Task processing method and apparatus, electronic device, and medium

Abstract

This invention provides a task processing method, apparatus, electronic device, and medium, relating to the field of data processing technology. The specific implementation includes: responding to obtaining requirement description information, performing intent recognition on the requirement description information to obtain an intent recognition result; responding to determining that the intent recognition result includes a target label, obtaining a reference task flow related to the intent recognition result, wherein the target label refers to a partial branch task function to be executed for a target scenario included in the intent recognition result, and the reference task flow includes multiple branch task flows determined for the target scenario and the execution dependencies between the multiple branch task flows; constructing a first task flow based on the target label and the reference task flow; and performing task processing based on the intent recognition result and the first task flow to obtain a first task execution result.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a task processing method, apparatus, electronic device, storage medium, and computer program product. Background Technology

[0002] Currently, to ensure the efficient execution of business scenario tasks in business systems, these tasks are typically orchestrated into task flows containing multiple operator nodes, and automated processing is achieved by executing these task flows. However, when business scenarios change, the corresponding task flows need to be re-orchestrated, resulting in high manpower and time costs and impacting task processing efficiency. Summary of the Invention

[0003] This invention provides a task processing method, apparatus, electronic device, storage medium, and computer program product.

[0004] According to one aspect of the present invention, a task processing method is provided, comprising: in response to obtaining requirement description information, performing intent recognition on the requirement description information to obtain an intent recognition result; in response to determining that the intent recognition result includes a target label, obtaining a reference task flow related to the intent recognition result, wherein the target label is used to refer to a partial branch task function to be executed for a target scenario included in the schematic diagram recognition result, and the reference task flow includes multiple branch task flows determined for the target scenario and the execution dependencies between the multiple branch task flows; constructing a first task flow according to the target label and the reference task flow; and performing task processing according to the intent recognition result and the first task flow to obtain a first task execution result.

[0005] According to another aspect of the present invention, a task processing apparatus is provided, comprising: an identification module, configured to, in response to obtaining requirement description information, perform intent identification on the requirement description information to obtain an intent identification result; an acquisition module, configured to, in response to determining that the intent identification result includes a target label, acquire a reference task flow related to the intent identification result, wherein the target label is used to refer to a partial branch task function to be executed for a target scenario included in the schematic diagram identification result, and the reference task flow includes multiple branch task flows determined for the target scenario and the execution dependencies between the multiple branch task flows; a construction module, configured to construct a first task flow based on the target label and the reference task flow; and a processing module, configured to perform task processing based on the intent identification result and the first task flow to obtain a first task execution result.

[0006] According to another aspect of the present invention, an electronic device is provided, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors perform a task processing method as described above.

[0007] According to another aspect of the present invention, a computer-readable storage medium is provided having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the task processing method described above.

[0008] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the task processing method described above. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0010] Figure 1 This is a flowchart of a task processing method according to an embodiment of the present invention;

[0011] Figure 2 These are examples of annotation information corresponding to some task instructions according to embodiments of the present invention;

[0012] Figure 3 This is a schematic diagram illustrating the process of constructing dialogue sample data according to an embodiment of the present invention;

[0013] Figure 4 This is a block diagram of a task processing apparatus according to an embodiment of the present invention;

[0014] Figure 5 This is a block diagram of an electronic device suitable for implementing a task processing method according to an embodiment of the present invention. Detailed Implementation

[0015] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments and accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0016] It should be noted that the sequence numbers of each operation in the following methods are merely for descriptive purposes and should not be construed as indicating the execution order of the operations. Unless explicitly stated otherwise, the method does not need to be executed in the exact order shown. Furthermore, in the description of this invention, terms such as "first," "second," etc. (if present) are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0017] In the technical solution of this invention, the collection, storage, use, processing, transmission, provision, disclosure, and application of data (including but not limited to user personal information) comply with the relevant laws and regulations of the relevant countries and regions, and do not violate public order and good morals. Furthermore, in the technical solution of this invention, authorization or consent from the data owner is obtained before acquiring or collecting any relevant data.

[0018] Figure 1 This is a flowchart of a task processing method according to an embodiment of the present invention.

