Intelligent management system and method for personnel files
Through deep learning technology, the personnel information of employees is subject to semantic understanding and context analysis, and the access rights of the personnel archives are intelligently allocated, which solves the shortcomings of the existing system in compliance with the use of compliant and access rights management, and realizes differentiated access to personnel archives and Compliant use.
Patent Information
- Application Number
- CN202411082209.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-08
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-08
AI Technical Summary
The existing information-based personnel file management system has shortcomings in compliance use and refined management of access rights, especially in large enterprises with diverse employee roles and responsibilities. How to ensure the compliant use of personnel information and avoid information leakage and abuse have become an urgent problem.
By extracting employee personnel files from employee personnel archives, extracting employee names, departments, positions and responsibilities information, and using deep learning technology for semantic understanding and context analysis, we can deeply understand employees' authority needs, and thus intelligently allocate them to their Access rights to the personnel archive.
It realizes differentiated access to personnel files, ensures compliant use of personnel information, avoids information leakage and abuse, and meets the efficient, accurate and compliant personnel information management needs of modern enterprises.
Smart Images

Figure CN118734339B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent management technology, and more specifically, to an intelligent management system and method for personnel files. Background Art
[0002] With the expansion of enterprise scale and the complexity of management, personnel file management has become an indispensable part of enterprise management. Traditional personnel file management methods mainly rely on manual operations, such as archiving and retrieving paper files, which are not only inefficient but also prone to errors, and it is difficult to meet the needs of modern enterprises for efficient, accurate and compliant personnel information management.
[0003] In the wave of digital and intelligent transformation, enterprises have gradually begun to use information technology to optimize the personnel file management process. However, the existing information-based personnel file management systems mostly focus on the digital storage and basic query functions of files, and there are still deficiencies in the compliant use of personnel information and the refined management of access rights. Especially in large enterprises with diverse employee roles and responsibilities, how to ensure the compliant use of personnel information and avoid information leakage and abuse has become an urgent problem to be solved. Therefore, an optimized personnel file intelligent management system and method are expected. Summary of the invention
[0004] In order to solve the above technical problems, this application is proposed. The embodiment of the present application provides a personnel file intelligent management system and method, which extracts the personnel files of employees to be processed from the employee personnel file library, and extracts the employee's name, department, position and responsibility information from the personnel file, uses deep learning technology to perform semantic understanding and context analysis, and deeply understands the employee's authority requirements, so as to intelligently assign access rights to the personnel file library. In this way, differentiated access to personnel files can be achieved according to the employee's position and work needs, ensuring the compliance of personnel information.
[0005] Accordingly, according to one aspect of the present application, there is provided a personnel file intelligent management system, which includes:
[0006] An employee personnel information extraction module is used to extract the personnel files of the employees to be processed from the employee personnel file library, and extract the employee personnel information from the personnel files of the employees to be processed, wherein the employee personnel information includes the name of the employee, department to which he belongs, position and responsibilities;
[0007] A personnel information semantic understanding module, used to semantically encode each information item in the employee personnel information to obtain a sequence of semantic encoding feature vectors of the employee personnel information items;
[0008] A personnel information context semantic association analysis module, used to perform multi-scale semantic association analysis on the sequence of semantic encoding feature vectors of employee personnel information items to obtain a multi-scale representation feature vector of employee personnel information semantic association fusion;
[0009] The access permission management module is used to determine the access permission label of the employee to be processed based on the semantic association of the employee's personnel information and the fusion of the multi-scale representation feature vector.
[0010] In the above-mentioned personnel file intelligent management system, the personnel information semantic understanding module is used to: semantically encode each information item in the employee personnel information through an information item semantic encoder based on the Bert model to obtain a sequence of semantic encoding feature vectors of the employee personnel information items.
[0011] In the above-mentioned personnel file intelligent management system, the personnel information context semantic association analysis module includes: a first-scale semantic association fusion unit, which is used to perform global context association encoding on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a first-scale employee personnel information context semantic association fusion feature vector; a second-scale semantic association fusion unit, which is used to perform weighted fusion on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a second-scale employee personnel information context semantic association fusion feature vector; a multi-scale fusion unit, which is used to fuse the first-scale employee personnel information context semantic association fusion feature vector and the second-scale employee personnel information context semantic association fusion feature vector to obtain the employee personnel information semantic association fusion multi-scale representation feature vector.
