A talent empowerment management system based on big data

Abstract

The present application discloses a talent empowerment management system based on big data, which relates to the field of information management technology, and includes: an operation distribution module, an instruction mapping module, a hierarchical processing module and a correction decision module; a talent status tree is generated through personnel information stored in a blockchain, and the operation sequence of each personnel and the project distribution instructions are identified according to the operation instructions of the talent status tree on each blockchain node; a target status node to be migrated is generated from the talent status tree, and a mapping relationship between the target status node and the talent status tree is established based on the migration information of each target status node; according to the mapping relationship between the target status node and the talent status tree, the forwarding receipt instructions of each target status node after the operation instruction is distributed are identified, the target status nodes are graded, and the correction type for each project for personnel operation is set; a management decision plan is set according to the correction type; the value conversion rate of human resources and management efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the field of information management technology, and in particular to a talent empowerment management system based on big data. Background Art

[0002] Traditional talent management models suffer from significant flaws. For one thing, there's a lack of a timely and effective response mechanism for deviations in personnel execution. This prevents flexible adjustments to personnel instructions based on actual circumstances. Consequently, when multiple tasks are being executed concurrently, personnel struggle to quickly and efficiently match task requirements, leading to low execution efficiency.

[0003] On the other hand, when it comes to tracking personnel capabilities, the specific circumstances of project distribution and execution are not fully considered. This lack of targeted approach also reduces the efficiency of distribution and execution. Furthermore, under the traditional model, personnel operational data is difficult to review retrospectively using technologies such as blockchain, further limiting the transparency and traceability of talent management.

[0004] For example, Chinese Patent Publication No. CN115392804A discloses a talent empowerment method and system based on big data, wherein the talent empowerment system includes: a data verification module that maintains unstructured data stored by a general-purpose computer and monitors the equipment associated with the general-purpose computer; a talent management module that determines the personnel characteristics of a staff member and sends them to the data verification module; and a task management module that assigns work item tasks related to maintaining the general-purpose computer and its associated equipment to the staff member through the data verification module. The present invention also relates to a talent empowerment method based on big data, including a data verification module that maintains unstructured data stored by a general-purpose computer and monitors the equipment associated with the general-purpose computer; a talent management module that is coupled to an operational capability extreme value testing module to determine the personnel characteristics of the staff member through the operational capability extreme value testing module; and a task management module that assigns work item tasks related to maintaining the general-purpose computer and its associated equipment to the staff member through the data verification module.

[0005] For example, Chinese patent publication number CN117291559A discloses an intelligent enterprise talent management system, which includes a talent pool management module, an enterprise architecture management module and an evaluation and analysis module; wherein the talent pool management module is used to establish an enterprise talent resource pool, and store and manage the talent files of enterprise talents based on the enterprise talent resource pool; the enterprise architecture management module is used to build an enterprise organizational structure according to actual conditions, wherein the enterprise organizational structure contains position information and information on employees corresponding to the positions, wherein the position information includes position descriptions and required conditions for the positions; the evaluation and analysis module is used to perform evaluation and analysis based on the position information in the enterprise organizational structure and the corresponding information on employees, and obtain evaluation and analysis results for employees; and / or perform evaluation and analysis based on the position information in the enterprise organizational structure and the talent files of enterprise talents in the enterprise talent resource pool, and obtain talent recommendation evaluation and analysis results.

[0006] The existing technologies respectively describe pushing special projects based on the behavioral characteristics of talents, and completing personnel interview management through methods such as image recognition. However, the personnel information obtained in the existing technologies is easily scattered across various systems, making global coordination difficult. At the same time, there is a lack of real-time status tracking of personnel, which leads to delays in subsequent personnel allocation and task distribution, causing resource conflicts or efficiency losses. Summary of the Invention

[0007] The embodiment of the present application provides a talent empowerment management system based on big data, including: an operation distribution module for generating a talent status tree based on personnel information stored in the blockchain, and identifying the operation sequence and project distribution instructions of each personnel based on the operation instructions of the talent status tree on each blockchain node.