[0019] like Figure 1 As shown, the task processing method 100 includes operations S110 to S140.

[0020] In operation S110, in response to obtaining the requirement description information, intent recognition is performed on the requirement description information to obtain the intent recognition result.

[0021] In operation S120, in response to determining that the intent recognition result includes the target label, a reference task flow related to the intent recognition result is obtained.

[0022] In operation S130, the first task flow is constructed based on the target label and the reference task flow.

[0023] In operation S140, task processing is performed based on the intent recognition result and the first task flow to obtain the first task execution result.

[0024] According to embodiments of the present invention, the requirement description information can be natural language information used to characterize the task processing requirements of the target object. The requirement description information can be obtained, for example, based on the target object's input operation to an information input box in the interactive interface; of course, other methods can also be used to obtain it. Embodiments of the present invention do not limit the method of obtaining the requirement description information.

[0025] To understand the task processing requirements of the target object, intent recognition can be further performed on the requirement description information to obtain intent recognition results. The intent recognition results characterize the target object's intent within the requirement description information, providing a more detailed intent description and thus laying the foundation for subsequent processing. In this embodiment of the invention, an intent recognition model can be used to perform intent recognition on the requirement description information. For example, the intent recognition model can be obtained by training a Large Language Model (LLM) based on dialogue sample data.

[0026] The following describes the process of constructing the dialogue sample data.

[0027] In this embodiment of the invention, pre-constructed labeled data can be used to construct dialogue sample data. This labeled data includes, for example, multiple task instructions and corresponding labeled information for each task instruction. Each task instruction is natural language information used to characterize the task processing needs of a target object. Multiple task instructions may include, for example, task instructions characterizing clear needs and task instructions characterizing ambiguous needs. The labeled information corresponding to each task instruction includes, for example, but is not limited to, the needs represented by the task instruction, a judgment on whether the needs represented by the task instruction are ambiguous, known attribute information, and missing attribute information in the task instruction. The known attribute information includes known attribute items, attribute identifiers of known attribute items, and attribute values ​​of known attribute items. The missing attribute information includes the target attribute item missing from the task instruction, the attribute identifier of the target attribute item, and at least one potential attribute value associated with the target attribute item. The attribute identifier is used to identify the importance of the attribute item, that is, the information value or keyness of the attribute item in the task instruction. For example, the attribute identifiers "required parameter" and "optional parameter" can be used to identify the importance of the attribute item. For example, you can use the attribute label "required parameter" to identify critical attribute items, and the attribute label "optional parameter" to identify general attribute items, and so on. It should be understood that attribute labels are not limited to the above examples, and can be set according to actual needs.

[0028] In this embodiment of the invention, the known and missing attribute information in each task instruction can be obtained by annotating it with reference attribute items (e.g., including key attribute items and general attribute items) that are compatible with the task instruction. The reference attribute items (e.g., including key attribute items and general attribute items) that are compatible with each task instruction can be set according to the business scenario to which the task instruction belongs.

[0029] In one example, when the business scenario to which the task instruction "xxxxx" belongs is "full-element feature interpretation," this business scenario supports initiating interpretation tasks for all feature types, including but not limited to farmland, grassland, woodland, individual buildings, and water bodies. Reference attribute items adapted to the task instruction "xxxxx" may include key attribute items "target area" and "task label," as well as general attribute items such as "image time," "band information," and "satellite source." The key attribute item "task label" indicates the task function to be executed or the feature type to be interpreted contained in the task instruction "xxxxx." Based on the aforementioned reference attribute items adapted to the task instruction "xxxxx," the known and missing attribute information in the task instruction "xxxxx" can be annotated.

[0030] In another example, when the business scenario to which the task instruction "xxx" belongs is "crop growth distribution inversion production," this business scenario supports inverting the growth status of the target crop based on observation indicators and producing a growth distribution map. Reference attribute items adapted to this task instruction "xxx" may include key attribute items such as "target area," "time information," "observation object," and "observation indicator," as well as general attribute items such as "resolution" and "derived information or analysis." Based on the above reference attribute items adapted to the task instruction "xxx," the known and missing attribute item information in the task instruction "xxx" can be annotated.

