Enterprise information query task allocation method and system
By constructing a subtask dependency graph and a time estimation model, combined with a node capability matrix and real-time monitoring, the problems of low resource utilization and poor stability in enterprise information query task allocation are solved, and fast and accurate task execution and result output are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHONGRUI FESCO OUTSOURCING (BEIJING) CO LTD
- Filing Date
- 2025-12-31
- Publication Date
- 2026-04-28
AI Technical Summary
The allocation of enterprise information query tasks suffers from issues such as rudimentary task breakdown and dependency analysis, idle or overloaded resources, and an imperfect task execution fault tolerance mechanism, resulting in low query efficiency and poor stability.
By constructing a subtask dependency graph, using a time estimation model and node capability matrix for task allocation, and monitoring preceding tasks in real time and triggering fallback strategies, we can ensure that the task execution order is reasonable and resource utilization is high.
It enables the rapid and accurate completion of enterprise information query tasks, improves resource utilization and task processing efficiency, reduces the impact of execution failures on the entire process, and ensures the stability of the query service.
Smart Images
Figure CN121935274A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method and system for allocating enterprise information query tasks. Background Technology
[0002] In enterprise information management scenarios, the rationality of information query task allocation has a significant impact on efficiency. Currently, enterprise information query task allocation often suffers from process shortcomings: task decomposition and dependency analysis are rudimentary, and the execution order of subtasks is prone to confusion, leading to increased overall time consumption; task processing time estimation relies on experience and lacks accurate model support; node selection only focuses on static performance and fails to dynamically adapt to real-time load and task characteristics, resulting in idle or overloaded resources; furthermore, the task execution fault tolerance mechanism is imperfect, and the failure of preceding tasks can easily trigger a complete process interruption, making it difficult to guarantee the stability and continuity of query services and failing to meet the enterprise's needs for efficient and reliable information query. Therefore, there is an urgent need for a smarter and more accurate method for allocating enterprise information query tasks. Summary of the Invention
[0003] In view of this, the present invention proposes a method and system for allocating enterprise information query tasks, which can achieve fast and accurate completion of enterprise information query tasks.
[0004] To achieve the above objectives, the present invention provides the following technical solution: A method for allocating enterprise information query tasks, comprising: Receive enterprise login request and perform enterprise identity authentication, obtain the target information query request submitted by the enterprise, identify the acquisition target corresponding to the target information query request, and the number of acquisition targets is at least one; Obtain a list of sub-tasks corresponding to each acquisition target based on a preset rule base; The task dependency analysis algorithm identifies the pre- and post-relationships between each subtask and outputs a subtask dependency graph. Construct a set of subtasks based on the number of acquisition targets, the subtask list, and the subtask dependency graph; The task processing time of each subtask in the subtask set is calculated based on a pre-built time estimation model. Real-time collection of node status data for each task processing node; Based on the pre-built node capability matrix, a set of candidate nodes is selected for each subtask. Based on the node status data and the task processing time, the optimal target node is selected from the set of candidate nodes, and each subtask is assigned to the corresponding optimal target node. For initial subtasks without prior dependencies, the corresponding optimal target node is triggered to execute immediately after task allocation is completed; For subtasks with prerequisites, monitor the execution status of the prerequisite subtasks. If the prerequisite subtask is completed and the result is verified, automatically activate the corresponding optimal target node to execute the subsequent subtask. If the prerequisite subtask fails to execute, trigger the corresponding fallback strategy according to the failure type. The results of each subtask are categorized and summarized according to the target obtained, and the enterprise information query results are output.
[0005] Based on the above technical solution, the present invention can be further improved as follows: Optionally, the acquisition objectives include acquiring the employee information table of the previous month, acquiring social security policy data, acquiring the policy change time series, acquiring the employee information difference list of the current month, acquiring the employee information table of the current month, acquiring the information of newly added employees this month, acquiring the information of employees who left this month, acquiring the information of employees whose information has changed this month, acquiring the employees whose social security needs to be adjusted this month, acquiring the social security adjustment costs of employees this month, and acquiring the employee relations management information of the current month.
