定任务执行请求需要多个执行步骤。响应于基于任务执行请求需要多个执行步骤，将任务执行请求拆分为多个子任务的请求，作为解析结果。
前已述及，应用管理平台110可以调用模型155，利用模型155的自然语言处理能力对用户输入进行解析。结合前述用户输入是“A款商品去年每个月的销量占比变化”的示例，应用管理平台110的解析结果为需要获取A款商品去年1至12个月每个月的销量，以及获取到所有商品去年1至12个月每个月的销量。由此，应用管理平台110可以确定任务执行请求需要多个执行步骤。如第一步骤是获取A商品去年1至12个月每个月的销量。第二步骤是获取所有商品去年1至12个月每个月的销量。如果再进行细分，应用管理平台110还可以将第一步骤细分为12个子步骤，也就是每个子步骤对应获取A商品在去年1个月中的销量。同理，应用管理平台110还可以将第二步骤细分为12个子步骤，也就是每个子步骤对应获取所有商品在去年1个月中的销量。
对于第一步骤的子步骤，应用管理平台110可以对应生成12条查询指令，每条查询指令被配置为获取A款商品1个月的销量数据。对于第二步骤的子步骤，应用管理平台110也可以生成12条查询指令，每条查询指令被配置为获取所有商品1个月的销量数据。通过将任务执行请求的拆分，可以将复杂问题拆分为多个简单问题。
在一些实施例中，应用管理平台110可以基于解析结果确定待获取数据的数据说明，数据说明用于指示待获取数据的术语和示例中的至少一项，以及利用目标模型，基于数据说明生成查询指令。
前已述及，解析可以通过应用管理平台110调用模型155完成，借助模型155的自然语言处理能力得到解析结果。一方面，解析结果是基于用户输入和模型155的自然语言处理能力完成的，因此解析结果通常是非结构化的。另一方面，以SQL查询语句为代表的查询指令是结构化的，由此可能会导致解析结果和查询指令之间存在差异。基于此，在得到解析结果后，应用管理平台110可以借助术语集展示区域302接收到的术语示例与解析结果进行匹配，从而将匹配成功的术语作为待获取数据的数据说明。
此外，应用管理平台110还可以在得到解析结果后，基于解析结果声明待获取数据的示例。待获取数据的示例可以用于指示待获取数据的数据格式、数据内容等。待获取数据的示例也可以作为待获取数据的数据说明。
在确定出待获取数据的数据说明后，模型155可以结合待获取数据的数据说明生成查询指令，可以使得查询指令更好的贴合于解析结果的数据获取需求。
在一些实施例中，在获取到查询指令得到的查询结果后，应用管理平台110还可以向通过查询指令得到的查询结果添加数据来源标识，数据来源标识用于指示查询结果在源数据中的出处。
在获取到查询指令得到的查询结果后，应用管理平台110还可以对查询指令得到的查询结果添加数据来源标识。数据来源标识用于指示查询结果在源数据中的出处。示例性地，数据来源标识可以用于指示查询指令得到的查询结果源自数据库的数据库名、数据库中的表、表中的列等信息。
添加数据来源标识的意义在于可以对数据真实性进行标注，由此可以规避掉没有出处的虚假数据。
在一些实施例中，应用管理平台110确定利用查询指令得到的查询结果未满足指定条件包括：响应于利用查询指令得到的查询结果无法解决任务执行请求所提出问题，确定原因，原因为无法解决任务执行请求所提出问题的原因；响应于原因指示查询指令的指令类别正确，对查询指令进行调整，得到调整后的查询指令，调整包括替换查询指令和增加新的查询指令中的一项；以及响应于调整满足预设的调整流程，且调整后的查询指令得到的查询结果仍无法解决任务执行请求所提出问题，确定原因指示查询指令的指令类别错误。
图4所示为任务处理方法的流程400的示意图。结合图4所示，在框401生成查询指令后，在框402，应用管理平台110需要对查询指令得到的查询结果能否解决任务执行请求所提出问题进行判定，也就是执行判断是否解决问题。判断查询结果可以包括可以解决或者无法解决。
如果判断查询结果是可以解决，那么表示生成的查询指令是正确的、无需调整的。由此在框403，可以向终端用户145进行反馈，也就是执行在交互窗口142进行反馈查询指令。反之，如果判断查询结果是无法解决，则应用管理平台110需要在框404进一步确定问题无法解决的原因，也就是执行判断原因是否是查询指令类别的错误。通常而言，问题无法解决的原因可以归类为指示查询指令的指令类别正确，但查询指令需要调整；或者问题无法解决的原因可以归类于查询指令的指令类别错误。如果确定无法解决的原因是查询指令的指令类别正确，那么可以表示当前的查询指令需要调整，也就是返回框401。调整可以包括替换或者增加新的查询指令。