[0031] In this embodiment of the invention, the determination of whether the intended demand represented by the task instruction is ambiguous in the annotation information corresponding to each task instruction is based on whether the task instruction is missing a key attribute item. That is, if a task instruction is missing a key attribute item, the intended demand represented by the task instruction is marked as ambiguous; otherwise, the intended demand represented by the task instruction is marked as clear.

[0032] Figure 2 Examples of annotation information corresponding to some task instructions are illustrated. Based on... Figure 2 After annotating each task instruction using the processing method described above, the annotated data can be obtained. Then, based on the annotated data, dialogue sample data can be constructed. The dialogue sample data can be dialogues between target objects and a large language model. The following section combines... Figure 3 The process of constructing the dialogue sample data is explained. It should be understood that... Figure 3 The contents shown are merely illustrative, and the embodiments of the present invention are not limited thereto.

[0033] like Figure 3As shown, for task instruction 311, "I want to interpret cultivated land and grassland. The image source should preferably be GF6. Since the objects to be interpreted are all vegetation objects, four-band imagery should be used for interpretation," the initial thought 301 is to construct a large language model using the annotation information 331 corresponding to task instruction 311 in the pre-constructed annotation data 330. This is to clarify whether the demand intention represented by task instruction 311 is ambiguous, and if the demand intention represented by task instruction 311 is ambiguous, to identify the missing target attribute items in task instruction 311, the attribute identifiers of the target attribute items, and to provide at least one potential attribute value associated with the target attribute items.

[0034] like Figure 3 As shown, based on preliminary consideration 301, the intended demand represented by task instruction 311 is ambiguous. The missing target attribute items include the mandatory parameter "target area" and the optional parameter "image time". Preliminary consideration 301 also provides at least one potential attribute value associated with each target attribute item. For example, the potential attribute values ​​associated with the mandatory parameter "target area" are "Area A", "Area B", and "Area C", etc., and the potential attribute values ​​associated with the optional parameter "image time" are "2025", "third quarter of 2025", and "May 2025", etc.

[0035] Next, based on the initial thinking 301, the task interaction information 321 is generated using the large language model.

[0036] like Figure 3 As shown, task interaction information 321 represents a query for all missing target attribute items (including the required parameter "target area" and the optional parameter "image time") in task instruction 311, and provides at least one potential attribute value associated with each target attribute item. These potential attribute values ​​can provide corresponding prompts to the target object, guiding the target object to provide task response information that is more suitable for task interaction information 321, thereby helping to reduce the number of dialogue interactions and improve the efficiency of obtaining missing information.

[0037] Next, the task response information 312 of the target object in response to the task interaction information 321 is obtained. The task response information 312 may include the target attribute values ​​corresponding to each of the target attribute items. In some embodiments, the task response information 312 may also include additional attribute values ​​for the known attribute items in the task instruction 311. The additional attribute values ​​of the known attribute items can be used to correct or supplement the attribute values ​​of the known attribute items, so that the corrected or supplemented attribute values ​​of the known attribute items better match the current intent of the target object.

[0038] like Figure 3As shown, the task response information 312 provided by the target object, "I need to identify the cultivated land and forest land in region A in December 2023," not only provides target attribute values ​​corresponding to all missing target attribute items, such as the target attribute value of "target area" being "region A" and the target attribute value of "image time" being "December 2023," but also provides additional attribute values ​​"cultivated land and forest land" for the known attribute item "task label." The additional attribute value "cultivated land and forest land" is used to correct the attribute value "cultivated land and grassland" of the known attribute item "task label" in task instruction 311, so that the corrected attribute value of the known attribute item "task label" is adapted to the target object's current intent.

[0039] It should be noted that queries for all missing target attribute items in the task instructions can also be achieved through multi-turn dialogues. Each turn of the dialogue can also query and provide prompts for only one target attribute item.

[0040] Next, we will use task instructions 311, task interaction information 321, and task response information 312 to construct a summary of our thoughts.

