Task allocation method and device, equipment and storage medium

Abstract

The invention provides a task allocation method and device, equipment and a storage medium, and relates to the technical field of computers. The method comprises the following steps: dividing a fault processing scene based on a business process to obtain a sub-scene set containing a plurality of sub-scenes, dividing a tool for processing the sub-scenes based on the sub-scenes to obtain a plurality of task sets corresponding to the sub-scenes, tool identifiers corresponding to the tools, and processing the sub-scenes according to the task sets. The tool matched with the task data is determined in the sub-scene set and the task set through the pre-trained intelligent model and the task data, and the accuracy of the tool matched with the task data is improved.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to a task allocation method, apparatus, device, and storage medium. Background Technology

[0002] With the development of the times, the application of artificial intelligence (AI) is becoming increasingly widespread. As AI advances, related technologies enable AI to perform multi-step operations through closed-loop task execution, supporting multimodal perception and operation in physical or digital environments. However, current AI can only complete tasks based on question-and-answer methods. It is unable to execute tasks in complex environments. How to improve the ability of AI to execute tasks in complex scenarios is a pressing problem to be solved in this field. Summary of the Invention

[0003] This disclosure provides a task allocation method, apparatus, device, and storage medium, which improves the efficiency of task execution to at least a certain extent.

[0004] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part from practice of this disclosure.

[0005] According to one aspect of this disclosure, a task allocation method is provided, comprising: Based on the business process, the fault handling scenario is divided into a set of sub-scenarios containing multiple sub-scenarios; The tools used to process sub-scenes are divided based on the sub-scenes, resulting in multiple task sets corresponding to the sub-scenes, with each tool having a corresponding tool identifier; A tool that uses pre-trained intelligent models and task data to determine matching task data in sub-scene sets and task sets.

[0006] In one embodiment of this disclosure, a tool for determining matching task data in a sub-scene set and a task set using a pre-trained intelligent model and task data includes: The sub-scenes that match the task data are determined by using the first self-attention network contained in the intelligent model and the task data. The tool that matches the sub-scene is determined by the sub-scene and the second self-attention network contained in the intelligent model.

[0007] In one embodiment of this disclosure, the method further includes: The tool that matches task data and the sub-scenes that match the task data are input into the sorting intelligent model to obtain the sorted tool.

[0008] In one embodiment of this disclosure, determining a sub-scene matching the task data using a first self-attention network included in the intelligent model and the task data includes: The task data and sub-scenes are vector-encoded through the input layer of the first self-attention network to obtain the encoded task data and sub-scene data. The encoded task data and sub-scene data are transformed by the fully connected layer of the first self-attention network to obtain the transformed task data and sub-scene data. The sub-scenes that match the task data are determined by using an attention probability calculation mechanism, transformed task data, and sub-scene data.

[0009] In one embodiment of this disclosure, determining a tool that matches a sub-scene using a sub-scene and a second self-attention network included in the intelligent model includes: The sub-scene and the tool are vector-encoded through the input layer of the second self-attention network to obtain the encoded sub-scene and the tool. The encoded sub-scene and tool are transformed by the fully connected layer in the second self-attention network to obtain the transformed sub-scene and tool. The tool that matches the sub-scene is determined by using an attention probability calculation mechanism, the transformed sub-scene, and the tool itself.

[0010] In one embodiment of this disclosure, the method further includes: Acquire training samples, which include historical task data, historical sub-scenes, and historical tools; Input historical task data into the intelligent model to be trained to obtain the sub-scenes and tools output by the intelligent model to be trained. The loss function value is determined by the similarity between the sub-scene output by the intelligent model to be trained and the historical sub-scene, and the similarity between the tool output by the intelligent model and the historical tool. The pre-trained intelligent model is obtained when the loss function value converges.

[0011] In one embodiment of this disclosure, the ranking intelligence model includes an input prompt pre-trained language model.

