Digital performance evaluation method and system based on artificial intelligence

Abstract

The invention provides a digital performance evaluation method and system based on artificial intelligence, and belongs to the technical field of performance evaluation, and the method comprises the following steps: S101, obtaining HR assessment record data, employee task log data and employee real-time behavior data; s102, establishing a performance evaluation ID matched with the data, and performing data cleaning operation; s103, establishing a unified performance evaluation data graph and a difficulty-contribution degree association network based on cross-modal data alignment operation; s104, performing atlas performance evaluation operation and secondary performance evaluation operation to obtain an atlas evaluation numerical value and a network evaluation numerical value; s105, obtaining an evaluation threshold value, and carrying out performance grade matching operation; s106, performing subsequent work improvement evaluation operation based on the graph neural network model and the time sequence prediction model, and generating an employee performance report; according to the method, objectivity and fairness of employee performance evaluation can be improved, and subsequent work improvement suggestions related to employees can be provided.

Description

Technical Field

[0001] The present invention belongs to the technical field of performance evaluation, and more specifically, particularly relates to a digital performance evaluation method and system based on artificial intelligence. Background Art

[0002] In today's highly competitive business environment, the efficient operation and development of enterprises are inseparable from a scientific and reasonable performance evaluation system. Traditional performance evaluation methods are often affected by subjective factors during the evaluation process. It is difficult to unify the standards and scales of different evaluators, resulting in the lack of objectivity and fairness in the evaluation results. Moreover, traditional performance evaluation methods often cannot provide suggestions for employees' subsequent work improvement. Summary of the Invention

[0003] To solve the above technical problems, the present invention provides a digital performance evaluation method and system based on artificial intelligence to solve the technical problems in the prior art that traditional performance evaluation methods often cannot provide suggestions for employees' subsequent work improvement and have low objectivity and fairness in performance evaluation.

[0004] The purpose and efficacy of a digital performance evaluation method and system based on artificial intelligence of the present invention are achieved by the following specific technical means: A digital performance evaluation method based on artificial intelligence includes the following steps: S101: Obtain an employee evaluation database. The employee evaluation database is used to record and store relevant data of employees' daily work. Establish an API interface, establish a connection with the employee evaluation database through the API interface, so as to call relevant data in the employee evaluation database, and obtain HR assessment record data, employee task log data, and employee real-time behavior data based on the employee evaluation database; Among them, the HR assessment record data is used to evaluate the daily work status of employees, the employee task log data is used to evaluate the work results of employees, and the employee real-time behavior data is used to evaluate the daily work activity of employees; S102: Establish a performance evaluation ID. The performance evaluation ID represents the unique authentication ID of each employee and the data matching the employee. Match and establish a connection between the HR assessment record data, employee task log data, and employee real-time behavior data with the performance evaluation ID, perform data cleaning operations on the relevant data of the performance evaluation ID, and record the performance evaluation IDs with abnormal data in the data cleaning operation; S103: Perform cross-modal data alignment operations on the performance evaluation IDs without abnormal data. Based on the cross-modal data alignment operations, establish a unified performance evaluation data map for each performance evaluation ID that matches. The unified performance evaluation data map represents the comprehensive work performance of each performance evaluation ID, and establish a difficulty-contribution degree correlation network. The difficulty-contribution degree correlation network is used to comprehensively consider the workload and work task difficulty of each performance evaluation ID; S104: Perform map performance evaluation operations on the performance evaluation IDs without abnormal data based on the unified performance evaluation data map. After the evaluation is completed, generate map evaluation values. Perform secondary performance evaluation operations on the performance evaluation IDs without abnormal data based on the difficulty-contribution value correlation network. After the evaluation is completed, generate network evaluation values, and match and connect the map evaluation values and network evaluation values with the performance evaluation IDs; S105: Obtain the evaluation threshold. The evaluation threshold is used to perform performance level matching on the performance evaluation IDs. The evaluation threshold is regularly changed according to the market environment, and perform performance level matching operations on the map evaluation values and network evaluation values with the evaluation threshold; The performance level matching operations include: S1051: If both the map evaluation value and the network evaluation value are lower than the evaluation threshold, mark the performance evaluation ID as performance unqualified; S1052: If both the map evaluation value and the network evaluation value are within the proximity interval of the evaluation threshold, mark the performance evaluation IDs within the proximity interval as performance medium; S1053: If a single value among the map evaluation value and the network evaluation value is lower than the evaluation threshold, mark the performance evaluation ID as pending review. After the marking is completed, import the performance evaluation IDs to be reviewed into the expert review mechanism; S1054: If both the map evaluation value and the network evaluation value are higher than the proximity interval of the evaluation threshold, mark the performance evaluation ID as performance good; S106: After the performance level matching operations are completed, import the marked performance evaluation IDs into the graph neural network model and the time series prediction model for subsequent work improvement evaluation operations. The graph neural network model is used to perform employee ability analysis operations on the performance evaluation IDs, and the time series prediction model is used to perform evaluation growth trend operations on the performance evaluation IDs. Generate an employee performance report that matches the performance evaluation IDs based on the graph neural network model and the time series prediction model. The employee performance report includes the performance level and suggestions for subsequent work improvement, and forward the employee performance report to the employees and enterprise management levels that match the performance evaluation IDs; Regularly execute S101 - S106 to perform performance evaluation operations on enterprise employees.