[0041] like Figure 3 As shown, for example, key information from task instructions 311, task interaction information 321, and task response information 312 (such as target attribute values ​​corresponding to all missing target attribute items, attribute values ​​of known attribute items, and additional attribute values ​​of known attribute items) can be used to construct summary reflection 302. Summary reflection 302 summarizes all the information collected so far (such as target area, image time, and task tags) and explains why the information collected so far is sufficient to clarify the intent of the target object.

[0042] Next, the large language model is used to generate intent recognition results 322 based on summary thinking 302.

[0043] like Figure 3 As shown, the intent recognition result 322, "Identify cultivated land and forest land in area A in December 2023. The image source is GF6, and the four-band image is used for interpretation," contains sufficient and necessary information to more comprehensively, accurately, and clearly represent the target object's needs and intentions.

[0044] After constructing multiple dialogue sample data using the methods described above, the large language model can be further pre-trained using these sample data. This allows the trained intent recognition model to determine whether the intent represented by the demand description information is ambiguous based on the natural language expression. If the intent is ambiguous, it identifies missing target attributes in the demand description and uses at least one information interaction between the intent recognition model and the target object to supplement and summarize the attribute values ​​of the missing attributes, thus obtaining an intent recognition result that represents a clear demand intent. Furthermore, if the intent recognition model processes the demand description information and determines that the demand intent represented by the demand description information is clear (i.e., the demand description information does not lack key attributes), then the demand description information can be used as the intent recognition result for subsequent processing.

[0045] For example, suppose the target object inputs a demand description of "extracting roads and building areas". Processing this demand description using the aforementioned intent recognition model, it can be determined that the demand intent represented by this description is ambiguous. Then, through at least one information interaction between the target object and the intent recognition model, the attribute values ​​of the missing target attribute items in the demand description are supplemented, thereby obtaining an intent recognition result that represents a clear demand intent, such as "using sub-meter GF imagery from May 2025 to extract roads and building areas in region C, with three bands."

[0046] In this embodiment of the invention, an intent recognition model is used to identify the intent of the demand description information. After obtaining the intent recognition result, it can be further determined whether the intent recognition result contains a target label. The target label refers to the partial branch task function to be executed for the target scenario contained in the schematic diagram recognition result. The target scenario refers to a business scenario that includes multiple branch tasks, such as including but not limited to the "full-element ground feature interpretation" business scenario and the "change detection" business scenario. Taking the above intent recognition result "using the sub-meter level GF image from May 2025 to extract the road and building area in region C, with three band information" as an example, the intent recognition result includes the target label "road and building area". This target label "road and building area" indicates that in the "full-element ground feature interpretation" business scenario, the road interpretation branch task and the building unit interpretation branch task need to be executed.

[0047] In some embodiments, if it is determined that the intent recognition result includes a target label, a reference task flow related to the intent recognition result can be obtained, and a first task flow matching the intent recognition result can be constructed based on the target label and the reference task flow.

[0048] According to an embodiment of the present invention, the reference task flow is a task flow applicable to a target scenario. The reference task flow includes multiple branch task flows determined for the target scenario and the execution dependencies between these branch task flows. Each branch task flow includes multiple operator nodes and the execution dependencies between these operator nodes. Each branch task flow is used to execute one branch task.

[0049] When the intent recognition result contains a target label, since the target label indicates the partial branch task function to be executed for the target scenario contained in the intent recognition result, a corresponding first task flow can be constructed based on the target label and the reference task flow related to the intent recognition result, thereby realizing the reuse of the reference task flow under the target scenario. This approach not only improves the efficiency of constructing task flows adapted to the intent recognition result but also avoids resource waste caused by repeatedly arranging task flows with the same processing logic, effectively improving resource utilization.

[0050] After obtaining the first task flow, further task processing can be performed based on the intent recognition result and the first task flow to obtain the first task execution result. The first task execution result may contain information that matches the demand intent represented by the intent recognition result.

[0051] According to the technical solution of the present invention, after obtaining the requirement description information, intent recognition is performed on the requirement description information to obtain the intent recognition result. The intent recognition result provides a more detailed intent description, which lays the foundation for accurate subsequent processing. Next, if the intent recognition result includes a target label, a corresponding first task flow is constructed based on the target label and a reference task flow related to the intent recognition result. This enables the reuse of the reference task flow in the target scenario, improves the efficiency of constructing a task flow adapted to the intent recognition result, and avoids resource waste caused by repeatedly arranging task flows with the same processing logic. Therefore, task processing is performed based on the first task flow and the intent recognition result, which can improve the efficiency and accuracy of task processing.