[0012] In the above-mentioned personnel file intelligent management system, the first-scale semantic association fusion unit is used to: input the sequence of semantic encoding feature vectors of the employee personnel information items into the employee personnel information context semantic association encoder based on the BiLSTM model to obtain the first-scale employee personnel information context semantic association fusion feature vector.
[0013] In the above-mentioned personnel file intelligent management system, the access permission management module includes: a feature optimization unit, which is used to optimize the feature distribution of the employee personnel information semantic association fusion multi-scale representation feature vector to obtain an optimized employee personnel information semantic association fusion multi-scale representation feature vector; an access permission generation unit, which is used to pass the optimized employee personnel information semantic association fusion multi-scale representation feature vector through a classifier-based access permission manager to obtain a management result, and the management result is used to represent the access permission label of the employee to be processed.
[0014] In the above-mentioned personnel file intelligent management system, the feature optimization unit is used to: perform inverse error analysis based on backward scaling optimization on the employee personnel information semantic association fusion multi-scale representation feature vector to obtain the optimized employee personnel information semantic association fusion multi-scale representation feature vector.
[0015] In the above-mentioned intelligent management system for personnel files, the feature optimization unit includes: a probabilistic activation subunit, which is used to perform probabilistic activation on the employee personnel information semantic association fusion multi-scale representation feature vector based on the Sigmoid function to obtain a probabilistic employee personnel information semantic association fusion multi-scale representation feature vector; a pre-classification subunit, which is used to input the employee personnel information semantic association fusion multi-scale representation feature vector into the classifier to obtain an initial class label probability value; an error measurement subunit, which is used to calculate the absolute value of the difference between the feature value of each position in the probabilistic employee personnel information semantic association fusion multi-scale representation feature vector and the initial class label probability value to obtain a backward machine learning fine-grained error measurement vector; an attention optimization subunit, which is used to use the backward machine learning fine-grained error measurement vector as The key vector and the employee personnel information semantic association fusion multi-scale representation feature vector are used as the query vector and the value vector, and the key vector, the query vector and the value vector are input into the converter structure to obtain the backward learning type attention optimization employee personnel information semantic association fusion multi-scale representation feature vector; a scaling factor calculation subunit is used to calculate the ratio between the sum of the backward machine learning fine-grained error measurement vector and the length of the backward machine learning fine-grained error measurement vector, and calculate the natural exponential function value with the ratio as the power as the optimization scaling factor; a scale scaling optimization subunit is used to calculate the product between the optimization scaling factor and the backward learning type attention optimization employee personnel information semantic association fusion multi-scale representation feature vector to obtain the optimized employee personnel information semantic association fusion multi-scale representation feature vector.
[0016] In the above-mentioned intelligent management system for personnel files, the access permission generation unit includes: a fully connected encoding subunit, which is used to use the fully connected layer of the access permission manager to fully connect the optimized employee personnel information semantic association fusion multi-scale representation feature vector to obtain a fully connected encoding feature vector; a probabilization subunit, which is used to pass the fully connected encoding feature vector through the Softmax classification function of the access permission manager to obtain the probability value of the optimized employee personnel information semantic association fusion multi-scale representation feature vector belonging to each access permission label; a management result determination subunit, which is used to determine the access permission label corresponding to the largest one among the probability values as the management result.
[0017] According to another aspect of the present application, there is provided a method for intelligent management of personnel files, which includes:
[0018] Extracting the personnel files of the employees to be processed from the employee personnel file library, and extracting the personnel information of the employees from the personnel files of the employees to be processed, wherein the personnel information of the employees includes the name, department, position and responsibilities of the employees;
[0019] Semantically encoding each information item in the employee personnel information to obtain a sequence of semantic encoding feature vectors of the employee personnel information items;
[0020] Performing multi-scale semantic association analysis on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a multi-scale representation feature vector of the employee personnel information semantic association fusion;
[0021] Based on the semantic association of the employee's personnel information and the fusion of the multi-scale representation feature vector, the access permission label of the employee to be processed is determined.
