Human resource data analysis method and system based on time sequence knowledge graph

Abstract

The invention provides a human resource data analysis method and system based on a time sequence knowledge graph, and belongs to the technical field of data process.The human resource data analysis method based on the time sequence knowledge graph integrates employee characteristics, adjacency relations and timestamp information of to-be-analyzed employees through construction of the time sequence knowledge graph, so that the time sequence knowledge graph is obtained; the method is advantaged in that employee behaviors can be more accurately analyzed, data of the time sequence knowledge graph is utilized to be inputted to the graph convolutional neural network and the Transform model, complex characteristic relations are effectively extracted and learned, and analysis accuracy and efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of technical analysis, and particularly to a method and system for human resource data analysis based on a temporal knowledge graph. Background Art

[0002] Traditional human resource analysis usually relies on static data and is difficult to reflect the performance changes of employees in actual work. At the same time, existing methods often ignore the relationships and interactions among employees, as well as the impact of time information on performance, resulting in the inability to effectively capture the dynamic characteristics of employee performance and potential productivity problems.

[0003] In the prior art, although traditional regression analysis and machine learning methods (such as random forests and support vector machines) can process some human resource data, their ability to dynamically analyze time changes and employee relationships is limited. In addition, although graph neural networks can capture complex relationships, they usually lack the ability to model time series. Summary of the Invention

[0004] The present invention provides a method and system for human resource data analysis based on a temporal knowledge graph to solve the problem of inaccurate human resource data analysis in the prior art.

[0005] The present invention provides a method for human resource data analysis based on a temporal knowledge graph, including: Constructing a human resource temporal knowledge graph, where the human resource temporal knowledge graph includes analysis entities, adjacency relationships, and timestamp information, the analysis entities include employees to be analyzed and characteristics of employees to be analyzed, the adjacency relationship is the relationship between the analysis entities, and the timestamp information is information recording the analysis entities; Preprocessing the characteristics of employees to be analyzed and the adjacency relationships in the human resource temporal knowledge graph and inputting them into a graph convolutional neural network model to obtain an embedded representation of the analysis entities; Encoding the timestamp information to obtain time numerical features; Combining the embedded representation and the time numerical features to obtain a temporal embedded representation; Inputting the temporal embedded representation into a Transformer model to obtain a score of the employee to be analyzed; Based on the score of the employee to be analyzed, obtaining a human resource analysis result.

[0006] According to the method for human resource data analysis based on a temporal knowledge graph provided by the present invention, the characteristics of employees to be analyzed include static characteristics of employees to be analyzed and dynamic characteristics of employees to be analyzed, the static characteristics of employees to be analyzed are characteristics that do not change with time of employees, and the dynamic characteristics of employees to be analyzed are characteristics that change with time of the employees.

[0007] According to a human resource data analysis method based on a temporal knowledge graph provided by the present invention, preprocessing the employee features to be analyzed and the adjacency relationship in the human resource temporal knowledge graph and inputting them into a graph convolutional neural network model to obtain an embedded representation of the analysis entity, including: Generating an adjacency matrix according to the employee features to be analyzed and the adjacency relationship; Inputting the adjacency matrix into the graph convolutional neural network model to obtain an embedded representation of the analysis entity.

[0008] According to a human resource data analysis method based on a temporal knowledge graph provided by the present invention, encoding the timestamp information to obtain time numerical features, including: Performing periodic encoding on the timestamp information using sine and cosine functions to obtain the time numerical features.

[0009] According to a human resource data analysis method based on a temporal knowledge graph provided by the present invention, the human resource analysis result includes the employee turnover prediction probability; Obtaining the human resource analysis result based on the score of the employee to be analyzed, including: Determining the employee turnover prediction probability of the employee to be analyzed according to the score of the employee to be analyzed.

[0010] The present invention also provides a human resource data analysis system based on a temporal knowledge graph, including: A first processing module for constructing a human resource temporal knowledge graph, where the human resource temporal knowledge graph includes analysis entities, adjacency relationships, and timestamp information, the analysis entities include employees to be analyzed and employee features to be analyzed, the adjacency relationship is the relationship between the analysis entities, and the timestamp information is the information recording the analysis entities; A second processing module for inputting the employee features to be analyzed and the adjacency relationship in the human resource temporal knowledge graph into a graph convolutional neural network model to obtain an embedded representation of the analysis entity; A third processing module for encoding the timestamp information to obtain time numerical features; A fourth processing module for merging the embedded representation and the time numerical features to obtain a temporal embedded representation; A fifth processing module for inputting the temporal embedded representation into a Transformer model to obtain the score of the employee to be analyzed; A sixth processing module for obtaining the human resource analysis result based on the score of the employee to be analyzed.