[0006] Optionally, the subtasks include a first subtask, a second subtask, a third subtask, a fourth subtask, a fifth subtask, a sixth subtask, a seventh subtask, an eighth subtask, a ninth subtask, a tenth subtask, and an eleventh subtask: The first subtask is to obtain the employee information table of the corresponding enterprise for the previous month based on the enterprise login request with enterprise identification; The second subtask is to obtain social security policy data, construct a policy change time series based on the social security policy data, and record the effective time, geographical scope, applicable population and specific clauses of each social security policy; The third subtask is to obtain the initial employee information table for this month; The fourth subtask is to calculate the first comprehensive difference value between the initial employee information table for this month and the employee information table for last month, calculate the second comprehensive difference value between the policy change time series and the initial employee information table for this month, and determine a difference list based on the first comprehensive difference value and the second comprehensive difference value; The fifth subtask is to obtain information on newly added employees based on the aforementioned discrepancy list; The sixth subtask is to obtain information on departing employees based on the aforementioned discrepancy list; The seventh subtask is to modify employee information based on the information obtained from the aforementioned discrepancy list; The eighth subtask is to obtain the employees whose social security contributions need to be adjusted based on the aforementioned discrepancy list; The ninth subtask is to generate the current month's employee information table based on the difference list, the initial current month's employee information table, and the previous month's employee information table; The tenth subtask is to calculate the social security adjustment cost for employees whose social security contributions need to be adjusted. The eleventh subtask is to generate the current month's employee relations management information based on the current month's employee information table and the historical employee information table for the past year.
[0007] Optionally, the step of identifying the pre- and post-relationships between each subtask using a task dependency analysis algorithm and outputting a subtask dependency graph includes: Calculate the dependency strength value between subtask k and subtask k+1. When the dependency strength value is 1, determine that subtask k is the predecessor of subtask k+1 and the successor of subtask k. When the dependency strength value is 0, it is determined that subtask k and subtask k+1 have no dependency relationship. Using each subtask as a node, and the preceding and following relationships of the subtasks as directed edges, with the direction of the directed edges pointing from the preceding subtask to the following subtask, a subtask dependency graph is constructed and output.
[0008] Optionally, constructing a subtask set based on the number of acquisition targets, the subtask list, and the subtask dependency graph includes: When the number of targets is one, the execution order of global subtasks is determined based on the subtask list and the subtask dependency graph, and a subtask set is generated. When there are multiple targets, the multiple subtask lists are merged, deduplication is performed by using the unique identifier of the subtask, the pre-relationships and post-relationships of the global subtasks after deduplication are sorted out based on the subtask dependency graph, the global execution order is determined, and a set of subtasks is generated.
[0009] Optionally, the calculation of the task processing time of each subtask in the subtask set based on the pre-built time estimation model includes: The task processing time of subtask k is calculated using formula (1); Formula (1); In the formula, Let k be the task processing time for subtask k. The data volume feature value for subtask k. Let be the complexity coefficient of subtask k. This refers to the performance coefficient of the task processing node. The historical execution efficiency of subtask k. represents the adaptation coefficient between subtask k and task processing node.
[0010] Optionally, the step of selecting the optimal target node from the candidate node set based on the node status data and the task processing time includes: The fitness value of task processing node m for processing subtask k is calculated using formula (2). Formula (2); In the formula, The fitness value for processing subtask k for task processing node m. represents the matching degree coefficient between task processing node m and the subtask processing model. This represents the base processing performance value for task processing node m. Let m be the number of tasks currently being handled by the task processing node. Let k be the estimated processing time for subtask k. As the task load weight, Time weighting; The task processing node with the highest fit value is determined as the optimal target node.
[0011] Optionally, if the preceding subtask fails, triggering a corresponding fallback strategy based on the failure type includes: If the task processing node fails, it will automatically switch to the backup task processing node to retry. If the error is due to the task processing node itself, terminate the execution of subsequent subtasks and push an exception notification to the enterprise.
[0012] A system for allocating enterprise information query tasks, comprising: The request parsing module is used to receive enterprise login requests and perform enterprise identity authentication, obtain the target information query requirements submitted by the enterprise, identify the acquisition targets corresponding to the target information query requirements, and the number of acquisition targets is at least one. The subtask generation module is used to obtain a list of subtasks corresponding to each acquisition target based on a preset rule base. The task analysis module is used to identify the pre- and post-relationships between each subtask through a task dependency analysis algorithm, and output a subtask dependency graph. A set construction module is used to construct a subtask set based on the number of acquisition targets, the subtask list, and the subtask dependency graph. The time calculation module is used to calculate the task processing time of each subtask in the subtask set based on a pre-built time estimation model; The data acquisition module is used to collect the node status data of each task processing node in real time. The node matching module is used to filter a set of candidate nodes for each subtask based on a pre-built node capability matrix, filter the optimal target node from the set of candidate nodes based on the node status data and the task processing time, and assign each subtask to the corresponding optimal target node. For initial subtasks without prior dependencies, the corresponding optimal target node is triggered to execute immediately after task allocation is completed; For subtasks with prerequisites, monitor the execution status of the prerequisite subtasks. If the prerequisite subtask is completed and the result is verified, automatically activate the corresponding optimal target node to execute the subsequent subtask. If the prerequisite subtask fails to execute, trigger the corresponding fallback strategy according to the failure type. The results output module is used to classify and summarize the execution results of each subtask according to the acquisition target, and output the enterprise information query results.