[0012] According to another aspect of this disclosure, a task allocation apparatus is provided, comprising: The first partitioning module is used to partition fault handling scenarios based on business processes, resulting in a set of sub-scenarios containing multiple sub-scenarios. The second partitioning module is used to partition the tools used to process sub-scenes based on the sub-scenes, resulting in multiple task sets corresponding to the sub-scenes, with each tool corresponding to a tool identifier; The first determining module is a tool used to determine matching task data in the sub-scene set and task set using pre-trained intelligent models and task data.

[0013] In one embodiment of this disclosure, the first determining module includes: The first determining submodule is used to determine the sub-scene that matches the task data through the first self-attention network contained in the intelligent model and the task data; The second determination submodule is used to determine the tool that matches the sub-scene through the sub-scene and the second self-attention network contained in the intelligent model.

[0014] In one embodiment of this disclosure, the apparatus further includes: The input module is used to input the tool that matches the task data and the sub-scenes that match the task data into the sorting intelligent model to obtain the sorted tool.

[0015] In one embodiment of this disclosure, the first determining submodule includes: The first encoding unit is used to perform vector encoding on the task data and sub-scenes through the input layer of the first self-attention network to obtain the encoded task data and sub-scene data. The first transformation unit is used to perform feature transformation on the encoded task data and sub-scene data through the fully connected layer of the first self-attention network to obtain the transformed task data and sub-scene data. The first determining unit is used to determine the sub-scene that matches the task data through the attention probability calculation mechanism, the transformed task data, and the sub-scene data.

[0016] In one embodiment of this disclosure, the second determining submodule includes: The second encoding unit is used to perform vector encoding on the sub-scene and the tool through the input layer of the second self-attention network to obtain the encoded sub-scene and the tool. The second transformation unit is used to perform feature transformation on the encoded sub-scene and tool through the fully connected layer in the second self-attention network to obtain the transformed sub-scene and tool. The second determining unit is used to determine the tool that matches the sub-scene by using an attention probability calculation mechanism, the transformed sub-scene, and the tool.

[0017] In one embodiment of this disclosure, the apparatus further includes: The acquisition module is used to acquire training samples, which include historical task data, historical sub-scenes, and historical tools. The input module is used to input historical task data into the intelligent model to be trained, and obtain the sub-scenes and tools output by the intelligent model to be trained. The second determining module is used to determine the loss function value based on the similarity between the sub-scene output by the intelligent model to be trained and the historical sub-scene, and the similarity between the tool output by the intelligent model and the historical tool. The third determination module is used to obtain the pre-trained intelligent model in response to the convergence of the loss function value.

[0018] In one embodiment of this disclosure, the ranking intelligence model includes an input prompt pre-trained language model.

[0019] According to another aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the above-described task allocation method by executing the executable instructions.

[0020] According to another aspect of this disclosure, a computer-readable storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the task allocation method described above.

[0021] The task allocation method, apparatus, device, and storage medium provided in the embodiments of this disclosure divide the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios. The tools used to handle the sub-scenarios are divided based on the sub-scenarios to obtain a set of tasks corresponding to multiple sub-scenarios. Each tool corresponds to a tool identifier. By using a pre-trained intelligent model and task data, the tool that matches the task data is determined in the set of sub-scenarios and the set of tasks, thereby improving the accuracy of matching the tool with the task data.

[0022] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0023] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0024] Figure 1 This diagram illustrates a task allocation system structure according to an embodiment of the present disclosure. Figure 2 This diagram illustrates a task allocation method according to an embodiment of the present disclosure. Figure 3 This diagram illustrates another task allocation method in an embodiment of the present disclosure. Figure 4 This illustration shows a flowchart of another task allocation method in an embodiment of the present disclosure; Figure 5 This diagram illustrates a flowchart of yet another task allocation method according to an embodiment of the present disclosure; Figure 6This diagram illustrates a flowchart of yet another task allocation method according to an embodiment of the present disclosure; Figure 7 This diagram illustrates an interaction diagram of yet another task allocation method in an embodiment of this disclosure. Figure 8 This diagram illustrates an apparatus for yet another task allocation method according to an embodiment of the present disclosure. Figure 9 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0025] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0026] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0027] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0028] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0029] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0030] It should be noted that, where there is no conflict, the embodiments of this disclosure and the technical features in the embodiments can be combined with each other.