以替换为示例，如果基于解析结果生成查询指令得到的结果无法解决任务执行请求所提出问题，那么应用管理平台110可以根据实际情况生成新的查询指令。利用新的查询指令代替原有的查询指令，以进行数据查询。所谓的实际情况可以是根据模型155分析的结果，也可以是根据与终端用户145交互后得到的指示。对于具体实际情况的获取过程在此不再赘述。
以增加新的查询指令为示例，结合前述用户输入是“本班级期末考试数学满分的同学名单”的示例。应用管理平台110可以根据获取到的数据为本班级期末考试所有同学的数学成绩。如果应用管理平台110确定出利用查询指令得到的查询结果无法解决任务执行请求所提出问题，并且原因指示查询指令的指令类别正确，那么可以选择增加新的查询指令。新的查询指令可以是在本班级期末考试所有同学的数学成绩中查询成绩为满分的同学。
预设的调整流程可以用于指示应用管理平台110的调整能力。例如，调整能力可以指示应用管理平台110生成新的查询指令的能力。如果生成新的查询指令的能力为可以生成m(m为正整数)个新的查询指令，但新生成的m个查询指令得到的查询结果仍然无法解决任务执行请求所提出问题，可以确定原因指示查询指令的指令类别错误。或者，调整能力可以指示调整次数。如调整n次后(n为正整数)，查询指令得到的查询结果仍然无法解决任务执行请求所提出问题，也可以确定无法解决任务执行请求所提出问题的原因为指示查询指令的指令类别错误。
在确定利用查询指令得到的查询结果无法解决任务执行请求所提出问题，且原因在于查询指令选择的不准确或者存在遗漏的场景，可以对查询指令进行调整，从而选择出更适合的查询指令。反之，如果通过对查询指令的调整仍然无法解决任务执行请求所提出问题，则可以确定为仅靠查询指令无法解决问题。由此可以触发引入分析指令，从而可以更好的完成任务执行请求。
在一些实施例中，应用管理平台110确定利用查询指令得到的查询结果未满足指定条件还包括：响应于原因指示查询指令的指令类别错误，确定利用查询指令得到的查询结果未满足指定条件。
对于原因指示查询指令的指令类别错误的情况，可以直接确定为利用查询指令得到的查询结果未满足指定条件。应用管理平台110对于指令类别错误的判定，可以借助模型155的分析能力完成。例如，可以将解析结果和查询指令得到的查询结果输入给模型155，通过模型155得到无法解决任务执行请求所提出问题的原因为指令类别错误。
在一些实施例中，应用管理平台110生成分析指令，可以包括：获取上下文信息，上下文信息包括用户输入、解析结果、查询指令、利用查询指令得到的查询结果和查询结果未满足指定条件的原因中的至少一项，以及基于上下文信息，生成分析指令。
仍然参考图4所示，在框405，应用管理平台110可以利用模型155参考上下文信息生成分析指令。示例性地，应用管理平台110可以获取上下文信息，也就是获取包括用户输入、对用户输入进行解析的解析结果、已经选用过的所有查询指令、各查询指令对应的查询结果以及查询结果未能满足指定条件的原因中的至少一项内容。
在获取到上下文信息后，利用模型155即可基于上下文信息确定分析指令。结合前述用户输入是“A款商品去年每个月的销量占比变化”的示例，在获取用户输入、解析结果、查询指令以及查询指令无法解决任务执行请求所提出问题的原因为查询指令的指令类别错误之后，应用管理平台110可以借助模型155执行分析，从而得出需要分析指令来解决问题。此外，还可以确定出分析指令可以被配置为进行占比问题的运算，以及被配置为对占比运算的结果进行分析。
不难理解，如果生成的分析指令仍然无法解决任务执行请求所提出问题，应用管理平台110可以利用模型155进行分析，从而对分析指令进行调整。调整目标为调整后的分析指令可以解决任务执行请求所提出问题。例如，对于销量占比变化的情况，第一轮分析指令可以确定出占比数据，但占比数据还无法等同于占比变化。因此还可以进行第二轮分析指令的生成，也就是根据多个统计时段的占比情况，得出占比变化的最终分析结果。
由此，可以通过多轮次分析指令进行问题的解决。仍然结合图4所示，在框406，确定出分析指令后，即可进行执行分析指令。
分析指令可以采用任意可以完成分析任务的编程语言进行代码编译。编程语言的基本语句中包含算术运算符、逻辑运算符等，因而面对如同比、环比、占比等复杂问题可以轻松应对。编程语言的代码编译过程可以借助模型155的代码编译能力完成。图5示出了根据本公开的一些实施例的展示数据结果的页面500的示例。在查询内容显示区域511，可以显示出用户输入。在数据结果显示区域512，可以显示出得到最终数据结果所涉及到的数据出处、得到最终数据结果的过程中所生成或采用的查询指令(查询指令生成)、到最终数据结果的过程中所生成或采用的分析指令(分析指令生成)，以及最终的分析结果。