[0005] As a further solution of the present invention, the expert review mechanism includes: Establish an artificial expert review panel. The artificial expert review panel conducts an artificial review on the performance evaluation ID to be reviewed. After the review is completed, the pending review mark of the performance evaluation ID is removed, and the performance evaluation ID is marked as good, medium or unqualified. After the marking is completed, the performance evaluation ID is re-imported into S106; Specifically, it includes the following steps: A1: Establish an artificial expert review panel. The artificial expert review panel includes multiple expert groups. Each of the multiple expert groups includes the human resources supervisor, department supervisor and data analysis expert of different departments; A2: Allocate the performance evaluation IDs to be reviewed of different departments to the corresponding expert groups, obtain relevant work data based on the performance evaluation IDs to be reviewed, and forward the relevant work data to the corresponding expert groups; A3: Conduct an independent review on the performance evaluation ID to be reviewed based on the expert group, generate a human resources score, a supervisor score and a data analysis score, calculate the difference between the human resources score, the supervisor score and the data analysis score, and screen the performance evaluation ID to be reviewed based on the difference calculation; A4: Calculate the average value of the human resources score, the supervisor score and the data analysis score for the remaining performance evaluation IDs to be reviewed, and generate a review score based on the average value calculation; A5: Remove the pending review mark, and mark the performance evaluation ID to be reviewed as good, medium or unqualified based on the review score, and re-import the performance evaluation ID after the marking is completed into step S106.

[0006] As a further solution of the present invention, calculating the difference between the human resources score, the supervisor score and the data analysis score, and screening the performance evaluation ID to be reviewed based on the difference calculation includes: Obtain three different differences. The calculation formulas for the three differences are: ;

[0007] Wherein, , and respectively represent three different groups of differences. If the absolute value of any one of the three groups of differences , and is too large, it is regarded as an abnormal review situation. Screen the performance evaluation ID, and after screening, the expert group opens a review meeting for the performance evaluation ID to conduct a secondary performance review and verification on the performance evaluation ID, so as to generate a unified review score. Import the performance evaluation ID after the review score is generated into step A5.

[0008] As a further solution of the present invention, a secondary performance review and audit is carried out on the performance evaluation ID, so as to generate a unified review score, including: If the expert panel cannot generate a unified review score during the review meeting, the performance evaluation of the performance evaluation ID is temporarily stopped. Subsequently, an observation period is set for the performance evaluation ID, and an observation period mark is made for the performance evaluation ID. After the end of the observation period, the relevant work data of the performance evaluation ID is updated, and the performance evaluation ID is re-imported into step S104.

[0009] As a further solution of the present invention, a difficulty-contribution degree correlation network is established, including: Model the difficulty of different work tasks and model the contribution degree of different work. After the modeling is completed, network nodes are set. The network nodes include task nodes and contribution degree nodes. The task nodes represent the work difficulty of different work tasks, and the contribution degree nodes represent the incidental contribution degree of different work tasks. Connections are established based on the task nodes and the contribution degree nodes, so as to generate a task-contribution degree connection. Based on the task-contribution degree connection and the network nodes, a task-contribution degree bipartite graph structure is established. The task nodes and the contribution degree nodes in the task-contribution degree bipartite graph structure are interconnected. Based on the network nodes, the task-contribution degree connection and the task-contribution degree bipartite graph structure, a difficulty-contribution degree correlation network is established. After the establishment is completed, features in the task-contribution degree bipartite graph structure are extracted based on the graph convolutional neural network, and the difficulty-contribution degree correlation network is optimized. Based on the difficulty-contribution degree correlation network, a network evaluation value is generated.

[0010] As a further solution of the present invention, the network evaluation value can be expressed as: ;

[0011] Wherein, represents the network evaluation value, represents the total number of employee task submission frequencies, represents the employee task completion rate data, represents the th submission frequency data of the employee task submitted by the employee, represents the th work difficulty of the work task submitted by the employee, represents the th incidental contribution degree of the work task submitted by the employee.

[0012] As a further solution of the present invention, the evaluation threshold is periodically changed according to the change of the market environment, including: Obtain market factors and enterprise internal factors. The market factors include industry economic situation data and macroeconomic indicator data, and the market factors are used to evaluate the impact of the external economy on the enterprise. The enterprise internal factors include enterprise profit rate data and enterprise revenue data, and the enterprise internal factors are used to evaluate the overall situation of the enterprise operation; Construct a threshold change formula based on the external market factors and enterprise internal factors. The threshold change formula can be expressed as: ; Among them, represents the evaluation threshold, represents the average value of the historical performance of enterprise employees, represents the market impact factor, represents the enterprise operation impact factor, and represents the weight; Regularly obtain the evaluation threshold that changes with the market environment based on the threshold change formula.

[0013] As a further solution of the present invention, obtain HR assessment record data, employee task log data and employee real-time behavior data based on the employee assessment database, including: The HR assessment record data includes employee attendance data, employee periodic assessment result data and employee periodic training result data; The employee task log data includes employee task completion rate data and employee task submission frequency data; The employee real-time behavior data includes meeting participation frequency data and working hours data.

[0014] As a further solution of the present invention, perform data cleaning operations on the relevant data of the performance evaluation ID, and record the performance evaluation ID with abnormal data in the data cleaning operation, including: After the data cleaning operation is completed, establish a defect list. The defect list is used to record the performance evaluation ID with abnormal data, and the defect list is sent to the operator in the form of a reminder pop-up window in real time to remind the operator to supplement the abnormal data; If the data supplement is completed, perform a data cleaning operation on the performance evaluation ID for which the data supplement is completed, and re-import the performance evaluation ID without abnormal data in the completed data supplement into S103.