[0031] The specific implementation of the embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0032] Figure 1 A schematic diagram of a task allocation system structure according to an embodiment of the present disclosure is shown. This system can apply the task allocation method or task allocation device of various embodiments of the present disclosure.

[0033] like Figure 1 As shown, the task allocation system 10 may include a user terminal 101 and a server 102 for configuring intelligent models. The user terminal 101 and server 102 may reside on two different devices. The user terminal 101 may be a module on an electronic device with data acquisition capabilities, such as a recording device with sound collection capabilities, a photographic device with image collection capabilities, or a computer with text information collection capabilities. The server 102 may be a module on an electronic device with processing capabilities, such as a computer. The user terminal 101 and server 102 may also reside on the same device. For example, the user terminal 101 and server 102 may be an input module and a processing module on a computer or mobile phone.

[0034] A communication connection is established between the user terminal 101 and the server 102 via a network, which can be a wired network or a wireless network.

[0035] Optionally, the aforementioned wireless or wired networks use standard communication technologies and / or protocols. The network is typically the Internet, but can also be any network, including but not limited to Local Area Networks (LANs), Metropolitan Area Networks (MANs), Wide Area Networks (WANs), mobile, wired or wireless networks, private networks, or any combination of virtual private networks. In some embodiments, technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc., are used to represent data exchanged over the network. Furthermore, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Networks (VPNs), and Internet Protocol Security (IPsec) can be used to encrypt all or some links. In other embodiments, custom and / or dedicated data communication technologies can be used to replace or supplement the aforementioned data communication technologies.

[0036] The following describes the case where user terminal 101 and server 102 are located on two different devices.

[0037] User terminal 101 can be a terminal device, which can be various electronic devices, including but not limited to smartphones, tablets, laptops, desktop computers, wearable devices, augmented reality devices, virtual reality devices, etc.

[0038] Optionally, the client for the application installed on different terminal devices can be the same, or the client for the same type of application based on different operating systems. Depending on the terminal platform, the specific form of the application client can also differ; for example, the application client can be a mobile client, a PC client, etc.

[0039] Server 102 can be a server that provides various services, such as a backend management server that supports the operation of devices by users using terminal devices. The backend management server can analyze and process received requests and other data, and feed the processing results back to the terminal device.

[0040] Optionally, the server can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0041] Those skilled in the art will know that Figure 1 The number of user terminals and servers shown is merely illustrative; any number of user terminals and servers can be used depending on actual needs. This disclosure does not limit this.

[0042] To address the aforementioned problems, embodiments of this disclosure provide a task allocation method, apparatus, device, and storage medium.

[0043] Figure 2 A flowchart of a task allocation method according to an embodiment of this disclosure is shown.

[0044] like Figure 2 As shown, the method may include: S210 divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios.

[0045] In some embodiments, a fault handling scenario may include a scenario based on the fault handling environment, the required data, the actions to be performed, and / or the determination of the handling result.

[0046] In some embodiments, a sub-scenario may include a scenario encompassed by a fault handling scenario. The method for dividing sub-scenarios may be the same as the method for determining fault handling scenarios, and will not be described in detail here.

[0047] For example, a fault handling scenario may include a wireless fault handling scenario. Sub-scenarios may include information extraction, data self-service, data query, root cause query, solution query, and fault self-check query.

[0048] S220 divides the tools used to process sub-scenes based on the sub-scenes to obtain multiple task sets corresponding to the sub-scenes, and the tools are identified by tool identifiers.

[0049] In some embodiments, the task set may include a collection of tools required for the corresponding scenario. For example, the task set may include a collection of tools including address extraction, cell ID / cell extraction, time extraction, base station ID / base station extraction, performance metric extraction, and alarm metric extraction. It may also include a collection of tools including address mapping, base station mapping, and time parsing. Furthermore, it may include a collection of tools including performance metric query, alarm metric query, and engineering parameter audit query.