[0011] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the method for human resource data analysis based on a temporal knowledge graph as described in any one of the above is implemented.

[0012] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for human resource data analysis based on a temporal knowledge graph as described in any one of the above is implemented.

[0013] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, the method for human resource data analysis based on a temporal knowledge graph as described in any one of the above is implemented.

[0014] The method and system for human resource data analysis based on a temporal knowledge graph provided by the present invention integrate the employee characteristics, adjacency relationships, and timestamp information of the employee to be analyzed by constructing a temporal knowledge graph, analyze the employee's behavior more accurately, and then input the data of the temporal knowledge graph into a graph convolutional neural network and a Transformer model to effectively extract and learn complex feature relationships, which can improve the accuracy and efficiency of the analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 is a schematic flowchart of the method for human resource data analysis based on a temporal knowledge graph provided by the present invention; Figure 2 is a schematic structural diagram of the system for human resource data analysis based on a temporal knowledge graph provided by the present invention; Figure 3 is a schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0018] The following will combine with Figures 1-3 to describe the human resource data analysis method and system based on the temporal knowledge graph of the present invention.

[0019] The human resource data analysis method based on the temporal knowledge graph in the embodiments of the present invention mainly includes step 110, step 120, step 130, step 140, step 150 and step 160.

[0020] Step 110: Construct a human resource temporal knowledge graph, where the human resource temporal knowledge graph includes analysis entities, adjacency relationships and timestamp information. The analysis entities include employees to be analyzed and characteristics of employees to be analyzed. The adjacency relationship is the relationship between the analysis entities, and the timestamp information is the information recording the analysis entities.

[0021] It can be understood that a temporal knowledge graph (TKG) is a concept that extends the traditional knowledge graph, introducing a time dimension on the basis of the existing knowledge graph that only contains entities and the relationships between entities, and recording the dynamic changes of entities and related relationships over time.

[0022] The human resource temporal knowledge graph of the present invention includes analysis entities, adjacency relationships and timestamp information. The analysis entities include employees to be analyzed and characteristics of employees to be analyzed. The adjacency relationship is the relationship between the analysis entities, and the timestamp information is the information recording the analysis entities. The analysis entities can also be positions, departments and performances.

[0023] In some embodiments, the characteristics of the employees to be analyzed include static characteristics of the employees to be analyzed and dynamic characteristics of the employees to be analyzed. The static characteristics of the employees to be analyzed are the characteristics that do not change with time, and the dynamic characteristics of the employees to be analyzed are the characteristics that change with time.

[0024] Specifically, the employees to be analyzed can be the employees in the employee list obtained through the company's human resource system. The static characteristics of the employees to be analyzed can be date of birth, gender, education level, and these characteristics do not change with time. The dynamic characteristics of the employees to be analyzed can be performance scores, current positions, salary levels, skill levels, and these characteristics change with time.

[0025] The adjacency relationship can be an employment relationship: the relationship between an employee and a position, recording the employment history and current position of the employee. Department relationship: the relationship between an employee and a department, indicating the team or department where the employee is located. Performance relationship: the relationship between an employee and his / her performance record, reflecting the performance of the employee in different periods. Colleague relationship: the cooperation or interaction relationship between employees, such as project cooperation, direct reporting relationship, etc.

[0026] The timestamp information is used to record the information of the analysis entity, that is, the time relationship between the employee to be analyzed and other analysis entities (position, department, and performance). For example, the timestamp information in the employment relationship: records the time when the employee starts and ends a certain position, which can help analyze the length of service of the employee in a certain position. The timestamp information in the department relationship: records the time when the employee joins or leaves a certain department. The timestamp information in the performance relationship: records the time point of performance evaluation to observe the performance change trend. The timestamp information in the colleague relationship: records the start and end time of cooperation between employees.

[0027] Step 120: Preprocess the characteristics of the employee to be analyzed and the adjacency relationship in the human resource time-series knowledge graph and input them into the graph convolutional neural network model to obtain the embedded representation of the analysis entity.

[0028] It can be understood that after constructing the human resource time-series knowledge graph, preprocessing will be carried out on the human resource time-series knowledge graph, and then the processed data will be input into the graph convolutional neural network model to obtain the embedded representation of the analysis entity.