本申请实施例中，在面对仅采用查询指令获取到的数据无法解决用户提出的问题的情况下，可以根据与用户的交流以及在交流过程中的事实判别结论引入分析指令。通过将分析指令与查询指令结合的方式解决用户的问题。
在一些实施例中，基于所述上下文信息确定提示信息，利用目标模型，基于所述提示信息生成所述分析指令。
获取到上下文信息后，可以基于上下文信息生成提示信息。例如，可以对上下文信息执行结构化处理，得到提示信息。结构化处理可以是首先确定关键字段，如用户输入、解析结果等。此后对不同关键字段添加对应的内容，从而得到提示信息。确定关键字段和/或添加对应内容的过程中，还可以包括文字去重、简化表述等处理。在获取到提示信息后，利用模型155即可基于提示信息确定分析指令。也就是利用模型155基于提示信息执行自然语言代码编译，得到对应的分析指令的代码。
在一些实施例中，应用管理平台110还可以执行以下步骤：将数据结果存储于指定存储区域，数据结果包括查询指令得到的查询结果和分析指令得到的分析结果中的至少一项；以及利用存储区域的地址表征数据结果。
应用管理平台110在利用查询指令或者分析指令进行数据处理获取到数据结果后，可以将数据结果存储于指定的存储区域。示例性地，指定区域可以是执行分析指令的沙箱中的缓存区域。在数据结果存储完毕后，应用管理平台110可以根据数据结果所存储的地址生成存储标识。存储标识可以用于表征数据。由此，应用管理平台110在生成新的查询指令或者生成的分析指令需要进行数据访问时，可以生成存储标识读取指令，从而利用存储标识读取指令读取指定存储区域中存储的数据结果。
通过上述过程，不同指令之间的数据访问仅通过存储标识进行，由此可以对应用管理平台110在模型155调用以进行查询指令的代码编译以及分析指令的代码编译时所涉及到的词元(token)的数据量降维。也就是说，在代码编译时，无需通过词元罗列所有待获取数据，只需利用存储区域的地址表征待读取的数据即可。在面对模型155上下文窗口数据量接收能力有限的情况下，可以大幅度降低模型的资源开销。
在一些实施例中，应用管理平台110还可以执行分析指令以得到分析结果。执行分析指令包括：将分析指令和查询结果有关的信息发送至指令执行环境，以及获取指令执行环境输出的分析结果，分析结果是基于分析指令处理查询结果得到的。
分析指令可以是在指令执行环境中运行的。示例性地，指令执行环境可以是沙箱环境、沙盒环境等。也就是说，应用管理平台110可以将分析指令对应的代码以及分析指令所需的查询结果发送至指令执行环境。结合前序示例，对于分析指令所需的查询结果，可以将查询结果对应的存储区域的地址发送至指令执行环境。从而在指令执行环境中，基于存储区域的地址获取到对应的查询结果。
在指令执行环境中运行分析指令以对查询结果执行处理，从而可以得到分析结果。分析结果可以由指令执行环境被发送至指定存储区域。在指定存储区域存储的分析结果可以是最终分析结果。或者，分析结果可以直接在应用管理平台110所呈现的用户界面显示。
图6示出了根据本公开的一些实施例的任务处理装置600的示意性结构框图。装置600例如可以被实现在或被包括在应用管理平台110中。装置600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600包括解析模块601，被配置为对接收到的用户输入进行解析，得到解析结果，用户输入指示针对目标应用的任务执行请求；查询指令生成模块602，被配置为基于解析结果生成查询指令；以及分析指令生成模块603，被配置为响应于确定利用查询指令得到的查询结果未满足指定条件，生成分析指令，分析指令用于对利用查询指令得到的查询结果进行分析。
在一些实施例中，解析模块601还被配置为可以包括：执行步骤识别子模块，该模块被配置为利用目标模型，基于用户输入确定任务执行请求需要多个执行步骤；拆分子模块，该模块被配置为响应于基于任务执行请求需要多个执行步骤，将任务执行请求拆分为多个子任务的请求，作为解析结果。
在一些实施例中，查询指令生成模块602还被配置为基于解析结果，确定待获取数据的数据说明，数据说明用于指示待获取数据的术语和示例中的至少一项；以及利用目标模型，基于数据说明生成查询指令。
在一些实施例中，装置600还包括数据来源标识添加模块，该模块被配置为向通过查询指令得到的查询结果添加数据来源标识，数据来源标识用于指示查询结果在源数据中的出处。