[0015] A digital performance evaluation system based on artificial intelligence, including: A data module, which includes a data acquisition module and a data processing module; The data acquisition module is used to establish an API interface to obtain the relevant data required; A data processing module, which is used to perform data cleaning operations on the relevant data of the performance evaluation ID, perform cross-modal data alignment operations on the performance evaluation IDs without abnormal data, and move the performance evaluation IDs with abnormal data into the defect list; A threshold module, which is used to obtain the evaluation threshold and adjust the evaluation threshold based on the threshold change formula, so that the evaluation threshold can be changed regularly with the change of the market environment; An authentication module, which is used to establish the performance evaluation ID, match and connect the relevant data with the performance evaluation ID of each employee, and moreover, can perform relevant marking operations on the performance evaluation ID during the performance evaluation operation; An evaluation module, which is used to establish a performance evaluation data graph and a difficulty-contribution degree correlation network, obtain the network evaluation value and the graph evaluation value based on the performance evaluation data graph and the difficulty-contribution degree correlation network, perform a performance level matching operation on the network evaluation value and the graph evaluation value based on the evaluation threshold. After the performance level matching operation is completed, perform a follow-up work improvement evaluation operation on the performance evaluation ID based on the graph neural network model and the time series prediction model, and generate an employee performance report that matches the performance evaluation ID. The employee performance report includes the performance level and suggestions for follow-up work improvement, and forward the employee performance report to the employee and the enterprise management layer that match the performance evaluation ID; An execution module, which regularly executes S101-S106 in this method to perform performance evaluation operations on enterprise employees; A review module, which is used to execute the expert review mechanism.

[0016] Compared with the prior art, the present invention has the following beneficial effects: First, obtain the employee evaluation database, establish a connection with the employee evaluation database through the API interface, and obtain the HR assessment record data, employee task log data, and employee real-time behavior data based on the employee evaluation database. After the acquisition is completed, establish a performance evaluation ID, match the HR assessment record data, employee task log data, and employee real-time behavior data with the performance evaluation ID and establish a connection, perform data cleaning operations, and record the performance evaluation IDs with abnormal data to prevent data errors during the evaluation process and improve the accuracy of subsequent performance evaluation operations. Then, based on the cross-modal data alignment operation, establish a unified performance evaluation data map for each performance evaluation ID that matches, and at the same time establish a difficulty-contribution value association network. After the establishment is completed, perform map performance evaluation operations and secondary performance evaluation operations to generate map evaluation values and network evaluation values, obtain the evaluation threshold, perform performance level matching operations based on the evaluation threshold, and introduce an expert review mechanism during the performance level matching operation. After the performance level matching operation is completed, perform subsequent work improvement evaluation operations based on the graph neural network model and the time series prediction model to generate an employee performance report that matches the performance evaluation ID. The employee performance report includes the performance level and suggestions for subsequent work improvement. Finally, forward the employee performance report to the employee and the enterprise management level that match the performance evaluation ID. When performing performance evaluation, this method can automatically read the relevant data required for employee performance evaluation through the introduction of models and association networks, and perform employee performance evaluation, improving the objectivity of employee performance evaluation, preventing the inability to unify the standards and scales of different evaluators in manual evaluation, and when performing employee performance evaluation, introducing an expert review mechanism to conduct a secondary evaluation of the performance of employees with medium and low-level performance by the expert review panel, improving fairness and being more user-friendly. At the same time, this method can also generate an employee performance report based on the relevant work data of the employee, and the report includes the performance level and suggestions for subsequent work improvement. Brief Description of the Drawings

[0017] Figure 1 is a flowchart of the steps of a digital performance evaluation method based on artificial intelligence according to the present invention; Figure 2 is a flowchart of the steps of the performance level matching operation in step S105 of a digital performance evaluation method based on artificial intelligence according to the present invention; Figure 3 is a flowchart of the steps of the expert review mechanism in a digital performance evaluation method based on artificial intelligence according to the present invention. Detailed Embodiments

[0018] The following further describes in detail the embodiments of the present invention in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the technical solutions of the present invention, but cannot be used to limit the protection scope of the present invention.

[0019] Example 1: As shown in the appended Figure 1 , Figure 2 , Figure 3 figures: The present invention provides a digital performance evaluation method based on artificial intelligence, applicable to performance evaluation, including the following steps: S101: Obtain an employee evaluation database, which is used to record and store relevant data of employees' daily work. Establish an API interface, establish a link with the employee evaluation database through the API interface, so as to call relevant data in the employee evaluation database, and obtain HR assessment record data, employee task log data, and employee real-time behavior data based on the employee evaluation database.

[0020] Among them, the HR assessment record data is used to evaluate the daily work status of employees, the employee task log data is used to evaluate the work results of employees, and the employee real-time behavior data is used to evaluate the daily work activity of employees.

[0021] Furthermore, the HR assessment record data includes employee attendance data, employee periodic assessment result data, and employee periodic training result data; through the selection of HR assessment record data, the attendance data and assessment scores are based on objective records, reducing human subjective intervention. Periodic assessments can reflect employees' long-term work performance, rather than just short-term fluctuations. Training data can measure whether employees are actively improving their own abilities and whether they are adapting to the development of the position.

[0022] The employee task log data includes employee task completion rate data and employee task submission frequency data; through the selection of employee task log data, the task completion rate is the most intuitive KPI, which can quantify employees' contributions. The task submission frequency can reflect employees' work rhythm and whether there are procrastination problems, and the task log is automatically recorded by the system, avoiding human tampering or misjudgment.

[0023] The employee real-time behavior data includes meeting participation frequency data and working hours data; through the selection of employee real-time behavior data, the frequency of participating in meetings can reflect whether employees are willing to communicate and collaborate, and the working hours data can reveal whether employees are working efficiently or working inefficiently overtime. If the task completion rate is high and the working hours are moderate, it indicates that employees have high work efficiency; otherwise, there may be problems of inefficient work.