[0022] Compared with the prior art, the personnel file intelligent management system and method provided by the present application extracts the personnel files of the employees to be processed from the employee personnel file library, and extracts the employee's name, department, position and responsibility information from the personnel file, uses deep learning technology to perform semantic understanding and context analysis, and deeply understands the employee's authority requirements, so as to intelligently assign access rights to the personnel file library. In this way, differentiated access to personnel files can be achieved according to the employee's position and work requirements, ensuring the compliance of personnel information. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0024] Figure 1 It is a block diagram of the intelligent management system of personnel files according to an embodiment of the present application.
[0025] Figure 2 Schematic diagram of the architecture of the intelligent management system for personnel files according to an embodiment of the present application.
[0026] Figure 3 It is a block diagram of a contextual semantic association analysis module of personnel information in the intelligent management system of personnel files according to an embodiment of the present application.
[0027] Figure 4 It is a block diagram of the access rights management module in the intelligent management system of personnel files according to an embodiment of the present application.
[0028] Figure 5 A block diagram of an access rights generation unit in the intelligent management system of personnel files according to an embodiment of the present application.
[0029] Figure 6 Flow chart of the personnel file intelligent management method according to an embodiment of the present application. DETAILED DESCRIPTION
[0030] Below, the embodiments of the present application will be described in more detail in conjunction with the accompanying drawings, and the above and other purposes, features and advantages of the present application will become more apparent. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments described herein.
[0031] Figure 1 It is a block diagram of the intelligent management system of personnel files according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the structure of the intelligent management system for personnel files according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to the embodiment of the present application, the intelligent management system 100 for personnel files includes: an employee personnel information extraction module 110, which is used to extract the personnel files of the employees to be processed from the employee personnel file library, and extract the employee personnel information from the personnel files of the employees to be processed, wherein the employee personnel information includes the name, department, position and responsibilities of the employee; a personnel information semantic understanding module 120, which is used to semantically encode each information item in the employee personnel information to obtain a sequence of semantic encoding feature vectors of the employee personnel information items; a personnel information context semantic association analysis module 130, which is used to perform multi-scale semantic association analysis on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a multi-scale representation feature vector of the employee personnel information semantic association fusion; an access permission management module 140, which is used to determine the access permission label of the employee to be processed based on the multi-scale representation feature vector of the employee personnel information semantic association fusion.
[0032] As mentioned in the above background technology, the existing information-based personnel file management system mostly focuses on the digital storage and basic query functions of files, and there are still deficiencies in the compliant use of personnel information and the refined management of access rights. Especially in large enterprises with diverse employee roles and different responsibilities, how to ensure the compliant use of personnel information and avoid information leakage and abuse has become a problem that needs to be solved urgently. In response to the above technical problems, the technical concept of this application is to extract the personnel files of pending employees from the employee personnel file library, and extract the name, department, position and responsibility information of the employee from the personnel file, use deep learning technology to perform semantic understanding and context analysis, deeply understand the authority requirements of employees, and thus intelligently allocate access rights to the personnel file library for them. In this way, differentiated access to personnel files can be achieved according to the employee's position and work requirements, ensuring the compliant use of personnel information.
[0033] In the above-mentioned personnel file intelligent management system 100, the employee personnel information extraction module 110 is used to extract the personnel files of the pending employees from the employee personnel file library, and extract the employee personnel information from the personnel files of the pending employees, wherein the employee personnel information includes the name, department, position and responsibilities of the employee. It should be understood that since different employees have different responsibilities and needs in the enterprise, in order to meet their work needs and avoid information leakage and abuse, it is necessary to determine their access rights to personnel files based on the specific information of the employees. For example, a department manager can view the files of employees in his department, while the human resources department can access the files of all employees. In the technical solution of the present application, by extracting the name, department, position and responsibilities information of the employees, appropriate access rights can be allocated according to the specific circumstances of each employee, so that employees can easily obtain the information and resources required for work, and avoid unauthorized access to sensitive information.