[0052] According to an embodiment of the present invention, in the above operation S130, constructing the first task flow based on the target label and the reference task flow includes the following steps.

[0053] First, based on the target label, select the corresponding operator node from the reference task flow to obtain multiple target operator nodes.

[0054] The process of obtaining multiple target operator nodes is explained below.

[0055] First, based on the target label, the baseline operator node is determined from the reference task flow. The baseline operator node is related to the target label.

[0056] For example, each operator node in the reference task flow has corresponding operator attribute information, such as, but not limited to, operator identifier, version information, category information, operator function, application scenarios supported by the operator node, and other descriptive information related to the operator node. Based on the key information contained in the target label, an operator node whose operator attribute information matches that key information can be selected from the reference task flow as the baseline operator node.

[0057] Next, taking the baseline operator node as the current operator node, query the reference task flow for operator nodes that have execution dependencies on the current operator node.

[0058] Since the reference task flow reflects the execution dependencies between each operator node, it is possible to trace the operator nodes that have execution dependencies on the base operator node based on these dependencies. For example, taking the base operator node as the current operator node, one can query the downstream operator nodes that receive the output data of the current operator node and the upstream operator nodes that provide input data to the current operator node from the reference task flow, thereby obtaining the operator nodes that have execution dependencies on the current operator node.

[0059] Next, the queried operator node is returned as the new current operator node to execute the query, until there is no operator node in the reference task flow that has an execution dependency relationship with the current operator node.

[0060] Next, the baseline operator node and all operator nodes obtained from the query that have execution dependencies on the current operator node are identified as multiple target operator nodes.

[0061] Next, based on the execution dependencies between multiple target operator nodes in the reference task flow, the multiple target operator nodes are combined to obtain the first task flow.

[0062] For example, based on the execution dependencies between multiple target operator nodes in the reference task flow, multiple target operator nodes are connected by connecting lines to obtain the first task flow.

[0063] According to an embodiment of the present invention, in the above operation S140, the task processing based on the intent recognition result and the first task flow to obtain the first task execution result includes: extracting task configuration parameters from the intent recognition result, generating a first target task flow based on the task configuration parameters and the first task flow, and then executing the first target task flow to obtain the first task execution result.

[0064] In this embodiment of the invention, task configuration parameters extracted from the intent recognition result can be assigned to the corresponding target operator nodes in the first task flow to update the operator parameters of the corresponding target operator nodes. This ensures that the updated operator parameters are adapted to the current task requirements, thereby obtaining a first target task flow that matches the intent recognition result. By executing the first target task flow, a more accurate first task execution result can be obtained.

[0065] In some embodiments, before executing the first target task flow, the task processing method further includes: displaying the first target task flow, responding to obtaining adjustment information for the first target task flow, adjusting the first target task flow based on the adjustment information to obtain a second target task flow, and executing the second target task flow to obtain a second task execution result.

[0066] According to an embodiment of the present invention, before executing the first target task process, the first target task process can be displayed through an interactive interface for the target object to confirm, thereby improving the matching degree between the first target task process and the true intention of the target object, and thus enabling the obtained task execution result to truly meet the needs of the target object.

[0067] According to embodiments of the present invention, the target object can also input adjustment information for the first target task flow through an interactive interface. The adjustment information may be, for example, natural language information instructing the deletion, modification, addition, or replacement of operator parameters of target operator nodes in the first target task flow. Of course, embodiments of the present invention are not limited to this; the adjustment information may also be, for example, natural language information instructing the correction of execution dependencies between target operator nodes in the first target task flow. After obtaining the adjustment information input by the target object, the electronic device can adjust the first target task flow based on the adjustment information to obtain a second target task flow that matches the target object's current needs and intentions, and then obtain a second task execution result that matches the target object's current needs and intentions based on the second target task flow.