[0050] For example, the table below shows the correspondence between sub-scenes and sets of tools.

[0051]

[0052] In some implementations, tool identification may include the tool's name, tool attributes, and tool parameters.

[0053] In some embodiments, the tool may include programs, software, and relevant data available for processing sub-scenes.

[0054] S230 is a tool that uses pre-trained intelligent models and task data to determine matching task data in sub-scene sets and task sets.

[0055] In some embodiments, the task assignment of the wireless fault handling agent can be discretized into a standard Markov Decision Process (MDP). The actions during the task allocation phase at time t The space consists of selectable sub-scene action spaces. and tool action space The policy network selects the appropriate sub-scene action based on the state at time t. Then select the appropriate tool action based on the sub-scene. .state Represents the real-time state, which is determined by the state of the sub-scene. and the state of tools composition The task allocation process t is a selection of a certain action. The new status has been updated to Reward function The reward consists of two parts: sub-scene and tool selection rewards. The reward is based on the accuracy of the selection. Sub-scene selection accounts for a higher proportion, with the sub-scene reward accounting for 0.65% and the tool selection reward accounting for 0.35%.

[0056] The tool that matches the task data in the sub-scene set and task set can be determined by inputting the task data, sub-scene set and task set into the intelligent model, obtaining the tool identifier output by the intelligent model, and determining the corresponding tool by the tool identifier output.

[0057] The task allocation method provided in the embodiments of this disclosure divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios. The tools used to handle the sub-scenarios are divided based on the sub-scenarios to obtain a set of tasks corresponding to multiple sub-scenarios. Each tool is identified by a tool identifier. By using a pre-trained intelligent model and task data, the tool that matches the task data is determined in the set of sub-scenarios and the set of tasks, thereby improving the accuracy of matching the tool with the task data.

[0058] Figure 3 A flowchart of another task allocation method in an embodiment of this disclosure is shown.

[0059] like Figure 3 As shown, the method may include: S310 divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios; S320 divides the tools used to process sub-scenes based on the sub-scenes, resulting in multiple task sets corresponding to the sub-scenes, with each tool having a corresponding tool identifier.

[0060] The S330 uses a first self-attention network included in the intelligent model and task data to determine sub-scenes that match the task data.

[0061] In some embodiments, the intelligent model may include an intelligent model with a two-layer self-attention network. The intelligent model can encode task data, sub-scenes, and tools separately using the two self-attention networks. For example, task data, sub-scenes, and tools can be encoded separately based on an encoding function. The encoding function can be: a task data encoding function. Sub-scene encoding function Tool encoding function Where S represents the vector dimension. Task data vector encoding can be represented as..., and sub-scene vector encoding can be represented as... The tool vector encoding can be represented as .

[0062] For example, the patent uses a Chinese vector encoding method based on chinese-bert-wwm, with a vector dimension of 768. In the first self-attention mechanism layer of the Actor network, the task data and sub-scenes are first vector-encoded, and then generated through a fully connected layer. and ,in and Representing the q-value matrix and k-value matrix in the self-attention mechanism, the probability of sub-scene selection is calculated. The action at time t is selected, i.e., the i-th sub-scene is chosen. A score of 1 is given when the correct sub-scene is selected, and 0 is given otherwise. Next, the selected sub-scene and tool are vector-encoded. Similar to the first layer of self-attention mechanism, the sub-scene and tool vectors from a fully connected layer are used as the q-value matrix and k-value matrix in the second layer of self-attention mechanism to calculate the probability of tool selection. The system outputs multiple tools based on probability and determines the reward score through a value function. For example, when the query question is "Help me query the VONR uplink packet loss rate alarm details for cell 123-c?", the first-layer self-attention network (sub-scene selection strategy) selects the "data query" sub-scene at a certain time t. The second-layer self-attention network (tool selection strategy) selects "alarm indicator query" from "alarm indicator query", "performance indicator query", and "engineering parameter audit query", and receives 1 reward score.