在一些实施例中，分析指令生成模块603还被配置为可以包括原因确定子模块，该模块被配置为响应于利用查询指令得到的查询结果无法解决任务执行请求所提出问题，确定原因，原因为无法解决任务执行请求所提出问题的原因；查询指令调整子模块，该模块被配置为响应于原因指示查询指令的指令类别正确，对查询指令进行调整，得到调整后的查询指令，调整包括替换查询指令和增加新的查询指令中的一项；以及指令类别错误判定子模块，该模块被配置为响应于调整满足预设的调整流程，且调整后的查询指令得到的查询结果仍无法解决任务执行请求所提出问题，确定原因指示查询指令的指令类别错误。
在一些实施例中，指令类别错误判定子模块还被配置为响应于原因指示查询指令的指令类别错误，确定利用查询指令得到的查询结果未满足指定条件。
在一些实施例中，分析指令生成模块603还被配置为获取上下文信息，上下文信息包括用户输入、解析结果、查询指令、利用查询指令得到的查询结果和查询结果未满足指定条件的原因中的至少一项；以及基于上下文信息，生成分析指令。
在一些实施例中，分析指令生成模块603还被配置为基于上下文信息确定提示信息，利用目标模型，基于提示信息生成分析指令。
在一些实施例中，装置600还包括数据表征模块，该模块被配置为将数据结果存储于指定存储区域，数据结果包括查询指令得到的查询结果和分析指令得到的分析结果中的至少一项；以及利用存储的地址表征数据结果。
在一些实施例中，装置600还包括分析指令运行模块，该模块被配置为：将分析指令和查询结果有关的信息发送至指令执行环境；以及获取指令执行环境输出的分析结果，分析结果是基于分析指令处理查询结果得到的。
图7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700的框图。应当理解，图7所示出的电子设备700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7所示出的电子设备700可以包括或被实现为图1的应用管理平台110，或图6的装置600。
如图7所示，电子设备700是通用电子设备的形式。电子设备700的组件可以包括但不限于一个或多个处理器710或处理单元、存储器720、存储设备730、一个或多个通信单元740、一个或多个输入设备750以及一个或多个输出设备760。处理器710可以是实际或虚拟处理器并且能够根据存储器720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700的并行处理能力。
电子设备700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700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720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730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700内被访问。
电子设备700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7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720可以包括计算机程序产品725，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740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700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700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750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760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700还可以根据需要通过通信单元740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700交互的设备进行通信，或者与使得电子设备700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，从而生成出一种机器，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 (14)