[0024] If a single type of data is used alone, it may lead to one-sided conclusions. Therefore, combining the three types of data can more accurately evaluate employees' real work status. This data-driven performance evaluation method can help enterprises optimize the performance evaluation system and improve the accuracy and fairness of employee management.

[0025] S102: Establish a performance evaluation ID. The performance evaluation ID represents the unique authentication ID for each employee and the data matching that employee. Match and establish a connection between the HR assessment record data, the employee task log data, and the employee real-time behavior data with the performance evaluation ID. Perform data cleaning operations on the relevant data of the performance evaluation ID, and record the performance evaluation IDs with abnormal data during the data cleaning operation.

[0026] It can be understood that the performance evaluation ID is the unique identity identifier of each employee in the performance evaluation system, mainly used to match and integrate data from different sources (task logs, behavior data, etc.). Through the setting of the uniquely authenticated performance evaluation ID, all performance-related data is bound to the ID, avoiding data loss or duplicate calculations. It can also avoid using sensitive information such as real names for data processing, reducing the risk of privacy leakage. When conducting cross-departmental analysis, it can ensure data consistency.

[0027] Specifically, data cleaning operations can be used to remove incorrect, redundant, or abnormal data to ensure the accuracy of performance evaluation. When selecting the specific algorithms for data cleaning operations, algorithms such as anomaly detection algorithms can be chosen. And if the user needs to prevent data tampering, encryption storage and permission control related to anti-tampering can be set to ensure objectivity and fairness when the method automatically evaluates employee performance.

[0028] Furthermore, after the data cleaning operation is completed, establish a defect list. The defect list is used to record the performance evaluation IDs with abnormal data and send the defect list to the operator in the form of a reminder pop-up window in real time, reminding the operator to supplement the abnormal data. After receiving the reminder pop-up window, the operator can promptly remind the employees matching the performance evaluation ID included in the defect list to supplement the missing data. The data cleaning operation can prevent data errors during the evaluation process and improve the accuracy of subsequent performance evaluation operations. If the data supplementation is completed, perform the data cleaning operation on the performance evaluation ID for which the data supplementation is completed again. Performing the data cleaning operation on the performance evaluation ID for which the data supplementation is completed again can prevent data errors again. For example, when an employee's task completion rate is greater than 100% or the working hours exceed 24 hours, the incorrect data will be cleaned to avoid sudden anomalies. If abnormal data still appears in the performance evaluation ID, repeat the above steps and re-import the performance evaluation IDs without abnormal data in the completed data supplementation into S103.

[0029] S103: Perform cross-modal data alignment operations on the performance evaluation IDs without abnormal data, establish a unified performance evaluation data map for each performance evaluation ID that matches the performance evaluation ID based on the cross-modal data alignment operations. The unified performance evaluation data map represents the comprehensive work performance of each performance evaluation ID, and establish a difficulty-contribution degree association network, which is used to comprehensively consider the workload and work task difficulty of each performance evaluation ID.

[0030] It can be understood that during the performance evaluation process, the data sources are diverse, including HR assessment record data, employee task log data, employee real-time behavior data, etc. These data usually belong to different modalities, that is, the data types, formats, and sources are different. For example, attendance data is structured tabular data, while employee behavior data may come from unstructured log records, which makes it difficult to directly fuse these data for analysis. Therefore, cross-modal data alignment is required to unify these data from different sources. When establishing a unified performance evaluation data map, a knowledge graph structure can be selected for construction. At the same time, when obtaining the graph evaluation value, the graph evaluation value is obtained based on the knowledge graph structure.

[0031] Furthermore, when establishing the difficulty-contribution degree association network: Model different work task difficulties and different work contribution degrees; After the modeling is completed, set network nodes, which include task nodes and contribution degree nodes. The task nodes represent the work difficulties of different work tasks, and the contribution degree nodes represent the incidental contribution degrees of different work tasks. Establish connections based on the task nodes and contribution degree nodes to generate task-contribution degree connections. Based on the task-contribution degree connections and network nodes, establish a task-contribution degree bipartite graph structure, in which the task nodes and contribution degree nodes in the task-contribution degree bipartite graph structure are interconnected; Based on the network nodes, task-contribution degree connections, and task-contribution degree bipartite graph structure, establish a difficulty-contribution degree association network. After the establishment is completed, extract the features in the task-contribution degree bipartite graph structure based on the graph convolutional neural network, and optimize the network of the difficulty-contribution degree association network. Generate network evaluation values based on the difficulty-contribution degree association network.

[0032] S104: Perform graph performance evaluation operations on the performance evaluation IDs without abnormal data based on the unified performance evaluation data map. After the evaluation is completed, generate graph evaluation values. Perform secondary performance evaluation operations on the performance evaluation IDs without abnormal data based on the difficulty-contribution value association network. After the evaluation is completed, generate network evaluation values, and match and connect the graph evaluation values and network evaluation values with the performance evaluation IDs; Among them, the network evaluation value can be expressed as: ; Among them, is expressed as the network evaluation value, is expressed as the total frequency of employees' task submissions, is expressed as the data of employees' task completion rate, is expressed as the th submission frequency data of employees' tasks for the th time submitted by employees, is expressed as the work difficulty of the work task submitted by employees for the th time, is expressed as the incidental contribution degree of the work task submitted by employees for the

[0033] S105: Obtain the evaluation threshold. The evaluation threshold is used to perform performance level matching on the performance evaluation ID. The evaluation threshold is regularly changed according to the market environment. Perform performance level matching operations on the atlas evaluation value and the network evaluation value with the evaluation threshold.