[0013] An electronic device includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the method described herein.
[0014] The present invention has the following advantages: The enterprise information query task allocation method in this invention ensures a reasonable task execution order and avoids process chaos by using a subtask dependency graph; it improves resource utilization and task processing efficiency by matching the optimal target node with a time estimation model, node capability matrix, and real-time node status; it reduces the impact of execution failure on the entire process and ensures service stability by improving the pre-task monitoring and fallback strategy; and it classifies the results according to the acquisition target and outputs enterprise information query results to ensure accurate output of enterprise information query results, thereby achieving fast and accurate completion of enterprise information query tasks. Attached Figure Description
[0015] For illustrative purposes and not limiting, the present invention will now be described in conjunction with embodiments and accompanying drawings, wherein: Figure 1 This is a flowchart illustrating the method for allocating enterprise information query tasks in an embodiment of the present invention. Figure 2 This is a schematic diagram of the main components of the enterprise information query task allocation system in an embodiment of the present invention; Figure 3 This is a schematic diagram of the physical structure of the electronic device provided by the present invention. Detailed Implementation
[0016] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0017] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be used interchangeably where appropriate for the embodiments of the invention described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0018] It should be noted that, where there is no conflict, the embodiments and features of the present invention can be combined with each other. The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
[0019] Figure 1 This is a flowchart illustrating the method for allocating enterprise information query tasks in an embodiment of the present invention, such as... Figure 1 As shown, the method for allocating enterprise information query tasks provided in this embodiment of the invention includes the following steps S101 to S110.
[0020] S101 receives the enterprise login request and performs enterprise identity authentication, obtains the target information query request submitted by the enterprise, and identifies the target to be obtained corresponding to the target information query request.
[0021] The system adopts a "company identifier + two-factor authentication" model: enterprise login requests must carry a unique enterprise identifier (such as a unified social credit code) and verify a dynamic password (such as an SMS verification code / key signature) to prevent unauthorized access.
[0022] After authentication is successful, the system retrieves the enterprise permission configuration table and only allows access to the target that the enterprise has permission to query (e.g., prohibiting cross-enterprise queries of employee information).
[0023] For natural language query requests submitted by enterprises (such as "query the social security adjustment costs and list of departing employees this month"), core keywords ("social security adjustment costs for employees this month" and "list of departing employees") are extracted, matched with a preset target database, and automatically mapped to the corresponding standardized target.
[0024] If duplicate targets are detected, they will be automatically removed; if there are related targets (such as "get this month's employee information table" and "get this month's newly added employee information"), they will be sorted by default according to "basic targets take precedence over derived targets".
[0025] The acquisition objectives include obtaining the employee information table for the previous month, obtaining social security policy data, obtaining the policy change time series, obtaining the employee information difference list for the current month, obtaining the employee information table for the current month, obtaining information on newly added employees for the current month, obtaining information on employees who left the company for the current month, obtaining information on employees whose information has changed for the current month, obtaining information on employees whose social security needs to be adjusted for the current month, obtaining the social security adjustment costs for employees for the current month, and obtaining employee relations management information for the current month.
[0026] S102, obtain the sub-task list corresponding to each acquisition target based on the preset rule base.
[0027] The subtasks include the first subtask, the second subtask, the third subtask, the fourth subtask, the fifth subtask, the sixth subtask, the seventh subtask, the eighth subtask, the ninth subtask, the tenth subtask, and the eleventh subtask: The first subtask is to obtain the employee information table of the corresponding enterprise for the previous month based on the enterprise login request with enterprise identification; The second subtask is to obtain social security policy data, construct a policy change time series based on the social security policy data, and record the effective time, geographical scope, applicable population and specific clauses of each social security policy; The third subtask is to obtain the initial employee information table for this month; The fourth subtask is to calculate the first comprehensive difference value between the initial employee information table for this month and the employee information table for last month, calculate the second comprehensive difference value between the policy change time series and the initial employee information table for this month, and determine a difference list based on the first comprehensive difference value and the second comprehensive difference value; The fifth subtask is to obtain information on newly added employees based on the aforementioned discrepancy list; The sixth subtask is to obtain information on departing employees based on the aforementioned discrepancy list; The seventh subtask is to modify employee information based on the information obtained from the aforementioned discrepancy list; The eighth subtask is to obtain the employees whose social security contributions need to be adjusted based on the aforementioned discrepancy list; The ninth subtask is to generate the current month's employee information table based on the difference list, the initial current month's employee information table, and the previous month's employee information table; The tenth subtask is to calculate the social security adjustment cost for employees whose social security contributions need to be adjusted. The eleventh subtask is to generate the current month's employee relations management information based on the current month's employee information table and the historical employee information table for the past year.