[0063] S340 is a tool that determines the matching of a sub-scene through a sub-scene and a second self-attention network contained in the intelligent model.

[0064] In some embodiments, a fully connected layer can be set after two self-attention networks. The pre-assignment strategy is obtained from the fully connected layer.

[0065] In this embodiment of the disclosure, multiple self-attention networks are used to separately determine the sub-scenes for task data matching and the tools for sub-scene matching, thereby improving the accuracy of the determined tools.

[0066] Figure 4 A flowchart of another task allocation method according to an embodiment of this disclosure is shown.

[0067] like Figure 4 As shown, the method may include: S410 divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios; S420 divides the tools used to process sub-scenes based on the sub-scenes, resulting in multiple task sets corresponding to the sub-scenes, with each tool having a corresponding tool identifier.

[0068] The S430 uses the first self-attention network contained in the intelligent model and the task data to determine the sub-scene that matches the task data.

[0069] S440 is a tool that determines the matching of a sub-scene through a sub-scene and a second self-attention network contained in the intelligent model.

[0070] S450 is a tool that inputs the matching task data and the sub-scenes that match the task data into the sorting intelligent model to obtain the sorted data.

[0071] In some embodiments, the ranking intelligent model can be a Prompt, which is a carrier of instructions for the user to interact with the Large Language Model (LLM). Essentially, it is a structured input used to guide the model to generate expected outputs. It is equivalent to a "programming language for the large model," and different ways of writing it can significantly affect the quality, style, and accuracy of the model's responses.

[0072] In some embodiments, corresponding prompt words can be set and input into the aforementioned intelligent sorting model. The tool that completes the sorting is then obtained by inputting the set prompt words into the intelligent sorting model. For example, the prompt word could be: Task data is: {Query}, to complete the {...}th sorting task data... The sub-scene of step} is: {Scene}, and the tools used to complete the sub-scene task are: {TOOLS}. Please output the execution order of these tools.

[0073] In some embodiments, the ranking intelligent model may also include a Bert_base model, whose inputs are sub-scene vector encoding and tool vector encoding of sub-scene task pre-assignment strategy. The vector encoding uses a 768-dimensional dimension, and the corresponding content is filled into the Prompt paradigm as the input of the model, and the output is the sequential position number of each tool.

[0074] In this embodiment of the disclosure, the tools for matching task data are sorted by a sorting intelligent model, which enables the tools based on the sorting sequence to perform task data more accurately.

[0075] Figure 5 A flowchart of yet another task allocation method according to an embodiment of this disclosure is shown.

[0076] like Figure 5 As shown, the method may include: S510 divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios; S520 divides the tools used to process sub-scenes based on the sub-scenes, resulting in multiple task sets corresponding to the sub-scenes, with each tool having a corresponding tool identifier.

[0077] S530 performs vector encoding on the task data and sub-scenes through the input layer of the first self-attention network to obtain the encoded task data and sub-scene data.

[0078] In some embodiments, vector encoding of task data and sub-scenes may include encoding using the methods described in the above embodiments, which will not be repeated here.

[0079] In some embodiments, vector encoding of task data and sub-scenes through the input layer of a first self-attention network may include converting task data and sub-scenes into vectors.

[0080] S540 performs feature transformation on the encoded task data and sub-scene data through the fully connected layer of the first self-attention network to obtain the transformed task data and sub-scene data.

[0081] The S550 uses an attention probability calculation mechanism, transformed task data, and sub-scene data to determine the sub-scene that matches the task data.

[0082] In some embodiments, the attention probability calculation mechanism may include .

[0083] S560 is a tool that determines the matching of a sub-scene through a sub-scene and a second self-attention network contained in the intelligent model.