[0029] Specifically, the static characteristics of the employee to be analyzed can be encoded into static quantitative characteristics, and the dynamic characteristics of the employee to be analyzed can be combined with timestamp information for processing to obtain multiple dynamic quantitative characteristics extracted and integrated from the dynamic characteristics of the employee to be analyzed. The static quantitative characteristics and dynamic quantitative characteristics are merged to obtain the characteristics of the employee to be analyzed, and the characteristics are represented as a feature vector x i , where x i ∈Rdx i ∈R d , d is the feature dimension. Construct an adjacency matrix A to represent the adjacency relationship between analysis entities, and the matrix element a ij represents whether there is an adjacency relationship between node i and node j .

[0030] Next, perform the first layer of graph convolution to update the dynamic characteristics of the employee to be analyzed through the following formula: ; In the above formula, is the adjacency matrix plus a self-loop, is the degree matrix after adding the self-loop, and W (0) is the learnable weight matrix of the first layer, σ is a non-linear activation function (such as ReLU). Subsequently, in the second layer of graph convolution, use the output of the first layer H (1), further update the node feature representation to generate the final embedded representation: ; In the above formula, W (1) is the learnable weight matrix of the second layer.

[0031] Step 130, encode the timestamp information to obtain time numerical features.

[0032] It can be understood that in the step of encoding the timestamp information in the temporal knowledge graph, periodic encoding (such as sine and cosine functions) or other suitable time encoding methods can be used to convert the timestamp information into numerical features. Assume the time feature t is periodically encoded as: t = [sin_hour, cos_hour, sin_day, cos_day, sin_week, cos_week, sin_month, cos_month] t = [sin_hour, cos_hour, sin_day, cos_day, sin_week, cos_week, sin_month, cos_month]; Among them, assume that the dimension of each periodic feature is 2 (sine and cosine), and we extract hours, days, weeks, and months. Therefore t has a dimension of 8.

[0033] Step 140, merge the embedded representation and the time numerical features to obtain a temporal embedded representation.

[0034] It can be understood that after obtaining the embedded representation, the embedded representation and the time numerical features will be merged by vector concatenation to obtain a temporal embedded representation including the time dimension. In deep learning and feature engineering, vector concatenation is a common operation used to combine features from different sources into a comprehensive feature vector.

[0035] For example, concatenate the embedded representation H (2) and the time feature t to form a temporal embedded representation H . The formula is as follows: ; In the above formula, ⊕ represents the concatenation operation.

[0036] The specific concatenation form is as follows: H =h 1, h 2, h 3, h 4, sin_hour, cos_hour, sin_day, cos_day, sin_week, cos_week, sin_month, cos_month]; Among them, the embedded representation H (2) has a dimension of 4, and the time feature t has a dimension of 8. Therefore, the merged time-series embedded representation H has a dimension of: ; Step 150, input the time-series embedded representation into the Transformer model to obtain the employee score to be analyzed.

[0037] It can be understood that after obtaining the time-series embedded representation, the format of the time-series embedded representation will be converted first to obtain the input data in the tensor format, and the data will be normalized to obtain the standard input data, and then the standard input data will be input into the trained Transformer model to obtain the employee score to be analyzed.

[0038] Step 160, obtain the human resources analysis result based on the employee score to be analyzed.

[0039] It can be understood that an employee score to be analyzed (i.e., the probability of employee turnover) can be obtained from the output of the Transformer model, and this value can be converted into a binary classification result according to the threshold set during model training (assumed to be 0.5).

[0040] Adjust the threshold according to business requirements. For example, if you want to be more sensitive to the turnover risk, you can set the threshold lower so that high-risk employees can be more easily identified. Use data visualization tools, such as Matplotlib, to display the relationship between key features and prediction results to help human resources managers understand the potential factors for employee turnover. Then integrate the prediction results to generate a system report, including employee scores, turnover risk assessment, recommended intervention measures, etc.

[0041] The human resources data analysis method based on the time-series knowledge graph of the present invention integrates the employee characteristics, adjacency relationships, and timestamp information of the employee to be analyzed by constructing a time-series knowledge graph, analyzes the behavior of employees more accurately, and then uses the data of the time-series knowledge graph to input into the graph convolutional neural network and the Transformer model, effectively extracts and learns complex feature relationships, and can improve the accuracy and efficiency of analysis.

[0042] The human resource data analysis system based on the temporal knowledge graph provided by the present invention will be described below. The human resource data analysis system based on the temporal knowledge graph described below can be correspondingly referred to the human resource data analysis method based on the temporal knowledge graph described above.