[0028] S103 uses a task dependency analysis algorithm to identify the pre- and post-relationships between each subtask and outputs a subtask dependency graph.
[0029] Calculate the dependency strength value between subtask k and subtask k+1. When the dependency strength value is 1, determine that subtask k is the predecessor of subtask k+1 and the successor of subtask k. When the dependency strength value is 0, it is determined that subtask k and subtask k+1 have no dependency relationship. Using each subtask as a node, and the preceding and following relationships of the subtasks as directed edges, with the direction of the directed edges pointing from the preceding subtask to the following subtask, a subtask dependency graph is constructed and output.
[0030] S104, construct a set of subtasks based on the number of targets to be acquired, the subtask list, and the subtask dependency graph.
[0031] When the number of targets is one, the execution order of global subtasks is determined based on the subtask list and the subtask dependency graph, and a subtask set is generated. When there are multiple targets, the multiple subtask lists are merged, deduplication is performed by using the unique identifier of the subtask, the pre-relationships and post-relationships of the global subtasks after deduplication are sorted out based on the subtask dependency graph, the global execution order is determined, and a set of subtasks is generated.
[0032] S105, calculate the task processing time of each subtask in the subtask set based on the pre-built time estimation model.
[0033] The task processing time of subtask k is calculated using formula (1); Formula (1); In the formula, Let k be the task processing time for subtask k. The data volume feature value for subtask k. Let be the complexity coefficient of subtask k. This refers to the performance coefficient of the task processing node. The historical execution efficiency of subtask k. represents the adaptation coefficient between subtask k and task processing node.
[0034] S106 collects node status data of each task processing node in real time.
[0035] The hardware performance and workload metrics of the data acquisition nodes are collected in real time, with a sampling frequency of 1 second per acquisition. Hardware performance metrics: CPU utilization, memory utilization, disk I / O, network bandwidth; Task load metrics: number of tasks currently being handled, task queue length, and average task response time.
[0036] The collected indicator data are normalized to eliminate differences in dimensions. A sliding window algorithm is used to filter out outliers (such as a sudden 100% CPU utilization) to ensure data stability.
[0037] S107: Based on the pre-built node capability matrix, a set of candidate nodes is selected for each subtask. Based on the node status data and task processing time, the optimal target node is selected from the set of candidate nodes, and each subtask is assigned to the corresponding optimal target node.
[0038] The fitness value of task processing node m for processing subtask k is calculated using formula (2). Formula (2); In the formula, The fitness value for processing subtask k for task processing node m. represents the matching degree coefficient between task processing node m and the subtask processing model. This represents the base processing performance value for task processing node m. Let m be the number of tasks currently being handled by the task processing node. Let k be the estimated processing time for subtask k. As the task load weight, Time weighting; The task processing node with the highest fit value is determined as the optimal target node.
[0039] If multiple nodes have the same fit value, a secondary filtering rule is triggered: the node with the shorter current task queue length is selected first.
[0040] S108: For starting subtasks without prior dependencies, the corresponding optimal target node is triggered to execute immediately after task allocation is completed.
[0041] The number of preceding dependent nodes of a subtask is queried based on the dependency graph. If the number is 0, it is determined to be the starting subtask. The starting subtask must meet the following conditions: node resources have been allocated and input data is ready.
[0042] For multiple initial subtasks without prior dependencies (such as the first and second subtasks), the system adopts a multi-threaded / distributed parallel execution mode, triggering the corresponding nodes to execute simultaneously, thereby improving task startup efficiency.
[0043] S109: For subtasks with prerequisite dependencies, monitor the execution status of the prerequisite subtasks. If the prerequisite subtask is completed and the result is verified, automatically activate the corresponding optimal target node to execute the subsequent subtask. If the prerequisite subtask fails to execute, trigger the corresponding fallback strategy according to the failure type.
[0044] A unified task status monitoring thread is deployed to collect the execution status ("pending execution / in execution / execution completed / execution failed") of the preceding subtasks in real time. After execution, the result verification algorithm is automatically triggered: verifying the completeness, format correctness, and logical consistency of the output data (e.g., the number of records in the difference list must match the number of records in the input table).
[0045] The fallback strategy is shown in Table 1; Table 1 If the task processing node fails, it will automatically switch to the backup task processing node to retry. If the error is due to the task processing node itself, terminate the execution of subsequent subtasks and push an exception notification to the enterprise.