[0034] Specifically, the evaluation threshold is regularly changed according to the market environment, including: Obtain market factors and internal enterprise factors. Market factors include industry economic situation data and macroeconomic indicator data. Market factors are used to evaluate the impact of the external economy on the enterprise. Internal enterprise factors include enterprise profit rate data and enterprise revenue data. Internal enterprise factors are used to evaluate the overall situation of the enterprise's operation. The thresholds in traditional performance evaluation methods are often static and cannot respond to fluctuations in the external environment in a timely manner. The performance standards of the enterprise should be dynamically adjusted according to changes in the market environment.

[0035] It can be understood that market factors include industry economic situation data and macroeconomic indicator data. Industry economic situation data evaluates industry trends through factors such as industry growth rate, market demand changes, and competitor data. Macroeconomic indicator data includes factors such as GDP growth rate, CPI (Consumer Price Index), PPI (Producer Price Index), and interest rate changes. Internal enterprise factors include enterprise profit rate data and enterprise revenue data. Enterprise profit rate and enterprise revenue data can reflect the overall situation of the enterprise's operation.

[0036] Based on external market factors and internal enterprise factors, construct a threshold change formula. The threshold change formula can be expressed as: ; Among them, is expressed as the evaluation threshold, is expressed as the average value of the historical performance of enterprise employees, is expressed as the market impact factor, is expressed as the enterprise operation impact factor, and Denoted as weights, the evaluation thresholds that change with the market environment are regularly obtained based on the threshold change formula.

[0037] Furthermore, the performance level matching operation includes: S1051: If both the graph evaluation value and the network evaluation value are lower than the evaluation threshold, mark the performance evaluation ID as failed.

[0038] S1052: If both the graph evaluation value and the network evaluation value are within the proximity interval of the evaluation threshold, mark the performance evaluation IDs within the proximity interval as medium.

[0039] S1053: If a single value among the graph evaluation value and the network evaluation value is lower than the evaluation threshold, mark the performance evaluation ID as pending review. After marking, import the performance evaluation ID pending review into the expert review mechanism.

[0040] Specifically, the expert review mechanism includes: Establish an artificial expert review panel. The artificial expert review panel conducts an artificial review on the performance evaluation ID pending review. After the review, remove the pending review mark of the performance evaluation ID, and mark the performance evaluation ID as good, medium, or failed. After marking, re-import the performance evaluation ID into S106; Specifically, it includes the following steps: A1: Establish an artificial expert review panel. The artificial expert review panel includes multiple expert groups. Each of the multiple expert groups includes human resource supervisors, department supervisors, and data analysis experts from different departments; A2: Allocate the performance evaluation IDs pending review from different departments to the corresponding expert groups. Obtain relevant work data based on the performance evaluation IDs pending review, and forward the relevant work data to the corresponding expert groups; A3: Conduct an independent review on the performance evaluation ID pending review by the expert group, generate a human resource score, a supervisor score, and a data analysis score. Calculate the difference between the human resource score, the supervisor score, and the data analysis score, and screen the performance evaluation ID pending review based on the difference calculation; A4: Calculate the average of the human resource score, the supervisor score, and the data analysis score for the remaining performance evaluation IDs pending review, and generate a review score based on the average calculation; A5: Remove the pending review mark, and mark the performance evaluation ID pending review as good, medium, or failed based on the review score. Re-import the performance evaluation ID after marking into step S106.

[0041] It is understandable that although artificial intelligence can provide efficient, data-driven performance evaluations, in some complex or boundary cases, relying solely on algorithms may lead to misjudgments. Therefore, introducing an expert review mechanism can effectively improve fairness. For example, adopting an expert review mechanism can consider special circumstances and avoid a one-size-fits-all approach. In the case of fully adopting artificial intelligence for traditional performance evaluation methods, the expert review mechanism can consider the special circumstances of individuals and conduct a secondary review to enhance fairness and reduce employee dissatisfaction and turnover rates.

[0042] For example, if an employee's short-term performance decreases due to participating in a high-difficulty project but has high long-term value, one of the values given by AI may be below the evaluation threshold, while the expert review mechanism can adjust the evaluation result based on the actual situation. For newly hired or temporarily transferred employees, the algorithm may give a low value due to insufficient data, and experts can make supplementary judgments from a business perspective.

[0043] Furthermore, obtain three different differences, and the calculation formulas for the three differences are: ;

[0044] where 、 and respectively represent three different groups of differences. If 、 and the absolute value of any one of the groups of differences is too large, it is regarded as an abnormal review situation, and the performance evaluation ID is screened. After screening, the expert group starts a review meeting for this performance evaluation ID, conducts a secondary performance review and verification for this performance evaluation ID, thereby generating a unified review score, and imports the performance evaluation ID after generating the review score into step A5.

[0045] Furthermore, if the expert group cannot generate a unified review score during the review meeting, the performance evaluation of this performance evaluation ID is temporarily stopped, and then an observation period is set for this performance evaluation ID. The observation period can be set from one to two months. By setting the observation period, employees can be given the opportunity for re-evaluation, avoiding unreasonable evaluation methods due to the expert group's inability to reach a unified review result, and marking the observation period for this performance evaluation ID. After the observation period ends, update the relevant work data of this performance evaluation ID and re-import this performance evaluation ID into step S104.

[0046] S1054: If both the graph evaluation value and the network evaluation value are higher than the proximity interval of the evaluation threshold, mark this performance evaluation ID as having good performance.