[0008] The instruction mapping module is used to generate target state nodes to be migrated from the talent state tree, and establish a mapping relationship between the target state nodes and the talent state tree based on the migration information of each target state node.

[0009] The hierarchical processing module is used to identify the forwarding receipt instructions of each target state node after the operation instruction is distributed according to the mapping relationship between the target state node and the talent state tree, classify the state of the target state node, and set the correction type for each project for personnel operations.

[0010] The correction decision module is used to regulate personnel management according to the correction type and set management decision plans.

[0011] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The present invention stores personnel information through blockchain, and uses its tamper-proof and decentralized characteristics to ensure data security; maps personnel information into a chain node set, and combines the nodes into a multi-level talent status tree by expanding them; to prevent data asynchrony problems when personnel are allocated data, and realize trusted traceability under multi-data task distribution, thereby completing database reliability management.

[0012] 2. The present invention calculates the matching degree between personnel and projects, determines the coupling extreme value through the maximum-minimum method, divides the maximum matching direction and the minimum matching direction, sets the forwarding level of each operation instruction according to the instructions in these directions, and then adjusts the priority of project distribution to improve the accuracy of the distribution and matching of each operation instruction and prevent problems such as matching errors.

[0013] 3. The present invention establishes a mapping relationship between the target state node and the talent state tree based on the migration information and migration path of each data, ensuring the traceability of the instruction status distributed to each person. At the same time, after setting scores for these traceable parts, the changes in the personnel information of each operation instruction under the number of migrations are quantified. After achieving differentiated grading of different personnel statuses, the execution plan of each person is adjusted to prevent resource conflicts and efficiency losses caused by the allocation of tasks related to operation instructions, thereby improving the efficiency of personnel management. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a system framework diagram of a talent empowerment management system based on big data in the present invention.

[0015] Figure 2 This is a flow chart of an operation distribution module of a talent empowerment management system based on big data in the present invention.

[0016] Figure 3 This is a flow chart of the instruction mapping module of a talent empowerment management system based on big data in the present invention.

[0017] Figure 4 This is a flow chart of a hierarchical processing module of a talent empowerment management system based on big data in the present invention.

[0018] Figure 5 This is a flow chart of a correction decision module of a talent empowerment management system based on big data according to the present invention. DETAILED DESCRIPTION

[0019] To facilitate understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains; the terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0021] like Figure 1 As shown, the present application provides a talent empowerment management system based on big data, including: an operation distribution module for generating a talent status tree based on the personnel information stored in the blockchain, and identifying the operation sequence and project distribution instructions of each personnel based on the operation instructions of the talent status tree on each blockchain node.

[0022] The instruction mapping module is used to generate target state nodes to be migrated from the talent state tree, and establish a mapping relationship between the target state nodes and the talent state tree based on the migration information of each target state node.

[0023] The hierarchical processing module is used to identify the forwarding receipt instructions of each target state node after the operation instruction is distributed according to the mapping relationship between the target state node and the talent state tree, classify the state of the target state node, and set the correction type for each project for personnel operations.

[0024] The correction decision module is used to regulate personnel management according to the correction type and set management decision plans.

[0025] In the operation distribution module, the talent status tree is a data structure built based on the personnel information stored in the blockchain. It resembles a tree, with nodes representing personnel information or personnel-related status data, such as basic information such as name, age, skills, current task availability, and ongoing tasks. This data is also reflected in competency assessment results, such as professional competency level and performance. This data is displayed in a tree-like structure, showcasing each person's actual capabilities. The tasks or instructions performed by each person are then annotated to identify their decision-making and management in empowerment scenarios.

[0026] Operational instructions are specific commands issued on blockchain nodes for the talent status tree or related personnel information. Examples include assigning a new task to the person represented by a node, updating a person's competency assessment data, or changing a person's task status from "in progress" to "completed." These instructions drive the talent management process and ensure that personnel information remains synchronized with actual business operations.