  1. 一种任务处理方法，包括：
    对接收到的用户输入进行解析，得到解析结果，所述用户输入指示针对目标应用的任务执行请求；
    基于所述解析结果生成查询指令；以及
    响应于确定利用所述查询指令得到的查询结果未满足指定条件，生成分析指令，所述分析指令用于对利用所述查询指令得到的查询结果进行分析。
  2. 根据权利要求1所述的方法，其中对接收到的用户输入进行解析包括：
    利用目标模型，基于所述用户输入确定所述任务执行请求需要多个执行步骤；以及
    响应于基于所述任务执行请求需要多个执行步骤，将所述任务执行请求拆分为多个子任务的请求，作为所述解析结果。
  3. 根据权利要求1所述的方法，其中基于所述解析结果生成查询指令包括：
    基于所述解析结果确定待获取数据的数据说明，所述数据说明用于指示所述待获取数据的术语和示例中的至少一项；以及
    利用目标模型，基于所述数据说明生成所述查询指令。
  4. 根据权利要求1所述的方法，还包括：
    向通过所述查询指令得到的查询结果添加数据来源标识，所述数据来源标识用于指示所述查询结果在源数据中的出处。
  5. 根据权利要求1所述的方法，其中确定利用所述查询指令得到的查询结果未满足指定条件包括：
    响应于利用所述查询指令得到的查询结果无法解决所述任务执行请求所提出问题，确定原因，所述原因为无法解决所述任务执行请求所提出问题的原因；
    响应于所述原因指示所述查询指令的指令类别正确，对所述查询指令进行调整，得到调整后的查询指令，所述调整包括替换查询指令和增加新的查询指令中的一项；以及
    响应于调整满足预设的调整流程，且所述调整后的查询指令得到的查询结果仍无法解决所述任务执行请求所提出问题，确定所述原因指示所述查询指令的指令类别错误。
  6. 根据权利要求5所述的方法，其中确定利用所述查询指令得到的查询结果未满足指定条件还包括：
    响应于所述原因指示所述查询指令的指令类别错误，确定利用所述查询指令得到的查询结果未满足指定条件。
  7. 根据权利要求1至6任一所述的方法，其中生成分析指令包括：
    获取上下文信息，所述上下文信息包括所述用户输入、所述解析结果、所述查询指令、利用所述查询指令得到的查询结果和所述查询结果未满足指定条件的原因中的至少一项；以及
    基于所述上下文信息，生成所述分析指令。
  8. 根据权利要求7所述的方法，其中基于所述上下文信息，生成所述分析指令包括：
    基于所述上下文信息确定提示信息；以及
    利用目标模型，基于所述提示信息生成所述分析指令。
  9. 根据权利要求1至6任一所述的方法，还包括：
    将数据结果存储于指定存储区域，所述数据结果包括查询指令得到的查询结果和所述分析指令得到的分析结果中的至少一项；以及
    利用所述存储的地址表征所述数据结果。
  10. 根据权利要求1所述的方法，还包括：
    将所述分析指令和所述查询结果有关的信息发送至指令执行环境；以及
    获取所述指令执行环境输出的分析结果，所述分析结果是基于所述分析指令处理所述查询结果得到的。
  11. 一种任务处理装置，包括：
    解析模块，被配置为对接收到的用户输入进行解析，得到解析结果，所述用户输入指示针对目标应用的任务执行请求；
    查询指令生成模块，被配置为基于所述解析结果生成查询指令；以及
    分析指令生成模块，被配置为响应于确定利用所述查询指令得到的查询结果未满足指定条件，生成分析指令，所述分析指令用于对利用所述查询指令得到的查询结果进行分析。
  12. 一种电子设备，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根据权利要求1至10中任一项的方法。
  13. 一种计算机可读存储介质，其上存储有计算机可执行指令，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10中任一项的方法。
  14. 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，计算机可执行指令在由设备执行时使设备执行根据权利要求1至10中任一项所述的方法。
PCT/CN2025/079845 2024-06-27 2025-02-28 任务处理方法、装置、设备和计算机可读存储介质 Pending WO2026001052A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202410851944.3A CN119005327B (zh) 2024-06-27 2024-06-27 任务处理方法、装置、设备和计算机可读存储介质
CN202410851944.3 2024-06-27