[0043] As Figure 2 shown, the human resource data analysis system based on the temporal knowledge graph in the embodiment of the present invention mainly includes a first processing module 210, a second processing module 220, a third processing module 230, a fourth processing module 240, a fifth processing module 250, and a sixth processing module 260.

[0044] The first processing module 210 is used to construct a human resource temporal knowledge graph. The human resource temporal knowledge graph includes analysis entities, adjacency relationships, and timestamp information. The analysis entities include employees to be analyzed and characteristics of employees to be analyzed. The adjacency relationship is the relationship between the analysis entities, and the timestamp information is the information recording the analysis entities. The second processing module 220 is used to input the characteristics of the employees to be analyzed and the adjacency relationships in the human resource temporal knowledge graph into a graph convolutional neural network model to obtain an embedded representation of the analysis entities. The third processing module 230 is used to encode the timestamp information to obtain time numerical features. The fourth processing module 240 is used to merge the embedded representation and the time numerical features to obtain a temporal embedded representation. The fifth processing module 250 is used to input the temporal embedded representation into a Transformer model to obtain a score of the employee to be analyzed. The sixth processing module 260 is used to obtain a human resource analysis result based on the score of the employee to be analyzed.

[0045] In some embodiments, the characteristics of the employees to be analyzed include static characteristics of the employees to be analyzed and dynamic characteristics of the employees to be analyzed. The static characteristics of the employees to be analyzed are the characteristics that do not change with time, and the dynamic characteristics of the employees to be analyzed are the characteristics that change with time.

[0046] In some embodiments, the second processing module 220 is further used to generate an adjacency matrix according to the characteristics of the employees to be analyzed and the adjacency relationships. Input the adjacency matrix into the graph convolutional neural network model to obtain an embedded representation of the analysis entities.

[0047] In some embodiments, the third processing module 230 is further used to perform periodic encoding on the timestamp information using sine and cosine functions to obtain the time numerical features.

[0048] In some embodiments, the sixth processing module 260 is further configured to determine the employee turnover prediction probability of the employee to be analyzed according to the score of the employee to be analyzed.

[0049] Figure 3 An example of the physical structure diagram of an electronic device is shown as Figure 3 shown. The electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340. Among them, the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 can call the logical instructions in the memory 330 to execute the human resource data analysis method based on the temporal knowledge graph provided by the above methods. The method includes: constructing a human resource temporal knowledge graph, where the human resource temporal knowledge graph includes analysis entities, adjacency relationships, and timestamp information. The analysis entities include employees to be analyzed and characteristics of employees to be analyzed. The adjacency relationship is the relationship between the analysis entities, and the timestamp information is the information recording the analysis entities; preprocessing the characteristics of the employees to be analyzed and the adjacency relationships in the human resource temporal knowledge graph and inputting them into a graph convolutional neural network model to obtain the embedded representation of the analysis entities; encoding the timestamp information to obtain time numerical features; merging the embedded representation and the time numerical features to obtain a temporal embedded representation; inputting the temporal embedded representation into a Transformer model to obtain the score of the employee to be analyzed; and obtaining a human resource analysis result based on the score of the employee to be analyzed.

[0050] In addition, when the logical instructions in the above-mentioned memory 330 are implemented in the form of software function units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0051] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the human resource data analysis method based on a temporal knowledge graph provided by the above-mentioned various methods. The method includes: constructing a human resource temporal knowledge graph, where the human resource temporal knowledge graph includes analysis entities, adjacency relationships, and timestamp information. The analysis entities include employees to be analyzed and characteristics of employees to be analyzed. The adjacency relationship is the relationship between the analysis entities, and the timestamp information is the information recording the analysis entities; preprocessing the characteristics of the employees to be analyzed and the adjacency relationships in the human resource temporal knowledge graph and inputting them into a graph convolutional neural network model to obtain an embedded representation of the analysis entities; encoding the timestamp information to obtain time numerical features; merging the embedded representation and the time numerical features to obtain a temporal embedded representation; inputting the temporal embedded representation into a Transformer model to obtain a score of the employee to be analyzed; and obtaining a human resource analysis result based on the score of the employee to be analyzed.