[0027] The operation sequence refers to the order in which the various operation instructions are executed during the operation of the talent status tree. Different operation sequences may lead to different results. For example, if you first execute the instruction of "update personnel skill information" and then execute the instruction of "assign new tasks based on new skills", the personnel task arrangement and project execution status obtained may be very different from assigning tasks first and then updating skill information. A reasonable operation sequence can ensure the logic and effectiveness of the talent management process. When setting multiple instruction tasks for a single person or multiple people, identifying the operation sequence can understand the work process and work content for the current operator, so as to facilitate the management of each employee's capability plan, so that in the empowerment scenario, the progress of each project and the allocation of personnel can be easily traced back.

[0028] A project dispatch instruction is a command that explicitly assigns a specific project to a relevant person. It contains detailed information about the project, such as the project name, objectives, requirements, etc., as well as the identity of the person to be assigned.

[0029] like Figure 2 As shown, the implementation method of the operation distribution module also includes: obtaining the regional chain database interface, using the personnel information at each interface, setting leaf nodes according to the number of personnel read, and connecting each leaf node with the connection relationship of the leaf node in the regional chain to form a chain node set.

[0030] Determine whether each leaf node in the chain node set is an extension node of each other, and combine each leaf node with the extension node, and use the timestamps of each leaf node after combination to form a talent status tree.

[0031] Extension nodes are nodes in a chained set that can establish additional connections or relationships with other nodes. They can represent relationships between people, such as partnerships, hierarchical relationships, or complementary skills. By combining the data from these existing nodes to determine the connections between multiple nodes, these connected nodes are considered a talent status tree.

[0032] For example, the extension nodes can be the skills or professional areas of the personnel, the roles or responsibilities of the personnel in the project, the historical cooperation relationship between the personnel, and the geographical location or time zone of the personnel.

[0033] The implementation method for determining whether each leaf node in the chain node set is an extension node of each other is as follows: if each leaf node is an extension node of each other, the data corresponding to the leaf node are formed into a data collection, the data collection is used as the root node, the extension node is used as its branch node, and the corresponding leaf node with the extension node relationship is used as its leaf node.

[0034] If any leaf nodes in a chain node set are not expansion nodes, the non-expansion leaf nodes are connected one by one according to the connection relationships in the regional chain to form a talent status tree. The connection relationships in the regional chain generally describe the order in which the leaf nodes are stored in the regional chain. This order can represent the order in which operation instructions are issued in certain scenarios and the order in which data is updated after the operation instructions are issued. Using this order as the composition of the talent status tree focuses on the execution process of each person's operation instructions in the time storage order.

[0035] When distributing operation instructions, the method of obtaining the project distribution instructions also includes: judging the matching degree between each person and the execution operation instruction based on the operation sequence, and taking the maximum and minimum values ​​during the matching degree screening as the project coupling extreme values.

[0036] Operation instructions are extracted according to the maximum matching direction and minimum matching direction in the project coupling extreme value. Based on the characteristic information contained in the operation instructions, the forwarding levels of the corresponding operation instructions in the maximum matching direction and the minimum matching direction are set. After distributing each forwarding level, the output is the project distribution instruction.

[0037] The operation sequence represents the series of steps required to complete a project, or the order in which operational instructions are executed. When determining the compatibility between personnel and the execution of operational instructions, the following dimensions can be used to calculate each personnel's compatibility. For example, compatibility can be scored based on factors such as personnel skills, experience, and availability. The weighted sum of these scores is used as the compatibility score. Weights are set in the order of skills, experience, and availability: 0.4, 0.3, and 0.3. These weights can also be set based on the average of the weights used in historical personnel compatibility calculations.

[0038] The specific dimensions of match degree describe historical performance data, such as whether the person possesses the required expertise to perform the operation, the person's past success rate in performing similar operations, and whether the person is currently available for assignment. Specifically, if a person possesses skills, their skill score is set using a logical value, such as 0 or 1. Their experience score is calculated as the ratio of their successful execution frequency to the total number of tasks performed by all other people of the same type. Their availability score is calculated as the ratio of their available time to their assigned time. This determines the match degree between each person and the current operation instruction.