Publications (1)

Publication Number Publication Date
WO2026001052A1 true WO2026001052A1 (zh) 2026-01-02

Family

ID=93469834

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2025/079845 Pending WO2026001052A1 (zh) 2024-06-27 2025-02-28 任务处理方法、装置、设备和计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN119005327B (zh)
WO (1) WO2026001052A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119005327B (zh) * 2024-06-27 2025-11-21 北京字跳网络技术有限公司 任务处理方法、装置、设备和计算机可读存储介质
CN120632169A (zh) * 2025-05-30 2025-09-12 北京字跳网络技术有限公司 用户查询处理方法、装置、设备、存储介质和程序产品

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110825767A (zh) * 2019-11-15 2020-02-21 江苏满运软件科技有限公司 数据查询方法、平台、设备及存储介质
CN116894078A (zh) * 2023-07-28 2023-10-17 广东保伦电子股份有限公司 一种信息交互方法、装置、电子设备及介质
CN117033540A (zh) * 2023-09-01 2023-11-10 中国工商银行股份有限公司 报表生成方法、装置、电子设备和介质
US20240112074A1 (en) * 2022-09-30 2024-04-04 International Business Machines Corporation Natural language query processing based on machine learning to perform a task
CN117931852A (zh) * 2023-12-22 2024-04-26 润建股份有限公司 一种基于自然语言处理的智能查询和可视化方法
CN118152423A (zh) * 2024-03-01 2024-06-07 吉利汽车研究院（宁波）有限公司 智能查询方法、装置、电子设备及可读存储介质
CN118170795A (zh) * 2024-02-28 2024-06-11 北京德塔精要信息技术有限公司 数据查询方法、装置、存储介质及电子设备
CN118210619A (zh) * 2024-01-26 2024-06-18 联想(北京)有限公司 一种任务处理方法及电子设备
CN119005327A (zh) * 2024-06-27 2024-11-22 北京字跳网络技术有限公司 任务处理方法、装置、设备和计算机可读存储介质