[0068] In some embodiments, after obtaining the execution result of the first task or the execution result of the second task, feedback information can be generated based on the execution result of the first task or the execution result of the second task. For example, the first execution result or the execution result of the second task can be used as feedback information, or a preset configuration template can be filled based on the first execution result or the execution result of the second task to obtain feedback information. Of course, other methods can also be used to determine the feedback information, and this invention does not limit this. Then, the feedback information can be displayed to the target object. In some embodiments, identifiers can also be added to each execution result and the corresponding target task process and stored in a database for later use.

[0069] According to an embodiment of the present invention, after obtaining the intent recognition result, if it is determined that the intent recognition result does not contain a target label, a scene service matching the intent recognition result can be invoked to obtain the corresponding scene task flow. Then, task processing is performed based on the intent recognition result and the scene task flow.

[0070] In this embodiment of the invention, based on the intent recognition result, a scene service matching the intent recognition result can be recalled from the scene service set, and the scene service matching the intent recognition result can be invoked to obtain the corresponding scene task flow. This scene task flow refers to a task flow that does not contain multiple branch tasks and is typically suitable for executing a single scene task. The scene task flow may include, for example, multiple operator nodes and execution dependencies between these operator nodes. Then, the task configuration parameters extracted from the intent recognition result are assigned to the corresponding operator nodes in the scene task flow to update the operator parameters of the corresponding operator nodes, thereby obtaining an updated scene task flow. By executing the updated scene task flow, the corresponding task execution result is obtained.

[0071] Figure 4 This is a block diagram of a task processing apparatus according to an embodiment of the present invention.

[0072] like Figure 4 As shown, the task processing device 400 includes an identification module 410, an acquisition module 420, a construction module 430, and a processing module 440.

[0073] The recognition module 410 is used to perform intent recognition on the requirement description information in response to obtaining the requirement description information, and obtain the intent recognition result.

[0074] The acquisition module 420 is used to acquire a reference task flow related to the intent recognition result in response to determining that the intent recognition result includes a target label. The target label is used to indicate the branch task function to be executed for the target scenario. The reference task flow includes multiple branch task flows determined for the target scenario and the execution dependencies between the multiple branch task flows.

[0075] Module 430 is used to build the first task flow based on the target label and the reference task flow.

[0076] The processing module 440 is used to process the task based on the intent recognition result and the first task flow to obtain the first task execution result.

[0077] It should be noted that the implementation methods, technical problems solved, functions achieved, and technical effects of each module in the device embodiment are the same as or similar to the implementation methods, technical problems solved, functions achieved, and technical effects of each corresponding step in the method embodiment, and will not be repeated here.

[0078] Figure 5 A block diagram of an electronic device suitable for implementing a task processing method according to an embodiment of the present invention is shown schematically.

[0079] like Figure 5 As shown, an electronic device 500 according to an embodiment of the present invention includes a processor 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage portion 508 into a random access memory (RAM) 503. The processor 501 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 501 may also include onboard memory for caching purposes. The processor 501 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0080] RAM 503 stores various programs and data required for the operation of electronic device 500. Processor 501, ROM 502, and RAM 503 are interconnected via bus 504. Processor 501 executes various operations of the method flow according to embodiments of the present invention by executing programs in ROM 502 and / or RAM 503. It should be noted that the programs may also be stored in one or more memories other than ROM 502 and RAM 503. Processor 501 may also execute various operations of the method flow according to embodiments of the present invention by executing programs stored in said one or more memories.

[0081] According to an embodiment of the present invention, the electronic device 500 may further include an input / output (I / O) interface 505, which is also connected to a bus 504. The electronic device 500 may also include one or more of the following components connected to the I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 510 as needed so that computer programs read from it can be installed into the storage section 508 as needed.

[0082] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the task processing method according to embodiments of the present invention.

[0083] According to embodiments of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of the present invention, the computer-readable storage medium may include ROM 502 and / or RAM 503 and / or one or more memories other than ROM 502 and RAM 503 described above.

[0084] Embodiments of the present invention also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement the task processing method provided by the embodiments of the present invention.