[0039] When subsequently screening candidates, the maximum and minimum values ​​of the current matching degree are determined. These values ​​represent the maximum and minimum values ​​that exist when each talent status tree displays each node in a connected relationship. For example, a maximum and minimum value are obtained for the left branch of the talent status tree, and a maximum and minimum value are also obtained for the right branch of the talent status tree. The maximum and minimum values ​​that exist for each path that the talent status tree can describe are used as the output project coupling extreme value. This project coupling extreme value is mainly used to identify whether there are corresponding mismatch issues in the corresponding nodes after the current talent status tree divides the data into multiple nodes.

[0040] Afterwards, when extracting operation instructions for the maximum matching direction and the minimum matching direction, they are processed according to the operation instructions with the highest matching degree and the instructions with the lowest matching degree. The maximum matching direction indicates that the node of the talent status verification tree continuously moves from its root node to the connection path of the leaf node. There is a connection path with the maximum and minimum values ​​after subtracting the maximum and minimum values ​​of the matching degree in the project coupling extreme value. The path with the largest subtraction between the maximum and minimum values ​​of the matching degree on the connection path is regarded as the maximum matching direction, and the path with the smallest value after subtraction is regarded as the minimum matching direction.

[0041] The maximum matching direction at this point is the path with the greatest mismatch, which directly identifies the paths with the most complex coupling relationships and the most significant fluctuations in the project. These paths are often areas of concentrated risk or efficiency bottlenecks in the project. For example, multiple modules are highly coupled, where a delay in a single link can trigger a chain reaction, or where key resources such as manpower and equipment conflict across multiple tasks. Identifying this direction allows for early prediction and intervention at high-risk nodes, such as adding redundant resources and adjusting task priorities. It also allows for the concentration of core resources, such as expert teams and high-performance equipment, on critical paths, avoiding fragmented investment.

[0042] The minimum matching direction is the path with the smallest mismatch, identifying the path with the most stable coupling and controllable execution. This path typically occupies a standardized process area or low-risk area within the project, such as a validated codebase or standardized operating procedures, or parallel tasks with no external dependencies. By identifying this path, resources can be freed up, facilitating the reallocation of human resources to support related tasks on other paths.

[0043] That is, the method of obtaining the maximum matching direction and the minimum matching direction includes: according to the project coupling extreme value, from the connection path of each node in the talent status tree, extracting the path with the maximum matching degree difference and the path with the minimum matching degree difference in the connection path as the maximum matching direction and the minimum matching direction.

[0044] When the structure of the talent status tree is in a chain state, the maximum matching direction and minimum matching direction it identifies will be processed based on part of the data in the talent status tree composed of multiple personnel information. That is, the nodes with the maximum and minimum matching degree extreme values ​​are used as the two ends of the data acquisition interval, and the data within the data interval is used as the data content of the maximum matching direction and the minimum matching direction. At this time, the minimum matching direction will select the part with the smallest matching degree difference. This part indicates that the matching degree of personnel is small when performing related tasks, which belongs to a relatively fixed mechanized execution task.

[0045] The implementation method for setting the forwarding level is as follows: the nodes existing in the maximum matching direction and the minimum matching direction are taken as the target nodes, and based on the matching value and connection relationship of the target node, the preset forwarding rules of the target node are extracted. At this time, the preset forwarding rules will set the interval according to the extreme difference generated by the matching degree. For example, if the matching degree extreme difference is greater than 80%, it is regarded as the highest priority, and the matching degree extreme difference between 30% and 80% is set as the secondary priority, and the matching degree extreme difference is less than 30% and the minimum priority is set; in addition to the current content, this preset forwarding rule can also be adjusted according to the matching degree extreme difference value interval extracted from the historical data, according to the 30% and 80% of its value interval to set the endpoints to adjust the currently set preset forwarding rules.

[0046] The forwarding level is set for the target node according to the corresponding priority in the preset forwarding rule, and the operation instruction after setting the forwarding level is output as the project distribution instruction, so as to illustrate the corresponding priority when the current personnel's operation instruction is set. At this time, the data processed is prone to risks and no risks when handled by personnel. As for other data not covered, the default priority of each operation instruction is used, which will be pre-set in the database.