[0084] The task allocation method provided in the embodiments of this disclosure divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios. The tools used to handle the sub-scenarios are divided based on the sub-scenarios to obtain a set of tasks corresponding to multiple sub-scenarios. Each tool is identified by a tool identifier. By using a pre-trained intelligent model and task data, the tool that matches the task data is determined in the set of sub-scenarios and the set of tasks, thereby improving the accuracy of matching the tool with the task data.

[0085] Figure 6 A flowchart of yet another task allocation method according to an embodiment of this disclosure is shown.

[0086] like Figure 6 As shown, the method may include: S610 divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios; S620 divides the tools used to process sub-scenes based on the sub-scenes, resulting in multiple task sets corresponding to the sub-scenes, with each tool having a corresponding tool identifier.

[0087] The S630 uses the first self-attention network contained in the intelligent model and the task data to determine the sub-scene that matches the task data.

[0088] S640, the sub-scene and the tool are vector-encoded through the input layer of the second self-attention network to obtain the encoded sub-scene and the tool; S650 uses a fully connected layer in the second self-attention network to perform feature transformation on the encoded sub-scene and tool to obtain the transformed sub-scene and tool. S660 uses an attention probability calculation mechanism, the transformed sub-scene, and the tool to determine the matching tool for the sub-scene.

[0089] The task allocation method provided in the embodiments of this disclosure divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios. The tools used to handle the sub-scenarios are divided based on the sub-scenarios to obtain a set of tasks corresponding to multiple sub-scenarios. Each tool is identified by a tool identifier. By using a pre-trained intelligent model and task data, the tool that matches the task data is determined in the set of sub-scenarios and the set of tasks, thereby improving the accuracy of matching the tool with the task data.

[0090] Figure 7 A flowchart of yet another task allocation method according to an embodiment of this disclosure is shown.

[0091] like Figure 7 As shown, the method may include: S710, acquire training samples, which include historical task data, historical sub-scenes and historical tools; The S720 inputs historical task data into the intelligent model to be trained, and obtains the sub-scenes and tools output by the intelligent model to be trained. S730 determines the loss function value by the similarity between the sub-scene output by the intelligent model to be trained and the historical sub-scene, and the similarity between the tool output by the intelligent model and the historical tool. S740, in response to the convergence of the loss function value, yields a pre-trained intelligent model.

[0092] S750 divides fault handling scenarios based on business processes to obtain a set of sub-scenarios containing multiple sub-scenarios; S760 divides the tools used to process sub-scenes based on the sub-scenes, resulting in multiple task sets corresponding to the sub-scenes, with each tool corresponding to a tool identifier; The S770 is a tool that uses pre-trained intelligent models and task data to determine matching task data in sub-scene sets and task sets.

[0093] The task allocation method provided in the embodiments of this disclosure divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios. The tools used to handle the sub-scenarios are divided based on the sub-scenarios to obtain a set of tasks corresponding to multiple sub-scenarios. Each tool is identified by a tool identifier. By using a pre-trained intelligent model and task data, the tool that matches the task data is determined in the set of sub-scenarios and the set of tasks, thereby improving the accuracy of matching the tool with the task data.

[0094] Based on the same inventive concept, this disclosure also provides a task allocation device, as shown in the following embodiment. Since the principle by which this device solves the problem is similar to that of the method embodiment described above, the implementation of this device embodiment can refer to the implementation of the method embodiment described above, and repeated details will not be described again.

[0095] Figure 8 A structural diagram of a task allocation device according to an embodiment of the present disclosure is shown.

[0096] like Figure 8 As shown, the device 800 may include: The first partitioning module 810 is used to partition the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios. The second partitioning module 820 is used to partition the tools used to process sub-scenes based on the sub-scenes to obtain multiple task sets corresponding to the sub-scenes, and the tools correspond to tool identifiers; The first determining module 830 is a tool for determining matching task data in a set of sub-scenes and a set of tasks using a pre-trained intelligent model and task data.

[0097] In one embodiment of this disclosure, the first determining module includes: The first determining submodule is used to determine the sub-scene that matches the task data through the first self-attention network contained in the intelligent model and the task data; The second determination submodule is used to determine the tool that matches the sub-scene through the sub-scene and the second self-attention network contained in the intelligent model.