[0046] S110: Classify and summarize the execution results of each subtask according to the acquisition target, and output the enterprise information query results.
[0047] Based on the target-subtask output mapping relationship, the subtask execution results are categorized into the corresponding target. For multiple sets of result data with the same acquisition target, a data fusion algorithm is executed: duplicate data is eliminated, missing fields are supplemented, and the data format is unified.
[0048] Supports multiple output formats: structured formats (Excel / JSON) and visualization formats (charts / reports); Execution permission double verification before output: results are only pushed to authorized enterprise accounts, and the download link for results can be encrypted and the validity period can be set.
[0049] One embodiment is as follows: A technology company needs to complete a preliminary information query for employee social security adjustments at the beginning of each month. The target information query needs are: to obtain the employee information table from the previous month, to obtain social security policy data, to obtain a list of employee information discrepancies for the current month, to identify employees whose social security needs are to be adjusted this month, and to obtain the cost of employee social security adjustments for the current month. Based on the method of this invention, the execution process is as follows: After a company submits a login request and the system completes identity authentication, it receives the company's information query request and accurately identifies the five acquisition targets mentioned above.
[0050] The system retrieves a preset rule base and generates a list of sub-tasks corresponding to the five target matching criteria, involving sub-tasks one through ten: First subtask: Obtain the company's employee information table for the previous month (including 500 employee records); The second sub-task: Obtain the latest social security policy data, construct a time series of policy changes, record the policy effective date as the 1st of the current month, and apply to local technology industry employees; Third subtask: Obtain the initial employee information table for this month entered into the HR system; Fourth subtask: Calculate the first comprehensive difference between the initial employee information table for this month and the table for last month, calculate the second comprehensive difference between the policy change time series and the initial employee information table for this month, and determine the difference list; Subtasks 5 through 8: Extract information on newly added, resigned, changed, and pending social security adjustments for employees from the discrepancy list; Ninth subtask: Generate this month's employee information table based on the discrepancy list, the initial current month table, and the previous month table; The tenth sub-task: Calculate the social security adjustment costs for employees whose social security needs to be adjusted, based on social security policies.
[0051] The system calculates the dependency strength between subtasks using a task dependency analysis algorithm: The employee information table for the previous month output by the first subtask is the required input data for the fourth and ninth subtasks, with a dependency strength value of 1. Therefore, the first subtask is determined to be a prerequisite subtask for the fourth and ninth subtasks. The policy change time series output by the second subtask is the necessary input data for the fourth and tenth subtasks, with a dependency strength value of 1. Therefore, the second subtask is determined to be a prerequisite subtask for the fourth and tenth subtasks. The list of differences output by the fourth subtask is the required input data for the fifth to ninth subtasks, with a dependency strength value of 1. Therefore, the fourth subtask is determined to be a prerequisite subtask for the fifth to ninth subtasks. The social security employee information to be adjusted output by the eighth subtask is the necessary input data for the tenth subtask, with a dependency strength value of 1. Therefore, the eighth subtask is determined to be a prerequisite subtask for the tenth subtask. The dependency strength between the remaining subtasks is 0, indicating no relationship. The system outputs a subtask dependency graph, with subtasks as nodes and dependencies as directed edges.
[0052] The system retrieved 5 targets. After merging the corresponding subtask lists and deduplicating them using their unique identifiers, it then organized the global execution order based on the dependency graph: the first and second subtasks were executed in parallel → the fourth subtask → the fifth to ninth subtasks were executed in parallel → the tenth subtask, generating the final set of subtasks.
[0053] The system call time estimation model calculates the processing time of each subtask using formula (1). Taking the fourth subtask as an example: Data volume characteristic value Item, complexity coefficient Node performance coefficient Historical execution efficiency Subtask and task processing node adaptation coefficient ; Substituting into the formula, the processing time for the fourth subtask is calculated to be 3.0 seconds.
[0054] Similarly, the time for the first subtask is calculated to be 2.0 seconds, the time for the second subtask to be 2.5 seconds, and the time for the tenth subtask to be 4.0 seconds.
[0055] Assumptions: Nodes 1, 2, 3, 4, and 5 are bound to the employee information processing model; Nodes 6 and 7 are bound to the social security policy calculation model; Nodes 8, 9, 10, and 11 are general nodes.
[0056] The matching degree coefficient of node 1 is 1, the basic processing performance value of node 1 is 1.6, and the number of tasks currently carried by node 1 is 3.
[0057] The matching degree coefficient of node 2 is 1, the basic processing performance value of node 1 is 1.5, and the number of tasks currently carried by node 1 is 1.