[0034] In the above-mentioned personnel file intelligent management system 100, the personnel information semantic understanding module 120 is used to semantically encode each information item in the employee personnel information to obtain a sequence of semantic encoding feature vectors of the employee personnel information items. In a specific example of the present application, the processing method of semantically encoding each information item in the employee personnel information is to semantically encode each information item in the employee personnel information through an information item semantic encoder based on the Bert model to obtain a sequence of semantic encoding feature vectors of the employee personnel information items. It should be understood that the BERT model is a pre-trained deep bidirectional Transformer encoder that can deeply understand the semantic information of the text by training on a large amount of text data. In the technical solution of the present application, by using the BERT model to semantically encode each information item in the employee personnel information (such as the employee's name, department, position and responsibilities), the powerful semantic understanding ability of the BERT model can be used to capture the deep content meaning of each information item, and convert each information item into a numerical vector representation, providing an effective information basis for the employee's authority management.
[0035] In the above-mentioned personnel file intelligent management system 100, the personnel information context semantic association analysis module 130 is used to perform multi-scale semantic association analysis on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a multi-scale representation feature vector of the semantic association fusion of the employee personnel information. It should be understood that, considering that there is a certain semantic association dependency relationship between the various information items in the employee personnel information, for example, there is a direct association between the employee's department and his position and responsibilities. Therefore, in order to utilize the correlation between the various information items and more fully understand the employee's authority requirements, it is necessary to further perform context semantic association analysis on the sequence of semantic encoding feature vectors of the employee personnel information items. At the same time, considering that semantic association analysis methods of different scales may capture association information at different levels, thereby improving the accuracy and comprehensiveness of authority allocation. Therefore, in the technical solution of the present application, a multi-scale semantic association analysis method is used to perform information interaction fusion on the sequence of semantic encoding feature vectors of the employee personnel information items, so as to more accurately understand the information access needs of employees.
[0036] Figure 3 FIG. 1 is a block diagram of a personnel information context semantic association analysis module in a personnel file intelligent management system according to an embodiment of the present application. Figure 3As shown, the personnel information context semantic association analysis module 130 includes: a first scale semantic association fusion unit 131, used to perform global context association encoding on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a first scale employee personnel information context semantic association fusion feature vector; a second scale semantic association fusion unit 132, used to perform weighted fusion on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a second scale employee personnel information context semantic association fusion feature vector; a multi-scale fusion unit 133, used to fuse the first scale employee personnel information context semantic association fusion feature vector and the second scale employee personnel information context semantic association fusion feature vector to obtain the employee personnel information semantic association fusion multi-scale representation feature vector.
[0037] Specifically, the first-scale semantic association fusion unit 131 is used to perform global context association encoding on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain the first-scale employee personnel information context semantic association fusion feature vector. In a specific example of the present application, the processing method of performing global context association encoding on the sequence of semantic encoding feature vectors of the employee personnel information items is to input the sequence of semantic encoding feature vectors of the employee personnel information items into the employee personnel information context semantic association encoder based on the BiLSTM model to obtain the first-scale employee personnel information context semantic association fusion feature vector. It should be understood that the BiLSTM model is a bidirectional long short-term memory network that can capture context dependency information in sequence data. The LSTM (Long Short-Term Memory Network) model can effectively handle long-term dependency problems in sequence data by introducing a gating mechanism to control the flow of information. The bidirectional LSTM simultaneously considers the forward and backward context information in the sequence data, and more comprehensively captures the semantic associations between information items. In the technical solution of the present application, by using the BiLSTM model to perform global context association encoding on the sequence of semantic encoding feature vectors of the employee personnel information items, the information item association features from a global perspective can be obtained, which helps to understand the global relevance of employee information.
[0038] Specifically, the second-scale semantic association fusion unit 132 is used to perform weighted fusion on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a second-scale employee personnel information context semantic association fusion feature vector. It should be understood that since the importance of each information item in the employee personnel information to authority allocation may be different, different weights can be given to different information items according to actual needs. For example, for certain key positions or sensitive departments, their position and responsibility information may be more important, so higher weights can be given during fusion. Through this weighted fusion processing method, the importance of each information item can be considered more flexibly to highlight the influence of key information items and improve the pertinence and accuracy of authority allocation.