[0047] S106: After the performance level matching operation is completed, the marked performance evaluation IDs will be imported into the graph neural network model and the time series prediction model for subsequent improvement evaluation operations of the work. The graph neural network model is used to analyze the employee capabilities for the performance evaluation IDs, and the time series prediction model is used to evaluate the growth trends for the performance evaluation IDs. Based on the graph neural network model and the time series prediction model, an employee performance report matching the performance evaluation IDs will be generated. The employee performance report includes the performance level and suggestions for subsequent work improvement. The employee performance report will be forwarded to the employees and the enterprise management level matching the performance evaluation IDs.

[0048] Specifically, the combination of the graph neural network model and the time series prediction model can be used to generate personalized suggestions for subsequent work improvement, rather than simply scoring, which can make employees more clearly understand how to improve themselves. For example, the suggestions for subsequent work improvement for technical talents may recommend that they participate in professional skill training to improve specific technical capabilities, while for management talents, it may be recommended that they enhance team communication and leadership skills and improve management skills through training or actual tasks. Moreover, the management level can also optimize talents based on the employee performance report. For example, certain employees can be promoted, the positions or departments of inappropriate employees can be adjusted, or employees can be optimized.

[0049] Regularly execute S101 - S106 to perform performance evaluation operations on enterprise employees.

[0050] A digital performance evaluation system based on artificial intelligence, including: A data module, which includes a data acquisition module and a data processing module; The data acquisition module is used to establish an API interface to obtain the relevant data required; The data processing module is used to perform data cleaning operations on the relevant data of the performance evaluation IDs, perform cross-modal data alignment operations on the performance evaluation IDs without abnormal data, and move the performance evaluation IDs with abnormal data into the defect list; A threshold module is used to obtain the evaluation threshold and adjust the evaluation threshold based on the threshold change formula, so that the evaluation threshold can be changed regularly with the change of the market environment; An authentication module is used to establish performance evaluation IDs, match and connect the relevant data with the performance evaluation IDs of each employee, and moreover, can perform relevant marking operations on the performance evaluation IDs during the performance evaluation operation; An evaluation module, which is used to establish a performance evaluation data graph and a difficulty - contribution degree association network, obtain a network evaluation value and a graph evaluation value based on the performance evaluation data graph and the difficulty - contribution degree association network, perform a performance level matching operation on the network evaluation value and the graph evaluation value based on an evaluation threshold. After the performance level matching operation is completed, perform a follow - up work improvement evaluation operation on the performance evaluation ID based on a graph neural network model and a time - series prediction model, and generate an employee performance report that matches the performance evaluation ID. The employee performance report includes a performance level and suggestions for subsequent work improvement, and forward the employee performance report to the employee and the enterprise management level that match the performance evaluation ID; An execution module, which regularly executes the performance evaluation operation on enterprise employees in S101 - S106 of this method; A review module, which is used to execute an expert review mechanism.

[0051] The specific usage and function of the first embodiment: First, obtain an employee evaluation database, establish a connection with the employee evaluation database through an API interface, obtain HR assessment record data, employee task log data, and employee real - time behavior data based on the employee evaluation database. After obtaining, establish a performance evaluation ID, match and establish a connection between the HR assessment record data, employee task log data, and employee real - time behavior data and the performance evaluation ID, perform a data cleaning operation, and record the performance evaluation ID with abnormal data to prevent data errors during the evaluation process and improve the accuracy of subsequent performance evaluation operations. Then, establish a unified performance evaluation data graph for each performance evaluation ID that matches based on cross - modal data alignment operations, and establish a difficulty - contribution value association network at the same time. After establishment, perform a graph performance evaluation operation and a secondary performance evaluation operation to generate a graph evaluation value and a network evaluation value, obtain an evaluation threshold, perform a performance level matching operation based on the evaluation threshold, and introduce an expert review mechanism when performing the performance level matching operation. After the performance level matching operation is completed, perform a follow - up work improvement evaluation operation based on a graph neural network and a time - series prediction model to generate an employee performance report that matches the performance evaluation ID. The employee performance report includes a performance level and suggestions for subsequent work improvement. Finally, forward the employee performance report to the employee and the enterprise management level that match the performance evaluation ID. When performing performance evaluation, this method can automatically read relevant data required for employee performance evaluation by introducing a model and an association network and perform employee performance evaluation, improving the objectivity of employee performance evaluation, preventing the inability to unify the standards and scales of different evaluators in manual evaluation. And when performing employee performance evaluation, an expert review mechanism is introduced to conduct a secondary evaluation of the performance of employees at the middle and lower levels by an expert review panel, improving fairness and being more user - friendly. At the same time, this method can also generate an employee performance report based on the relevant work data of employees, and the report includes a performance level and suggestions for subsequent work improvement.

[0052] An electronic device, comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method proposed in Embodiment 1 of the present invention.

[0053] The following is a specific introduction to the various components of the electronic device: Among them, the processor is the control center of the electronic device, which can be a single processor or a collective term for multiple processing elements. For example, the processor is one or more central processing units (CPUs), or can be an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement Embodiment 1 of the present invention, such as: one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs).

[0054] Among them, the processor can execute various functions of the electronic device by running or executing software programs stored in the memory and calling data stored in the memory.

[0055] The memory is used to store the software program for implementing the solution of the present invention and is controlled by the processor for execution. The specific implementation manner can refer to the above method embodiment and will not be elaborated here.

[0056] The memory may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or may also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only (CD-ROM), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may be integrated with the processor or exist independently and be coupled to the processor through the interface circuit of the electronic device. The embodiments of the present invention do not make specific limitations on this.