[0058] The matching degree coefficient of node 3 is 1, the basic processing performance value of node 1 is 1.7, and the number of tasks currently carried by node 1 is 2.
[0059] The matching coefficient of node 4 is 1, the basic processing performance value of node 1 is 1.4, and the number of tasks currently carried by node 1 is 0.
[0060] The matching degree coefficient of node 5 is 1, the basic processing performance value of node 1 is 1.8, and the number of tasks currently carried by node 1 is 4.
[0061] The matching degree coefficient of node 6 is 1, the basic processing performance value of node 1 is 1.9, and the number of tasks currently carried by node 1 is 2.
[0062] The matching coefficient of node 7 is 1, the basic processing performance value of node 1 is 2.0, and the number of tasks currently carried by node 1 is 1.
[0063] The matching coefficient of node 8 is 0.8, the basic processing performance value of node 1 is 1.2, and the number of tasks currently carried by node 1 is 0.
[0064] The matching coefficient of node 9 is 0.8, the basic processing performance value of node 1 is 1.1, and the number of tasks currently carried by node 1 is 1.
[0065] The matching coefficient of node 10 is 0.8, the basic processing performance value of node 1 is 1.3, and the number of tasks currently carried by node 1 is 2.
[0066] The matching coefficient of node 11 is 0.8, the basic processing performance value of node 1 is 1.0, and the number of tasks currently carried by node 1 is 0.
[0067] Based on the node capability matrix and subtask type, the system first filters the candidate node set for each subtask, and then calculates the fitness value using formula (2) to determine the optimal target node.
[0068] Unified configuration of weight parameters: task load weight is 0.5, and time weight is 0.5.
[0069] The fourth subtask (employee information difference calculation) is filtered, with candidate nodes being nodes 1-5 bound to the employee information processing model. Known parameters: the estimated processing time for the fourth subtask is 3 seconds. The calculated fitness scores are 0.53 for node 1, 0.75 for node 2, 0.68 for node 3, 0.93 for node 4, and 0.51 for node 5. Node 4 has the highest fitness value and is therefore determined as the optimal target node for the fourth subtask.
[0070] The calculation method for other subtasks is the same as above; The first and second subtasks have no prerequisites. After the system completes the allocation, it immediately triggers node 4 to execute the first subtask and node 7 to execute the second subtask, thus initiating the task processing flow.
[0071] There are prerequisite subtasks that require execution and fallback handling: Normal execution process: After the first subtask is completed and the result is verified, the system automatically activates node 4 to execute the fourth subtask. After the second subtask is completed and the result is verified, wait for the eighth subtask to output the social security information of employees to be adjusted; After the fourth subtask is completed, activate node 4 to execute the ninth subtask, and activate node 8 to execute the fifth to eighth subtasks; After the eighth subtask is completed, activate node 7 to execute the tenth subtask.
[0072] Example of triggering the fallback strategy: Suppose a node hardware failure occurs when node 8 is executing the seventh subtask. The system triggers the fallback strategy: automatically switch to the backup general node 11, reacquire the difference list data and execute the seventh subtask to ensure that the process is not interrupted.
[0073] S110: Summary Output of Results The system summarizes the results of each subtask according to five acquisition objectives, generates a structured query report, and outputs it to the technology company. Employee information sheet for last month (500 complete records); Latest social security policy data and timeline of policy changes; This month's employee information discrepancy list (including details of personnel additions, deletions, and information changes); List of employees whose social security contributions need to be adjusted this month (accurately matching policy requirements); This month's employee social security adjustment cost details and total amount.
[0074] Figure 2 This is a schematic diagram illustrating the main components of the enterprise information query task allocation system in an embodiment of the present invention. (See diagram below.) Figure 2 As shown, the enterprise information query task allocation system 1 provided in this embodiment of the invention includes a request parsing module 10, a subtask generation module 20, a task analysis module 30, a set construction module 40, a time calculation module 50, a collection module 60, a node matching module 70, and a result output module 80.