[0098] In one embodiment of this disclosure, the apparatus further includes: The input module is used to input the tool that matches the task data and the sub-scenes that match the task data into the sorting intelligent model to obtain the sorted tool.

[0099] In one embodiment of this disclosure, the first determining submodule includes: The first encoding unit is used to perform vector encoding on the task data and sub-scenes through the input layer of the first self-attention network to obtain the encoded task data and sub-scene data. The first transformation unit is used to perform feature transformation on the encoded task data and sub-scene data through the fully connected layer of the first self-attention network to obtain the transformed task data and sub-scene data. The first determining unit is used to determine the sub-scene that matches the task data through the attention probability calculation mechanism, the transformed task data, and the sub-scene data.

[0100] In one embodiment of this disclosure, the second determining submodule includes: The second encoding unit is used to perform vector encoding on the sub-scene and the tool through the input layer of the second self-attention network to obtain the encoded sub-scene and the tool. The second transformation unit is used to perform feature transformation on the encoded sub-scene and tool through the fully connected layer in the second self-attention network to obtain the transformed sub-scene and tool. The second determining unit is used to determine the tool that matches the sub-scene by using an attention probability calculation mechanism, the transformed sub-scene, and the tool.

[0101] In one embodiment of this disclosure, the apparatus further includes: The acquisition module is used to acquire training samples, which include historical task data, historical sub-scenes, and historical tools. The input module is used to input historical task data into the intelligent model to be trained, and obtain the sub-scenes and tools output by the intelligent model to be trained. The second determining module is used to determine the loss function value based on the similarity between the sub-scene output by the intelligent model to be trained and the historical sub-scene, and the similarity between the tool output by the intelligent model and the historical tool. The third determination module is used to obtain the pre-trained intelligent model in response to the convergence of the loss function value.

[0102] In one embodiment of this disclosure, the ranking intelligence model includes an input prompt pre-trained language model.

[0103] The task allocation device provided in the embodiments of this disclosure divides the fault handling scenario based on the business process to obtain a set of sub-scenarios containing multiple sub-scenarios. The tools used to handle the sub-scenarios are divided based on the sub-scenarios to obtain a set of tasks corresponding to multiple sub-scenarios. Each tool is identified by a tool identifier. By using a pre-trained intelligent model and task data, the tool that matches the task data is determined in the set of sub-scenarios and the set of tasks, thereby improving the accuracy of matching the tool with the task data.

[0104] The task allocation device provided in this embodiment can be used to execute the task allocation methods provided in the above-described method embodiments. The implementation principle and technical effect are similar, and for the sake of simplicity, they will not be described in detail here.

[0105] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."

[0106] The following reference Figure 9 To describe an electronic device 900 according to such an embodiment of the present disclosure. Figure 9 The electronic device 900 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0107] like Figure 9 As shown, the electronic device 900 is manifested in the form of a general-purpose computing device. The components of the electronic device 900 may include, but are not limited to: at least one processing unit 99, at least one storage unit 920, and a bus 930 connecting different system components (including the storage unit 920 and the processing unit 99).

[0108] The storage unit stores program code, which can be executed by the processing unit 99 to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure. For example, the processing unit 99 can perform the following steps of the above method embodiments: Receive the user's first access domain request, which includes the domain tag and the first domain; Determine if cached DNS data exists corresponding to the first domain name; In the absence of cached parsed data, determine whether the domain tag matches the preset domain tag; If the domain name tag matches the preset domain name tag, the first domain name will be sent to the security authority server so that the security authority server can resolve the first domain name.

[0109] Storage unit 920 may include readable media in the form of volatile storage units, such as random access memory (RAM) 9201 and / or cache memory 9202, and may further include read-only memory (ROM) 9203.