[0057] The above embodiments may be implemented in whole or in part by software, hardware (such as circuits), firmware, or any other combination. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more collections of available media. The available medium may be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state drive.

[0058] It should be understood that the term "and / or" in this text is merely a description of the relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character " / " in this text generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship. The specific meaning can be understood by referring to the context before and after.

[0059] It should be understood that in the embodiments of the present invention, the magnitudes of the sequence numbers of the above processes do not imply the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0060] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A digital performance evaluation method based on artificial intelligence, characterized in that: The following steps are included: S101: Acquire an employee evaluation database, which is used to record and store data related to employees' daily work, establish an API interface, establish a link with the employee evaluation database through the API interface, so as to call relevant data in the employee evaluation database, and obtain HR assessment record data, employee task log data and employee real-time behavior data based on the employee evaluation database; Among them, HR assessment record data is used to evaluate employees' daily work status, employee task log data is used to evaluate employees' work results, and employee real-time behavior data is used to evaluate employees' daily work activity; S102: Establishing a performance evaluation ID, which represents a unique authentication ID for each employee and the data that matches the employee, matching and connecting the HR assessment record data, employee task log data, and employee real-time behavior data with the performance evaluation ID, performing data cleaning operations on the data related to the performance evaluation ID, and recording the performance evaluation ID for which abnormal data appears during the data cleaning operation; S103: Perform a cross-modal data alignment operation on the performance evaluation IDs that do not have abnormal data, and establish a unified performance evaluation data map that matches each performance evaluation ID based on the cross-modal data alignment operation. The unified performance evaluation data map represents the comprehensive work performance of each performance evaluation ID, and establishes a difficulty-contribution association network. The difficulty-contribution association network is used to comprehensively consider the workload and difficulty of each performance evaluation ID; S104: Based on the unified performance evaluation data map, a map performance evaluation operation is performed on the performance evaluation ID without abnormal data. After the evaluation is completed, a map evaluation value is generated. Based on the difficulty-contribution value association network, a secondary performance evaluation operation is performed on the performance evaluation ID without abnormal data. After the evaluation is completed, a network evaluation value is generated, and the map evaluation value and the network evaluation value are matched and connected with the performance evaluation ID; S105: Obtaining an evaluation threshold, which is used to match the performance evaluation ID to a performance level. The evaluation threshold is changed regularly as the market environment changes, and the graph evaluation value and the network evaluation value are matched with the evaluation threshold to perform a performance level matching operation; Performance level matching operations include: S1051: If both the graph evaluation value and the network evaluation value are lower than the evaluation threshold, the performance evaluation ID is marked as unqualified; S1052: If both the graph evaluation value and the network evaluation value are in a range close to the evaluation threshold, the performance evaluation ID in the range is marked as medium performance; S1053: If a single value of the graph evaluation value and the network evaluation value is lower than the evaluation threshold, the performance evaluation ID is marked as pending for review. After the marking is completed, the performance evaluation ID to be reviewed is imported into the expert review mechanism; S1054: If the graph evaluation value and the network evaluation value are both higher than the approximate interval of the evaluation threshold, the performance evaluation ID is marked as having good performance; S106: After the performance level matching operation is completed, the marked performance evaluation ID is imported into the graph neural network model and the time series prediction model for subsequent work improvement evaluation operations. The graph neural network model is used to perform employee capability analysis operations on the performance evaluation ID, and the time series prediction model is used to evaluate the growth trend of the performance evaluation ID. An employee performance report matching the performance evaluation ID is generated based on the graph neural network model and the time series prediction model. The employee performance report includes the performance level and subsequent work improvement suggestions, and the employee performance report is forwarded to the employee matching the performance evaluation ID and the enterprise management; Regularly execute S101-S106 to conduct performance evaluation of corporate employees.

2. According to the artificial intelligence-based digital performance evaluation method of claim 1, it is characterized in that: The expert review mechanism includes: Establish a manual expert review panel, and have the manual expert review panel manually review the performance evaluation ID to be reviewed. After the review is completed, remove the pending review mark of the performance evaluation ID, and mark the performance evaluation ID as good, medium or unqualified. After the marking is completed, re-import the performance evaluation ID into S106; The specific steps include: A1: Establish a human expert review panel, which includes multiple expert groups, including human resources managers, department heads and data analysis experts from different departments; A2: Assign the performance evaluation IDs to be reviewed by different departments to the corresponding expert groups, obtain relevant work data based on the performance evaluation IDs to be reviewed, and forward the relevant work data to the corresponding expert groups; A3: The expert group conducts an independent review of the performance evaluation IDs to be reviewed, generates human resource scores, supervisor scores and data analysis scores, calculates the difference between the human resource scores, supervisor scores and data analysis scores, and screens the performance evaluation IDs to be reviewed based on the difference calculation; A4: Calculate the average of the human resources score, supervisor score and data analysis score for the performance evaluation ID that is retained for review, and generate the review score based on the average calculation; A5: Remove the pending review mark, and mark the pending performance evaluation ID as good, medium or unqualified based on the review score, and re-import the marked performance evaluation ID into step S106.

3. According to the artificial intelligence-based digital performance evaluation method of claim 2, it is characterized in that: Calculate the difference between the human resources score, supervisor score and data analysis score, and filter the performance evaluation IDs to be reviewed based on the difference calculation, including: Get three different sets of differences. The calculation formula for the three sets of differences is: ； in, , and Represent three different sets of differences respectively. If , and If the absolute value of any group of differences is too large, it is considered an abnormal review situation and the performance evaluation ID is screened. After the screening, an expert group will start a review meeting for the performance evaluation ID and conduct a second performance review on the performance evaluation ID to generate a unified review score. The performance evaluation ID after the review score is generated is imported into step A5.