[0075] Request parsing module 10 is used to receive enterprise login requests and perform enterprise identity authentication, obtain the target information query requirements submitted by the enterprise, identify the acquisition targets corresponding to the target information query requirements, and the number of acquisition targets is at least one; Subtask generation module 20 is used to obtain a list of subtasks corresponding to each acquisition target based on a preset rule base; The task analysis module 30 is used to identify the pre- and post-relationships between each subtask through a task dependency analysis algorithm and output a subtask dependency graph. Set construction module 40 is used to construct a subtask set based on the number of acquisition targets, the subtask list, and the subtask dependency graph; Time calculation module 50 is used to calculate the task processing time of each subtask in the subtask set based on a pre-built time estimation model; The acquisition module 60 is used to collect the node status data of each task processing node in real time. The node matching module 70 is used to filter a set of candidate nodes for each subtask based on a pre-built node capability matrix, filter the optimal target node from the set of candidate nodes based on the node status data and the task processing time, and assign each subtask to the corresponding optimal target node. For initial subtasks without prior dependencies, the corresponding optimal target node is triggered to execute immediately after task allocation is completed; For subtasks with prerequisites, monitor the execution status of the prerequisite subtasks. If the prerequisite subtask is completed and the result is verified, automatically activate the corresponding optimal target node to execute the subsequent subtask. If the prerequisite subtask fails to execute, trigger the corresponding fallback strategy according to the failure type. The results output module 80 is used to classify and summarize the execution results of each subtask according to the acquisition target, and output the enterprise information query results.
[0076] Figure 3 This is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention, such as... Figure 3As shown, the electronic device 90 includes: a processor 901, a memory 902, and a bus 903; The processor 901 and the memory 902 communicate with each other via the bus 903. The processor 901 is used to call program instructions in the memory 902 to execute the methods provided in the above-described method embodiments, and to execute the methods provided in the embodiments of the present invention.
[0077] This embodiment provides a non-transitory computer-readable storage medium that stores computer instructions, which cause a computer to execute the method provided in this embodiment of the invention.
[0078] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various storage media capable of storing program code, such as ROM, RAM, magnetic disk, or optical disk.
[0079] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.
Claims
1. A method for allocating enterprise information query tasks, characterized in that, include: Receive enterprise login request and perform enterprise identity authentication, obtain the target information query request submitted by the enterprise, identify the acquisition target corresponding to the target information query request, and the number of acquisition targets is at least one; Obtain a list of sub-tasks corresponding to each acquisition target based on a preset rule base; The task dependency analysis algorithm identifies the pre- and post-relationships between each subtask and outputs a subtask dependency graph. Construct a set of subtasks based on the number of acquisition targets, the subtask list, and the subtask dependency graph; The task processing time of each subtask in the subtask set is calculated based on a pre-built time estimation model. Real-time collection of node status data for each task processing node; Based on the pre-built node capability matrix, a set of candidate nodes is selected for each subtask. Based on the node status data and the task processing time, the optimal target node is selected from the set of candidate nodes, and each subtask is assigned to the corresponding optimal target node. For initial subtasks without prior dependencies, the corresponding optimal target node is triggered to execute immediately after task allocation is completed; For subtasks with prerequisites, monitor the execution status of the prerequisite subtasks. If the prerequisite subtasks are completed and the results are verified, automatically activate the corresponding optimal target node to execute the subsequent subtasks. If the preceding subtask fails to execute, the corresponding fallback strategy is triggered based on the failure type. The results of each subtask are categorized and summarized according to the target obtained, and the enterprise information query results are output.
2. The method for allocating enterprise information query tasks according to claim 1, characterized in that, The acquisition objectives include obtaining the employee information table for the previous month, obtaining social security policy data, obtaining the policy change time series, obtaining the employee information difference list for the current month, obtaining the employee information table for the current month, obtaining information on newly added employees for the current month, obtaining information on employees who left the company for the current month, obtaining information on employees whose information has changed for the current month, obtaining information on employees whose social security needs to be adjusted for the current month, obtaining the social security adjustment costs for employees for the current month, and obtaining employee relations management information for the current month.
3. The method for allocating enterprise information query tasks according to claim 1, characterized in that, The subtasks include the first subtask, the second subtask, the third subtask, the fourth subtask, the fifth subtask, the sixth subtask, the seventh subtask, the eighth subtask, the ninth subtask, the tenth subtask, and the eleventh subtask: The first subtask is to obtain the employee information table of the corresponding enterprise for the previous month based on the enterprise login request with enterprise identification; The second sub-task is to obtain social security policy data, construct a policy change time series based on the social security policy data, and record the effective time, geographical scope, applicable population and specific clauses of each social security policy; The third subtask is to obtain the initial employee information table for this month; The fourth subtask is to calculate the first comprehensive difference value between the initial employee information table for this month and the employee information table for last month, calculate the second comprehensive difference value between the policy change time series and the initial employee information table for this month, and determine a difference list based on the first comprehensive difference value and the second comprehensive difference value; The fifth subtask is to obtain information on newly added employees based on the aforementioned discrepancy list; The sixth subtask is to obtain information on departing employees based on the aforementioned discrepancy list; The seventh subtask is to modify employee information based on the information obtained from the aforementioned discrepancy list; The eighth subtask is to obtain the employees whose social security contributions need to be adjusted based on the aforementioned discrepancy list; The ninth subtask is to generate the current month's employee information table based on the difference list, the initial current month's employee information table, and the previous month's employee information table; The tenth subtask is to calculate the social security adjustment cost for employees whose social security contributions need to be adjusted. The eleventh subtask is to generate the current month's employee relations management information based on the current month's employee information table and the historical employee information table for the past year.