[0047] In the instruction mapping module, the main task is to identify the migration path of the target state node and determine how the target state node is migrated. For example, starting from the talent state tree, two migration paths are divided, such as skill upgrade migration and project experience migration. These two migration paths point to two target state nodes. The obtained target state nodes represent the changes in the capabilities of the corresponding personnel under the talent empowerment system or the adjustments of the recorded data in the regional chain. These mapping relationships are then stored to record the content of each personnel's update each time. Status synchronization is achieved through the cross-chain protocol to complete the synchronization of personnel management information under multiple data sources.

[0048] like Figure 3 As shown, the implementation method of the instruction mapping module includes: obtaining a preset migration status rule, screening the talent status tree with the migration status rule, and obtaining the target status node to be migrated.

[0049] Extract the migration path of the target state node, map the migration information contained in the migration path according to the position of each target state node, and set the mapping relationship between the target state node and the talent state tree.

[0050] The migration status rule may be to filter the parts that need to be migrated in multiple dimensions such as skills and experience using the migration status rule to identify the parts of the current personnel to be migrated.

[0051] Migration status rules are criteria used to filter the portions of the talent status tree that require migration. These rules can be based on multiple dimensions, including but not limited to skills, experience, project participation, and timestamps. For example, the skill dimension might filter out skill nodes with a certain proficiency level; the experience dimension might filter out nodes with project participation of a certain type or duration exceeding a threshold; and the time dimension might filter out status nodes updated or created within a specific timeframe. The migration status rules are then used to traverse each node in the talent status tree, searching for nodes that meet the migration status rules.

[0052] The migration information described above refers to the core data units that can be synchronized to other data systems after screening of personnel-related capabilities, and is used to represent the manifestation of personnel's own capabilities in different dimensions.

[0053] Specifically, the implementation method for filtering the talent status tree using migration status rules also includes: obtaining at least one data migration range from the talent status tree based on the migration status rules, and obtaining a time series node set based on the chronological order of the data within the data migration range. The time series node set represents a data set consisting of multiple nodes within the data migration range. The data migration range represents the set of nodes in the talent status tree that need to be migrated, as filtered by the migration status rules. This range can be one or more subtrees or specific types of nodes, such as skill nodes or project nodes.

[0054] The corresponding starting and ending points in the time series node set are output as the target state nodes, and the subtrees of the talent state tree corresponding to each data migration range are output as the migration path. The starting and ending points in the time series node set represent the main factors affecting the personnel information migration or the data included after the data migration range is converted into a time series node set. They can include major changes or key events during data migration to identify the main circumstances of the current personnel information migration. Selecting the starting and ending points ensures the integrity of data processing and facilitates understanding of how each personnel information is migrated when executing operation instructions.

[0055] When extracting the migration path, for each target state node to be migrated, the path from the root node to the current node is extracted; the attributes, dependencies, and timestamps of the nodes contained in the path are used as part of establishing the mapping relationship. The extracted migration path is then used as the silver-gray part to ensure that the migrated node can be traced back to its position and context in the original talent status tree to identify whether the information and status contained in the personnel have changed. Based on these changes, the relative status of each person can be dynamically adjusted to facilitate the subsequent adjustment of the final output content for the talent status tree of different personnel, improve the efficiency of talent management, and reduce manual intervention and errors. The updated content of the relevant information status of the personnel is focused on as the basis for identifying their capabilities, achieving adaptability to personnel empowerment management.

[0056] At the same time, for the mapping relationship between the target status node and the talent status tree, it is also necessary to ensure that the node where the personnel information is stored can complete the processing of its data after the predecessor node is transferred. The predecessor node generally indicates the previous node that the current personnel information depends on. That is, when constructing the mapping relationship, it is also necessary to set the mapping relationship between the target status node and the talent status tree in the form of a bidirectional construction.

[0057] In the hierarchical processing module, the information of each person is hierarchically processed, and the forwarding receipt instructions present therein are identified. After the hierarchical processing, the correction type of the current person's operation is explained.