[0110] The storage unit 920 may also include a program / utility 9204 having a set (at least one) program module 9205, such program module 9205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0111] Bus 930 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0112] Electronic device 900 can also communicate with one or more external devices 940 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 900, and / or with any device that enables electronic device 900 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 950. Furthermore, electronic device 900 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 990. As shown, network adapter 990 communicates with other modules of electronic device 900 via bus 930. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 900, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0113] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0114] In exemplary embodiments of this disclosure, a computer-readable storage medium is also provided, which may be a readable signal medium or a readable storage medium. A program product capable of implementing the methods described above is stored thereon. In some possible implementations, various aspects of this disclosure may also be implemented as a program product including program code, which, when run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.

[0115] More specific examples of computer-readable storage media in this disclosure may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0116] In this disclosure, a computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, wherein readable program code is carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting a program for use by or in connection with an instruction execution system, apparatus, or device.

[0117] Optionally, the program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0118] In practice, program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and 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).

[0119] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0120] Furthermore, although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0121] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, mobile terminal, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0122] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the appended claims.

Claims

1. A task allocation method, characterized in that, include: Based on the business process, the fault handling scenario is divided into a set of sub-scenarios containing multiple sub-scenarios; The tools used to process sub-scenes are divided based on the sub-scenes to obtain multiple task sets corresponding to the sub-scenes, and the tools are identified by tool identifiers. A tool that uses pre-trained intelligent models and task data to determine matching task data in the sub-scene set and task set.

2. The method according to claim 1, characterized in that, The tool for determining matching task data in the sub-scene set and task set using pre-trained intelligent models and task data includes: The sub-scene matching the task data is determined by the first self-attention network included in the intelligent model and the task data. The tool matching the sub-scene is determined by the sub-scene and the second self-attention network contained in the intelligent model.

3. The method according to claim 2, characterized in that, The method further includes: The tool that matches the task data and the sub-scenes that match the task data are input into the sorting intelligent model to obtain the sorted tool.

4. The method according to claim 2, characterized in that, The step of determining the sub-scene matching the task data using the first self-attention network included in the intelligent model and any task data includes: The task data and sub-scenes are vector-encoded through the input layer of the first self-attention network to obtain the encoded task data and sub-scene data. The encoded task data and sub-scene data are transformed by the fully connected layer of the first self-attention network to obtain the transformed task data and sub-scene data. The sub-scene that matches the task data is determined by using an attention probability calculation mechanism, the transformed task data, and the sub-scene data.

5. The method according to claim 2, characterized in that, The tool for determining the match between the sub-scene and the second self-attention network included in the intelligent model includes: The sub-scene and the tool are vector-encoded through the input layer of the second self-attention network to obtain the encoded sub-scene and tool; The encoded sub-scene and tool are transformed by the fully connected layer in the second self-attention network to obtain the transformed sub-scene and tool. The tool that matches the sub-scene is determined by using an attention probability calculation mechanism, the transformed sub-scene, and the tool.

6. The method according to claim 1, characterized in that, The method further includes: Obtain training samples, which include historical task data, historical sub-scenes, and historical tools; The historical task data is input into the intelligent model to be trained to obtain the sub-scenes and tools output by the intelligent model to be trained. The loss function value is determined by the similarity between the sub-scene output by the intelligent model to be trained and the historical sub-scene, and the similarity between the tool output by the intelligent model and the historical tool. The pre-trained intelligent model is obtained in response to the convergence of the loss function value.

7. The method according to claim 3, characterized in that, The sorting intelligent model includes the input prompt Prompt pre-trained language model.

8. A task allocation device, characterized in that, include: The first partitioning module is used to partition fault handling scenarios based on business processes, resulting in a set of sub-scenarios containing multiple sub-scenarios. The second partitioning module is used to partition the tools used to process sub-scenes based on the sub-scenes to obtain multiple task sets corresponding to the sub-scenes, wherein the tools correspond to tool identifiers; The first determining module is a tool used to determine matching task data in the sub-scene set and task set using a pre-trained intelligent model and task data.

9. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the task allocation method according to any one of claims 1 to 7 by executing the executable instructions.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the task allocation method according to any one of claims 1 to 7.

Images