[0085] When the computer program is executed by the processor 501, it performs the functions defined in the system / apparatus of this invention. According to embodiments of the invention, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0086] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 509, and / or installed from a removable medium 511. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0087] In such an embodiment, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by processor 501, it performs the functions defined in the system of this embodiment of the invention. According to embodiments of the invention, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0088] According to embodiments of the present invention, program code for executing the computer programs provided in the embodiments of the present invention can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0089] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0090] Those skilled in the art will understand that the features described in the various embodiments of the present invention can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, the features described in the various embodiments of the present invention can be combined and / or combined in various ways without departing from the spirit and teachings of the present invention. All such combinations and / or combinations fall within the scope of the present invention.

[0091] The embodiments of the present invention have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of the invention. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of the invention, and all such substitutions and modifications should fall within the scope of the invention.

Claims

1. A task processing method, comprising: In response to obtaining the requirement description information, the requirement description information is used to perform intent recognition to obtain the intent recognition result; In response to determining that the intent recognition result includes a target label, a reference task flow related to the intent recognition result is obtained, wherein the target label is used to indicate a partial branch task function to be executed for a target scenario contained in the intent recognition result, and the reference task flow includes multiple branch task flows determined for the target scenario and the execution dependencies between the multiple branch task flows; Based on the target label and the reference task flow, a first task flow is constructed; Based on the intent recognition result and the first task flow, the task is processed to obtain the first task execution result.

2. The method according to claim 1, wherein, Each of the aforementioned branch task processes includes multiple operator nodes and the execution dependencies between these multiple operator nodes; The step of constructing the first task flow based on the target label and the reference task flow includes: Based on the target label, select the corresponding operator node from the reference task flow to obtain multiple target operator nodes; Based on the execution dependencies between multiple target operator nodes in the reference task flow, the multiple target operator nodes are combined to obtain the first task flow.

3. The method according to claim 2, wherein, The step of selecting corresponding operator nodes from the reference task flow based on the target label to obtain multiple target operator nodes includes: Based on the target label, a baseline operator node is determined from the reference task flow, the baseline operator node being related to the target label; Using the baseline operator node as the current operator node, query the reference task flow for operator nodes that have execution dependencies on the current operator node; The queried operator node is returned as the new current operator node to execute the query until there are no operator nodes in the reference task flow that have an execution dependency relationship with the current operator node. The baseline operator node and all operator nodes that have execution dependencies on the current operator node are identified as multiple target operator nodes.

4. The method according to any one of claims 1 to 3, wherein, The step of processing the task based on the intent recognition result and the first task flow to obtain the first task execution result includes: Extract task configuration parameters from the intent recognition result, and generate a first target task flow based on the task configuration parameters and the first task flow; The first target task process is executed to obtain the first task execution result.

5. The method according to claim 4, further comprising, before executing the first target task process: Show the flow of the first target task; In response to obtaining adjustment information for the first target task process, the first target task process is adjusted based on the adjustment information to obtain the second target task process; The second target task process is executed to obtain the second task execution result.

6. The method according to any one of claims 1 to 5, further comprising: In response to determining that the intent recognition result does not contain the target label, the scene service that matches the intent recognition result is invoked to obtain the corresponding scene task flow; Task processing is performed based on the intent recognition results and the scenario task flow.

7. A task processing apparatus, comprising: The identification module is used to perform intent recognition on the requirement description information in response to obtaining the requirement description information, and obtain the intent recognition result. The acquisition module is configured to, in response to determining that the intent recognition result includes a target label, acquire a reference task flow related to the intent recognition result, wherein the target label is used to indicate a partial branch task function to be executed for a target scenario contained in the intent recognition result, and the reference task flow includes multiple branch task flows determined for the target scenario and the execution dependencies between the multiple branch task flows; A construction module is used to construct a first task flow based on the target label and the reference task flow; The processing module is used to perform task processing based on the intent recognition result and the first task flow to obtain the first task execution result.

8. An electronic device, comprising: One or more processors; Memory, used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors perform the task processing method according to any one of claims 1 to 6.

9. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the task processing method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program that, when executed by a processor, implements the task processing method according to any one of claims 1 to 6.

Images