[0058] The migration information of the target state node will be used to identify whether there are dependencies between the target state nodes. For example, independent nodes without dependencies represent relevant content of basic skill certification, nodes with a single parent node dependency can be represented as relevant content of project experience, and there are nodes that require multiple loops. Assuming that the target state node mainly contains project A experience at this time, the traversal starts from the project A experience, traverses to project management skills, traverses from project management skills to project B experience, and then traverses from project B experience back to project A experience. At this time, there is a corresponding traversal loop. At this time, it is necessary to obtain the migration information. After identifying these convenient contents, generate a forwarding receipt instruction for each target state node, and then grade each target node to know whether the currently executed operation instruction needs to be corrected, thereby completing the management and task allocation of each personnel information in the blockchain and other methods.

[0059] The implementation method of the hierarchical processing module also includes: based on the mapping relationship between the target state node and the talent state tree, migrating the target node for authentication, and identifying the inspection items of the current target detection node in the form of continuous timestamp verification.

[0060] The inspection items include node integrity, dependency continuity and timestamp consistency. Node integrity indicates whether the personnel information contained in the current target state node is complete. The completeness at this time indicates whether the personnel information is lost or missing after being divided into each node. Dependency continuity is to identify the dependency relationship corresponding to the current target state node. This dependency relationship can be verified by whether other nodes can be mapped to the current target state node when traversing the loop, and whether there is an extension node in the target state node. When the dependency relationship does not exist and the mapping is empty, this part of the inspection is considered normal. As for timestamp consistency, it is to verify whether there is a time point inconsistency in the data contained in the target state node. If they are completely consistent, the current inspection item is considered normal and migration certification can be performed.

[0061] Using the inspection items of the current target state node, the mapping relationship between the target state nodes before and after the migration is output as a forwarding receipt instruction. The forwarding receipt instruction is used to describe the relevant signs of skill updates and project progress after the corresponding data of the current node is migrated, so as to complete the forwarding of each operation instruction and monitor its implementation.

[0062] For example, the forwarding receipt instruction may include: an operation type code, a state change field, and an exception flag bit.

[0063] For example, the operation type code: 0x01 = skill update, 0x02 = project completion; the key-value pair set of the status change field, such as: {"proficiency":"expert level"}; the binary bit mask represented by the exception flag bit, such as inconsistent data in bit 0, abnormal permissions in bit 1, etc. The exception flag is used to identify the content of the target status node check through its check items.

[0064] The final target state node status classification can describe the instructions distributed by the current operation instructions to each person based on completion timeliness, resource consumption rate and quality compliance.

[0065] The status score is set by weighted summing the ratio of the time corresponding to completion timeliness to the standard time, the ratio of resources consumed in the resource consumption rate to the standard resources, and the ratio of the quality of completed operation instructions in the quality compliance to the average value of historical data. This score can quantify the content of the completed operation instructions to describe the score under its status classification. The usage weights can be set to 0.4, 0.3, and 0.3 respectively.

[0066] like Figure 4 As shown, the implementation method of the hierarchical processing module also includes: setting a status score based on the completion timeliness, resource consumption rate and quality compliance of the target status node.

[0067] The state score and marking time of the target state node are extracted based on the migration times of the target node, and the state volatility is set based on the standard deviation of the state score in the corresponding time series.

[0068] Record the time interval between the previous status score and the current status score, extract the status interval time, and set the grading weight based on the maximum number of migrations for each target node.

[0069] Using the grading weight and status score, each target node is divided to obtain the status classification after division, and the correction type is set based on the value range of the status classification.

[0070] The above migration times represent the cumulative number of times the target state node changes after receiving the operation instruction, such as starting, paused, in progress and completed, to record the completion status of each person for the operation instruction.

[0071] Its state volatility can be expressed as, for each target state node, collecting its state score, such as 、 ,,, ; and its corresponding timestamp , at this time n represents the number of collected data, then the state fluctuation rate It is expressed as follows.