[0039] Specifically, the multi-scale fusion unit 133 is used to fuse the first-scale employee personnel information context semantic association fusion feature vector and the second-scale employee personnel information context semantic association fusion feature vector to obtain the employee personnel information semantic association fusion multi-scale representation feature vector. It should be understood that the first-scale employee personnel information context semantic association fusion feature vector focuses on expressing the context semantic association in employee personnel information, while the second-scale employee personnel information context semantic association fusion feature vector emphasizes the importance of different information items through weighted fusion. The two are complementary in feature representation. Therefore, in order to more comprehensively reflect the employee's personnel information and avoid the loss or deviation of information caused by a single scale feature, the first-scale employee personnel information context semantic association fusion feature vector and the second-scale employee personnel information context semantic association fusion feature vector are further fused. In an embodiment of the present application, the first-scale employee personnel information context semantic association fusion feature vector and the second-scale employee personnel information context semantic association fusion feature vector can be feature spliced and fused in a feature cascade manner to maximize the retention and integration of original information and reduce information loss.
[0040] In the above-mentioned personnel file intelligent management system 100, the access permission management module 140 is used to determine the access permission label of the employee to be processed based on the semantic association of the employee personnel information and the fusion of the multi-scale representation feature vector. Figure 4 FIG. 1 is a block diagram of an access rights management module in a personnel file intelligent management system according to an embodiment of the present application. Figure 4As shown, the access permission management module 140 includes: a feature optimization unit 141, which is used to optimize the feature distribution of the employee personnel information semantic association fusion multi-scale representation feature vector to obtain an optimized employee personnel information semantic association fusion multi-scale representation feature vector; an access permission generation unit 142, which is used to pass the optimized employee personnel information semantic association fusion multi-scale representation feature vector through a classifier-based access permission manager to obtain a management result, and the management result is used to represent the access permission label of the employee to be processed.
[0041] Specifically, the feature optimization unit 141 is used to optimize the feature distribution of the semantic association fusion multi-scale representation feature vector of the employee personnel information to obtain an optimized semantic association fusion multi-scale representation feature vector of the employee personnel information. In particular, in the technical solution of the present application, considering that when the employee's personnel information is complex or ambiguous, the information item semantic encoder based on the Bert model may not be able to fully and accurately capture all the semantic information in the employee's personnel information. Although the BiLSTM model can process the contextual information of sequence data, it may not be able to fully capture all the complex contextual associations related to employee positions and responsibilities. At the same time, when the first scale and the second scale employee personnel information context semantic association fusion feature vectors are fused, additional noise or errors may be introduced, affecting the quality of the final feature, thereby causing the employee personnel information semantic association fusion multi-scale representation feature vector to have a category harmony offset relative to the predetermined class label of the classifier. Based on this, in order to reduce the category harmony deviation of the employee personnel information semantic association fusion multi-scale representation feature vector, in a specific example of the present application, the employee personnel information semantic association fusion multi-scale representation feature vector is subjected to a reverse error analysis based on backward scaling optimization to obtain the optimized employee personnel information semantic association fusion multi-scale representation feature vector.
[0042] Specifically, the feature optimization unit 141 includes: a probabilistic activation subunit, which is used to perform probabilistic activation on the employee personnel information semantic association fusion multi-scale representation feature vector based on the Sigmoid function to obtain a probabilistic employee personnel information semantic association fusion multi-scale representation feature vector; a pre-classification subunit, which is used to input the employee personnel information semantic association fusion multi-scale representation feature vector into the classifier to obtain an initial class label probability value; an error measurement subunit, which is used to calculate the absolute value of the difference between the feature value of each position in the probabilistic employee personnel information semantic association fusion multi-scale representation feature vector and the initial class label probability value to obtain a backward machine learning fine-grained error measurement vector; an attention optimization subunit, which is used to use the backward machine learning fine-grained error measurement vector as a key vector, The employee personnel information semantic association fusion multi-scale representation feature vector is used as a query vector and a value vector, and the key vector, the query vector and the value vector are input into a converter structure to obtain a backward learning type attention optimized employee personnel information semantic association fusion multi-scale representation feature vector; a scaling factor calculation subunit is used to calculate the ratio between the sum of the backward machine learning fine-grained error measurement vector and the length of the backward machine learning fine-grained error measurement vector, and calculate the natural exponential function value with the ratio as a power as the optimization scaling factor; a scale scaling optimization subunit is used to calculate the product between the optimization scaling factor and the backward learning type attention optimized employee personnel information semantic association fusion multi-scale representation feature vector to obtain the optimized employee personnel information semantic association fusion multi-scale representation feature vector.