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102315350B1 (ko) * 2015-08-27 2021-10-20 삼성전자주식회사 질의 처리 자동화 장치 및 방법
US10776700B2 (en) * 2015-09-10 2020-09-15 Senseforth, Inc. Method and system for automatic resolution of user queries
US10740373B2 (en) * 2017-02-08 2020-08-11 International Business Machines Corporation Dialog mechanism responsive to query context
CN109614400A (zh) * 2018-11-30 2019-04-12 深圳前海微众银行股份有限公司 失败任务的影响及溯源分析方法、装置、设备及存储介质
CN109739878B (zh) * 2018-12-13 2020-12-01 北京达佳互联信息技术有限公司 大数据查询方法、装置、服务器及存储介质
CN113609154B (zh) * 2021-08-06 2023-08-29 网易（杭州）网络有限公司 一种数据查询方法、装置、电子设备和存储介质
CN114647631A (zh) * 2022-03-01 2022-06-21 阿波罗智联(北京)科技有限公司 基于安全外壳协议终端的日志查询方法、装置和设备
CN118093962A (zh) * 2024-03-31 2024-05-28 苏州元脑智能科技有限公司 数据检索方法、装置、系统、电子设备及可读存储介质

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110825767A (zh) * 2019-11-15 2020-02-21 江苏满运软件科技有限公司 数据查询方法、平台、设备及存储介质
US20240112074A1 (en) * 2022-09-30 2024-04-04 International Business Machines Corporation Natural language query processing based on machine learning to perform a task
CN116894078A (zh) * 2023-07-28 2023-10-17 广东保伦电子股份有限公司 一种信息交互方法、装置、电子设备及介质
CN117033540A (zh) * 2023-09-01 2023-11-10 中国工商银行股份有限公司 报表生成方法、装置、电子设备和介质
CN117931852A (zh) * 2023-12-22 2024-04-26 润建股份有限公司 一种基于自然语言处理的智能查询和可视化方法
CN118210619A (zh) * 2024-01-26 2024-06-18 联想(北京)有限公司 一种任务处理方法及电子设备
CN118170795A (zh) * 2024-02-28 2024-06-11 北京德塔精要信息技术有限公司 数据查询方法、装置、存储介质及电子设备
CN118152423A (zh) * 2024-03-01 2024-06-07 吉利汽车研究院（宁波）有限公司 智能查询方法、装置、电子设备及可读存储介质
CN119005327A (zh) * 2024-06-27 2024-11-22 北京字跳网络技术有限公司 任务处理方法、装置、设备和计算机可读存储介质

Also Published As

Publication number Publication date
CN119005327B (zh) 2025-11-21
CN119005327A (zh) 2024-11-22

Similar Documents

Publication Publication Date Title
US10769056B2 (en) System for autonomously testing a computer system
WO2026001052A1 (zh) 任务处理方法、装置、设备和计算机可读存储介质
CN119002767B (zh) 信息处理方法、装置、设备和存储介质
EP4715585A1 (en) Method and apparatus for information processing, and device and storage medium
US20250200475A1 (en) Declarative code-less workflow using llm
WO2026040475A1 (zh) 数据处理方法、装置、设备和可读存储介质
WO2026001042A1 (zh) 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
WO2025223030A1 (zh) 信息处理方法、装置、设备和存储介质
WO2026001060A1 (zh) 信息处理方法、装置、设备和存储介质
WO2026040472A1 (zh) 应用配置方法、信息处理方法、装置、设备和存储介质
WO2025223031A1 (zh) 信息处理方法、装置、设备和存储介质
WO2026000958A1 (zh) 用于人机对话的方法、装置、设备、介质和程序产品
JP2026512322A (ja) アプリケーション作成のための方法、装置、機器及び記憶媒体
EP4517518A1 (en) Conversational digital assistant
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
CN119002745B (zh) 用于会话交互的方法、装置、设备和存储介质
US20260050746A1 (en) Message processing
US20260057007A1 (en) Method, apparatus, device, medium and program product for searching information
US20250147735A1 (en) Method, apparatus, device and storage medium for message processing
WO2026090923A1 (en) Method, system, apparatus, device, and medium for protecting sensitive information
WO2026007771A1 (zh) 用于接口处理的方法、装置、设备、存储介质和程序产品
WO2025108115A1 (zh) 数据处理的方法、装置、设备和存储介质

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: 25824447

Country of ref document: EP

Kind code of ref document: A1