[0072] ;in, represents the i-th state score, Represents the average state score. The state volatility obtained at this time is used to identify whether the state score of each target state node is relatively stable after executing the corresponding operation instruction. The state interval indicates the time period between two state score reductions when obtaining. The subsequent grading weight is the ratio of 1 minus the number of migrations of the current target state node minus the maximum number of migrations among each state node. After using this value as the grading weight, the product of this grading weight and the state score is used as the number of times to identify the state grade value, and the correction type is output according to the interval it falls in.

[0073] As shown in Table 1, its correction type can be expressed as follows.

[0074] Table 1. Correction type diagram

[0075]

[0076] The correction type is based on the range of values ​​calculated after the status score is used for the data under the status classification. For example, if the accuracy value is calculated by the judgment matrix and is in the range of 0.9 to 1, it is set as no correction is required, and 0.7 to 0.9 is set as a slight correction to mark its implementation content. For 0.5 to 0.7, it is necessary to verify whether the process is reasonable under the current project distribution, whether there is some node information that has not been updated, resulting in the current distribution content not meeting the current project requirements. As for 0 to 0.5, it means that readjustment is needed, some data anomalies exist, and the data needs to be processed. At the same time, machine learning models such as isolation forests can be introduced to detect abnormal classification values ​​to complete the classification processing.

[0077] In the correction decision module, decisions are made based on the correction type, with the data associated with the target state node as input. That is, a decision tree model is constructed based on correction type → influencing factor analysis → solution generation → conflict detection → solution optimization. Ultimately, the distribution and regulation of project instructions based on different personnel information is completed, thereby improving personnel management efficiency.

[0078] like Figure 5 As shown, the implementation method of the correction decision module includes: classifying each correction type based on the acquired correction type, performing data mapping with the influencing factors corresponding to the classified correction type, and generating candidate solutions corresponding to the current correction type.

[0079] Determine whether each candidate solution meets the conflict detection requirements, and use the optimal solution under conflict detection as the output management decision solution.

[0080] The above-mentioned conflict detection can be completed by checking the time overlap, and checking whether there is a time overlap part in each candidate solution. If not, the corresponding candidate solution is used as the part to complete the conflict detection.

[0081] When classifying the correction type, the data set represented by the correction type is used as the root node of the decision tree, and the root node is divided using Gini impurity to obtain the intermediate nodes of the decision tree. The leaf nodes of the decision tree are then divided using Gini impurity. When all nodes in the decision tree are optimal, the current correction type classification is considered completed, and the influencing factors contained in each leaf node are regarded as the influencing factors corresponding to its correction type. At the same time, when obtaining candidate solutions, the current influencing factors and the correction type expressed on the leaf node are searched from the preset solution library. When the cosine similarity value calculated between the influencing factors and the data of the correction type and the preset solution library is the largest, the corresponding solution is regarded as a candidate solution. At the same time, the optimal solution under conflict detection represents the solution that meets the conflict detection requirements and has the highest task completion efficiency. It is regarded as the optimal solution and is output as the management decision solution.

[0082] It should be noted that when obtaining influencing factors, the high-frequency factor combination in the leaf nodes of the decision tree is used for extraction. For example, the description content of the leaf nodes about insufficient personnel skills and task dependency conflicts is used to extract the frequent item sets in each leaf node, and the part with the highest frequency of occurrence in the frequent camera is used as the identified influencing factor, which will be used as the part for subsequent screening of influencing factors.

[0083] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A talent empowerment management system based on big data, characterized by: include: The operation distribution module is used to generate a talent status tree based on the personnel information stored in the blockchain, and identify the operation sequence and project distribution instructions of each personnel based on the operation instructions on each blockchain node in the talent status tree; An instruction mapping module is used to generate target state nodes to be migrated from the talent state tree, and to establish a mapping relationship between the target state nodes and the talent state tree based on the migration information of each target state node; The grading processing module is used to identify the forwarding receipt instructions of each target state node after the operation instruction is distributed based on the mapping relationship between the target state node and the talent state tree, classify the state of the target state node, and set the correction type for each project for personnel operations; The correction decision module is used to regulate personnel management according to the correction type and set management decision plans; Other ways to obtain project distribution instructions include: The matching degree between each person and the execution instruction is judged based on the operation sequence, and the maximum and minimum values ​​during the matching screening are used as the project coupling extreme values; Operation instructions are extracted according to the maximum matching direction and minimum matching direction in the project coupling extreme value. Based on the characteristic information contained in the operation instructions, the forwarding levels of the corresponding operation instructions in the maximum matching direction and the minimum matching direction are set. After distributing each forwarding level, the output is the project distribution instruction.