[0043] That is, the following feature optimization formula is used to perform a reverse error analysis based on backward scaling optimization on the employee personnel information semantic association fusion multi-scale representation feature vector to obtain the optimized employee personnel information semantic association fusion multi-scale representation feature vector, wherein the feature optimization formula is: ; ;in, represents the probabilistic employee personnel information semantic association fusion multi-scale representation feature vector obtained after the employee personnel information semantic association fusion multi-scale representation feature vector is probabilistically activated, represents the initial class label probability value, It means subtracting by position. represents the fine-grained error metric vector for backward machine learning, represents the first The eigenvalues at the positions, The value of is the length of the backward machine learning fine-grained error metric vector, represents the transpose of a vector, represents vector multiplication, represents the normalized exponential function, It means point multiplication by position. Represents the optimized employee personnel information semantic association fusion multi-scale representation feature vector.
[0044] Based on this, in the technical solution of the present application, the employee personnel information semantic association fusion multi-scale representation feature vector is subjected to reverse error analysis based on backward scaling optimization. First, the employee personnel information semantic association fusion multi-scale representation feature vector is subjected to probabilistic activation based on the Sigmoid function to obtain a probabilistic employee personnel information semantic association fusion multi-scale representation feature vector, and the employee personnel information semantic association fusion multi-scale representation feature vector is input into the classifier to obtain the initial class label probability value. Then, the error between the probabilistic employee personnel information semantic association fusion multi-scale representation feature vector and the initial class label probability value is calculated through backward error analysis to generate a fine-grained error metric vector, which quantifies the viewpoint metric contribution of each feature position to the classification error.
[0045] Then, the backward machine learning fine-grained error metric vector is used as the key vector, and the employee personnel information semantic association fusion multi-scale representation feature vector is used as the query vector and the value vector, and the key vector, the query vector and the value vector are input into the transformer structure to perform error metric-based attention-based backward machine learning on the employee personnel information semantic association fusion multi-scale representation feature vector through a Transformer-like architecture. In this way, a backward interpretability analysis is performed on the employee personnel information semantic association fusion multi-scale representation feature vector so that the obtained backward learning-like attention-optimized employee personnel information semantic association fusion multi-scale representation feature vector has a backward learning sensitive preference relative to the predetermined class label of the classifier. And further calculate the ratio of the sum of the error metric vector and the vector length, and use this ratio as the exponential to calculate the natural exponential function value to obtain the optimized scaling factor to perform reverse scaling optimization (Backward Optimization) on the backward learning type attention optimized employee personnel information semantic association fusion multi-scale representation feature vector. In this way, the sensitivity preference and robustness of the employee personnel information semantic association fusion multi-scale representation feature vector relative to the predetermined class label of the classifier can be effectively improved, thereby improving the accuracy of the classification result.
[0046] Specifically, the access permission generation unit 142 is used to pass the optimized employee personnel information semantic association fusion multi-scale representation feature vector through a classifier-based access permission manager to obtain a management result, and the management result is used to represent the access permission label of the employee to be processed. That is, the feature vector of the optimized employee personnel information semantic association fusion multi-scale representation is analyzed by the classifier to identify the deep-level features related to the access permission of the employee to be processed, and based on the relationship between employee personnel information and permission mapping learned by the classifier during the training process, the feature vector of the optimized employee personnel information semantic association fusion multi-scale representation is mapped to different access permission labels, thereby realizing accurate labeling of the access permission of the employee to be processed.