[0052] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the human resource data analysis method based on a temporal knowledge graph provided by the above-mentioned various methods. The method includes: constructing a human resource temporal knowledge graph, where the human resource temporal knowledge graph includes analysis entities, adjacency relationships, and timestamp information. The analysis entities include employees to be analyzed and characteristics of employees to be analyzed. The adjacency relationship is the relationship between the analysis entities, and the timestamp information is the information recording the analysis entities; preprocessing the characteristics of the employees to be analyzed and the adjacency relationships in the human resource temporal knowledge graph and inputting them into a graph convolutional neural network model to obtain an embedded representation of the analysis entities; encoding the timestamp information to obtain time numerical features; merging the embedded representation and the time numerical features to obtain a temporal embedded representation; inputting the temporal embedded representation into a Transformer model to obtain a score of the employee to be analyzed; and obtaining a human resource analysis result based on the score of the employee to be analyzed.

[0053] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0054] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended 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 described in the foregoing embodiments or perform equivalent replacements for some of the technical features. However, 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 each embodiment of the present invention.

Claims

1. A method for human resource data analysis based on a temporal knowledge graph, characterized in that Comprising: Construct a human resource time-series knowledge graph, where the human resource time-series knowledge graph includes analysis entities, adjacency relationships, and timestamp information. The analysis entities include employees to be analyzed and characteristics of employees to be analyzed. The adjacency relationship is the relationship between the analysis entities, and the timestamp information is the information recording the analysis entities; Preprocess the characteristics of the employees to be analyzed and the adjacency relationships in the human resource time-series knowledge graph and input them into a graph convolutional neural network model to obtain the embedded representation of the analysis entities; Encode the timestamp information to obtain time numerical features; Merge the embedded representation and the time numerical features to obtain a time-series embedded representation; Input the time-series embedded representation into a Transformer model to obtain the score of the employee to be analyzed; Based on the score of the employee to be analyzed, obtain the human resource analysis result.

2. The human resource data analysis method based on a temporal knowledge graph according to claim 1, wherein The characteristics of the employees to be analyzed include static characteristics of the employees to be analyzed and dynamic characteristics of the employees to be analyzed. The static characteristics of the employees to be analyzed are the characteristics that do not change with time, and the dynamic characteristics of the employees to be analyzed are the characteristics that change with time.

3. The method for human resource data analysis based on a temporal knowledge graph according to claim 1, wherein The preprocessing the characteristics of the employees to be analyzed and the adjacency relationships in the human resource time-series knowledge graph and inputting them into a graph convolutional neural network model to obtain the embedded representation of the analysis entities includes: Generate an adjacency matrix according to the characteristics of the employees to be analyzed and the adjacency relationships; Input the adjacency matrix into the graph convolutional neural network model to obtain the embedded representation of the analysis entities.

4. The human resource data analysis method based on the temporal knowledge graph according to claim 1, wherein, The encoding the timestamp information to obtain time numerical features includes: Perform periodic encoding on the timestamp information using sine and cosine functions to obtain the time numerical features.

5. The human resource data analysis method based on a temporal knowledge graph according to claim 1, wherein The human resource analysis result includes the employee turnover prediction probability; The obtaining the human resource analysis result based on the score of the employee to be analyzed includes: Determine the employee turnover prediction probability of the employee to be analyzed according to the score of the employee to be analyzed.

6. A human resource data analysis system based on a temporal knowledge graph, characterized in that, Comprising: A first processing module for constructing a human resource time-series knowledge graph, where the human resource time-series knowledge graph includes analysis entities, adjacency relationships, and timestamp information. The analysis entities include employees to be analyzed and characteristics of employees to be analyzed. The adjacency relationship is the relationship between the analysis entities, and the timestamp information is the information recording the analysis entities; A second processing module for inputting the characteristics of the employees to be analyzed and the adjacency relationships in the human resource time-series knowledge graph into a graph convolutional neural network model to obtain the embedded representation of the analysis entities; A third processing module for encoding the timestamp information to obtain time numerical features; A fourth processing module for merging the embedded representation and the time numerical features to obtain a time-series embedded representation; A fifth processing module for inputting the time-series embedded representation into a Transformer model to obtain the score of the employee to be analyzed; A sixth processing module for obtaining the human resource analysis result based on the score of the employee to be analyzed.

7. The human resource data analysis system based on the temporal knowledge graph according to claim 6, characterized in that, The second processing module is further configured to generate an adjacency matrix according to the employee characteristics to be analyzed and the adjacency relationship; input the adjacency matrix into the graph convolutional neural network model to obtain an embedded representation of the analysis entity.

8. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the human resource data analysis method based on the temporal knowledge graph according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the human resource data analysis method based on the temporal knowledge graph according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the human resource data analysis method based on the temporal knowledge graph according to any one of claims 1 to 6.

Images