2. A talent empowerment management system based on big data according to claim 1, characterized in that: The implementation of the operation distribution module also includes: Obtain the regional chain database interface, use the personnel information at each interface, set leaf nodes according to the number of personnel read, and connect each leaf node based on the connection relationship of the leaf nodes in the regional chain to form a chain node set; Determine whether each leaf node in the chain node set is an extension node of each other, and combine each leaf node with the extension node, and use the timestamps of each leaf node after combination to form a talent status tree.

3. A talent empowerment management system based on big data as claimed in claim 2, characterized in that: The implementation method for determining whether each leaf node in the chain node set is an extension node of each other is as follows: If each leaf node is an extension node of another, the data corresponding to the leaf node is combined into a data collection, and the data collection is used as the root node, the extension node is used as its branch node, and the leaf node with the corresponding extension node relationship is used as its leaf node; If there are leaf nodes in the chain node set that are not expansion nodes, the leaf nodes that are not expansion nodes will be connected one by one according to the connection relationship in the regional chain to form a talent status tree.

4. A talent empowerment management system based on big data according to claim 1, characterized in that: The implementation method of setting the forwarding level is as follows: The nodes in the maximum matching direction and the minimum matching direction are taken as target nodes, and the preset forwarding rules of the target nodes are extracted based on the matching degree values ​​and connection relationships of the target nodes; The forwarding level is set for the target node according to the corresponding priority in the preset forwarding rule, and the operation instruction after setting the forwarding level is output as the project distribution instruction.

5. The talent empowerment management system based on big data according to claim 1, characterized in that: The implementation of the instruction mapping module includes: Obtain the preset migration status rules, filter the talent status tree based on the migration status rules, and obtain the target status node to be migrated; Extract the migration path of the target state node, map the migration information contained in the migration path according to the position of each target state node, and set the mapping relationship between the target state node and the talent state tree.

6. A talent empowerment management system based on big data according to claim 5, characterized in that: Other ways to implement filtering the talent status tree based on migration status rules include: Obtain at least one data migration range from the talent status tree according to the migration status rule, and obtain a time series node set according to the time sequence of each data in the data migration range; The corresponding starting point and end point in the time series node set are output as the target state node, and the subtree of the talent state tree corresponding to each data migration range is output as the migration path.

7. The talent empowerment management system based on big data according to claim 1, characterized in that: The implementation of the hierarchical processing module also includes: Based on the mapping relationship between the target state node and the talent state tree, the target node is migrated and authenticated, and the inspection items of the current target detection node are identified in the form of continuous timestamp verification; Using the check items of the current target state node, the mapping relationship between the target state nodes before and after the migration is output as a forwarding receipt instruction.

8. The talent empowerment management system based on big data according to claim 1, characterized in that: The implementation of the hierarchical processing module also includes: Set status scores based on the completion timeliness, resource consumption rate, and quality compliance of the target status node; The state score and marking time of the target state node are extracted based on the migration times of the target node. The state volatility is set based on the standard deviation of the state score in the corresponding time series. Record the time interval between the previous status score and the current status score, extract the status interval time, and set the grading weight based on the maximum number of migrations for each target node; Using the grading weight and status score, each target node is divided to obtain the status grade after the division, and the correction type is set based on the value range of the status grade.

9. The talent empowerment management system based on big data according to claim 1, characterized in that: The implementation of the revised decision module includes: Based on the acquired correction types, each correction type is classified, and data mapping is performed with the influencing factors corresponding to the classified correction types to generate candidate solutions corresponding to the current correction type; Determine whether each candidate solution meets the conflict detection requirements, and use the optimal solution under conflict detection as the output management decision solution.

Images