4. The method for allocating enterprise information query tasks according to claim 1, characterized in that, The process involves using a task dependency analysis algorithm to identify the pre- and post-relationships between each subtask and outputting a subtask dependency graph, including: Calculate the dependency strength value between subtask k and subtask k+1. When the dependency strength value is 1, determine that subtask k is the predecessor of subtask k+1 and the successor of subtask k. When the dependency strength value is 0, it is determined that subtask k and subtask k+1 have no dependency relationship. Using each subtask as a node, and the preceding and following relationships of the subtasks as directed edges, with the direction of the directed edges pointing from the preceding subtask to the following subtask, a subtask dependency graph is constructed and output.
5. The method for allocating enterprise information query tasks according to claim 1, characterized in that, The construction of a subtask set based on the number of acquisition targets, the subtask list, and the subtask dependency graph includes: When the number of targets is one, the execution order of global subtasks is determined based on the subtask list and the subtask dependency graph, and a subtask set is generated. When there are multiple targets, the multiple sub-task lists are merged, deduplication is performed by using the unique identifier of the sub-task, the pre-relationships and post-relationships of the global sub-tasks after deduplication are sorted out based on the sub-task dependency graph, the global execution order is determined, and a set of sub-tasks is generated.
6. The method for allocating enterprise information query tasks according to claim 1, characterized in that, The calculation of the task processing time of each subtask in the subtask set based on the pre-built time estimation model includes: The task processing time of subtask k is calculated using formula (1); Official (1); In the formula, Let k be the task processing time for subtask k. The data volume feature value for subtask k. Let be the complexity coefficient of subtask k. The performance coefficient of the task processing node. The historical execution efficiency of subtask k. represents the adaptation coefficient between subtask k and task processing node.
7. The method for allocating enterprise information query tasks according to claim 6, characterized in that, The step of selecting the optimal target node from the candidate node set based on the node status data and the task processing time includes: The fitness value of task processing node m for processing subtask k is calculated using formula (2). Official (2); In the formula, The fitness value for processing subtask k for task processing node m. represents the matching degree coefficient between task processing node m and the subtask processing model. This represents the base processing performance value for task processing node m. Let m be the number of tasks currently being handled by the task processing node. Let k be the estimated processing time for subtask k. As the task load weight, Time weighting; The task processing node with the highest fit value is determined as the optimal target node.
8. The method for allocating enterprise information query tasks according to claim 1, characterized in that, If the preceding subtask fails, a fallback strategy is triggered based on the failure type, including: If the task processing node fails, it will automatically switch to the backup task processing node to retry. If the error is due to the task processing node itself, terminate the execution of subsequent subtasks and push an exception notification to the enterprise.
9. A system for allocating enterprise information query tasks, characterized in that, include: The request parsing module is used to receive enterprise login requests and perform enterprise identity authentication, obtain the target information query requirements submitted by the enterprise, identify the acquisition targets corresponding to the target information query requirements, and the number of acquisition targets is at least one. The subtask generation module is used to obtain a list of subtasks corresponding to each acquisition target based on a preset rule base. The task analysis module is used to identify the pre- and post-relationships between each subtask through a task dependency analysis algorithm, and output a subtask dependency graph. A set construction module is used to construct a subtask set based on the number of acquisition targets, the subtask list, and the subtask dependency graph. The time calculation module is used to calculate the task processing time of each subtask in the subtask set based on a pre-built time estimation model; The data acquisition module is used to collect the node status data of each task processing node in real time. The node matching module is used to filter a set of candidate nodes for each subtask based on a pre-built node capability matrix, filter the optimal target node from the set of candidate nodes based on the node status data and the task processing time, and assign each subtask to the corresponding optimal target node. For initial subtasks without prior dependencies, the corresponding optimal target node is triggered to execute immediately after task allocation is completed; For subtasks with prerequisites, monitor the execution status of the prerequisite subtasks. If the prerequisite subtasks are completed and the results are verified, automatically activate the corresponding optimal target node to execute the subsequent subtasks. If the preceding subtask fails to execute, the corresponding fallback strategy is triggered based on the failure type. The results output module is used to classify and summarize the execution results of each subtask according to the acquisition target, and output the enterprise information query results.
10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 8.