4. According to claim 3, a digital performance evaluation method based on artificial intelligence is characterized in that: Conduct a second performance review on the performance evaluation ID to generate a unified review score, including: If the expert group is unable to generate a unified review score during the review meeting, the performance evaluation of the performance evaluation ID will be temporarily suspended, and an observation period will be set for the performance evaluation ID, and the performance evaluation ID will be marked for the observation period. After the observation period, the relevant work data of the performance evaluation ID will be updated, and the performance evaluation ID will be re-imported into step S104.

5. According to the artificial intelligence-based digital performance evaluation method of claim 1, it is characterized in that: Build a difficulty-contribution correlation network, including: Model different work task difficulties and different work contributions; After the modeling is completed, the network nodes are set. The network nodes include task nodes and contribution nodes. The task nodes represent the difficulty of different work tasks, and the contribution nodes represent the incidental contribution of different work tasks. Based on the task nodes and contribution nodes, connections are established to generate task-contribution connections. Based on the task-contribution connections and network nodes, a task-contribution bidirectional graph structure is established. In the task-contribution bidirectional graph structure, the task nodes and contribution nodes are connected to each other. A difficulty-contribution association network is established based on network nodes, task-contribution connections and task-contribution bidirectional graph structures. After the establishment is completed, the features in the task-contribution bidirectional graph structure are extracted based on the graph convolutional neural network, and the difficulty-contribution association network is optimized, and a network evaluation value is generated based on the difficulty-contribution association network.

6. The digital performance evaluation method based on artificial intelligence according to claim 1 is characterized in that: The network evaluation value can be expressed as: ； in, Expressed as the network evaluation value, Indicates the total number of employee task submission frequencies. Represented as employee task completion rate data, Indicates that the employee The frequency data of employee task submissions submitted by times. Indicates that the employee The difficulty of the work task submitted Indicates that the employee The incidental contribution of the work task submitted.

7. The digital performance evaluation method based on artificial intelligence according to claim 1 is characterized in that: Assessment thresholds are regularly changed as market conditions change, including: Obtain market factors and internal factors of enterprises. Market factors include industry economic situation data and macroeconomic indicator data. Market factors are used to evaluate the impact of external economy on enterprises. Internal factors of enterprises include enterprise profit margin data and enterprise revenue data. Internal factors of enterprises are used to evaluate the overall situation of enterprise operations. The threshold change formula is constructed based on external market factors and internal enterprise factors. The threshold change formula can be expressed as: ； in, Denotes the evaluation threshold, It is expressed as the average value of the historical performance of the enterprise employees. Expressed as the market impact factor, It is expressed as the business operation impact factor of the enterprise. and Expressed as weight; Based on the threshold change formula, the evaluation threshold that changes with the market environment is regularly obtained.

8. The digital performance evaluation method based on artificial intelligence according to claim 1 is characterized in that: Based on the employee evaluation database, HR assessment record data, employee task log data and employee real-time behavior data are obtained, including: HR assessment record data includes employee attendance data, employee periodic assessment result data, and employee periodic training result data; Employee task log data includes employee task completion rate data and employee task submission frequency data; Employees’ real-time behavior data includes meeting attendance frequency data and working hours data.

9. The digital performance evaluation method based on artificial intelligence according to claim 1 is characterized in that: Perform data cleaning operations on the data related to the performance evaluation ID, and record the performance evaluation IDs that have abnormal data during the data cleaning operation, including: After the data cleaning operation is completed, a defect list is created to record the performance evaluation IDs of abnormal data. The defect list is sent to the operator in real time in the form of a reminder pop-up box to remind the operator to supplement the abnormal data. If the data supplementation is completed, the data cleaning operation is performed again on the performance evaluation IDs that have completed the data supplementation, and the performance evaluation IDs that have completed the data supplementation and have no abnormal data are re-imported into S103.

10. A digital performance evaluation system based on artificial intelligence, characterized in that: include: Data module, the data module includes a data acquisition module and a data processing module; The data acquisition module is used to establish an API interface to acquire the required relevant data; The data processing module is used to clean the data related to the performance evaluation ID, perform cross-modal data alignment on the performance evaluation ID without abnormal data, and move the performance evaluation ID with abnormal data into the defect list; A threshold module is used to obtain an evaluation threshold and adjust the evaluation threshold based on a threshold change formula, so that the evaluation threshold can be changed regularly as the market environment changes; The authentication module is used to establish a performance evaluation ID and match and connect the relevant data with the performance evaluation ID of each employee, and can perform relevant marking operations on the performance evaluation ID that is undergoing performance evaluation operations; An evaluation module is used to establish a performance evaluation data map and a difficulty-contribution correlation network, and obtain network evaluation values ​​and map evaluation values ​​based on the performance evaluation data map and the difficulty-contribution correlation network, and perform a performance level matching operation on the network evaluation values ​​and the map evaluation values ​​based on the evaluation threshold. After the performance level matching operation is completed, the performance evaluation ID is evaluated for subsequent work improvements based on the graph neural network model and the time series prediction model, and an employee performance report matching the performance evaluation ID is generated. The employee performance report includes the performance level and subsequent work improvement suggestions, and the employee performance report is forwarded to the employee matching the performance evaluation ID and the enterprise management; An execution module, which regularly executes S101-S106 in the method to perform performance evaluation operations on enterprise employees; The review module is used to implement the expert review mechanism.

Images