[0047] Figure 5 FIG. 1 is a block diagram of an access rights generation unit in a personnel file intelligent management system according to an embodiment of the present application. Figure 5 As shown, the access permission generating unit 142 includes: a fully connected encoding subunit 1421, used to use the fully connected layer of the access permission manager to fully connect the encoding feature vector of the optimized employee personnel information semantic association fusion multi-scale representation to obtain a fully connected encoding feature vector; a probabilization subunit 1422, used to pass the fully connected encoding feature vector through the Softmax classification function of the access permission manager to obtain the probability value of the optimized employee personnel information semantic association fusion multi-scale representation feature vector belonging to each access permission label; a management result determining subunit 1423, used to determine the access permission label corresponding to the largest one among the probability values as the management result.
[0048] In summary, according to the embodiment of the present application, the intelligent personnel file management system is explained, which extracts the personnel files of the employees to be processed from the employee personnel file library, and extracts the employee's name, department, position and responsibility information from the personnel file, uses deep learning technology to perform semantic understanding and context analysis, and deeply understands the employee's authority requirements, so as to intelligently assign access rights to the personnel file library. In this way, differentiated access to personnel files can be achieved according to the employee's position and work needs, ensuring the compliance of personnel information.
[0049] Figure 6 Flowchart of the personnel file intelligent management method according to the embodiment of the present application. Figure 6As shown, according to the embodiment of the present application, the intelligent personnel file management method includes the following steps: S110, extracting the personnel file of the employee to be processed from the employee personnel file library, and extracting the employee personnel information from the personnel file of the employee to be processed, wherein the employee personnel information includes the employee's name, department, position and responsibilities; S120, semantically encoding each information item in the employee personnel information to obtain a sequence of semantic encoding feature vectors of the employee personnel information items; S130, performing multi-scale semantic association analysis on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a semantic association fusion multi-scale representation feature vector of the employee personnel information; S140, determining the access permission label of the employee to be processed based on the semantic association fusion multi-scale representation feature vector of the employee personnel information.
[0050] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned personnel file intelligent management method have been described in the above reference. Figures 1 to 5 The description of the personnel file intelligent management system has been introduced in detail, and therefore, its repeated description will be omitted.
[0051] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to being implemented by adopting the above specific details.
[0052] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the module division is only a logical function division, and there may be other division methods in actual implementation. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.
[0053] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0054] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.
[0055] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference to a figure in a claim should not be considered as limiting the claim to which it relates.
[0056] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units stated in the system claims can also be implemented by one unit through software or hardware.
[0057] Finally, it should be noted that the above description has been given for the purpose of illustration and description. In addition, the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.
Claims
1. An intelligent management system for personnel files, characterized in that: include: An employee personnel information extraction module is used to extract the personnel files of the employees to be processed from the employee personnel file library, and extract the employee personnel information from the personnel files of the employees to be processed, wherein the employee personnel information includes the name of the employee, department to which he belongs, position and responsibilities; A personnel information semantic understanding module, used to semantically encode each information item in the employee personnel information to obtain a sequence of semantic encoding feature vectors of the employee personnel information items; A personnel information context semantic association analysis module, used to perform multi-scale semantic association analysis on the sequence of semantic encoding feature vectors of employee personnel information items to obtain a multi-scale representation feature vector of employee personnel information semantic association fusion; An access rights management module, configured to determine the access rights label of the employee to be processed based on the semantic association of the employee's personnel information and the fusion of the multi-scale representation feature vector; The personnel information context semantic association analysis module includes: A first-scale semantic association fusion unit, configured to perform global context association coding on the sequence of semantic encoding feature vectors of employee personnel information items to obtain a first-scale employee personnel information context semantic association fusion feature vector; A second-scale semantic association fusion unit, used for weighted fusion of the sequence of semantic encoding feature vectors of employee personnel information items to obtain a second-scale employee personnel information context semantic association fusion feature vector; A multi-scale fusion unit, used for fusing the first-scale employee personnel information context semantic association fusion feature vector and the second-scale employee personnel information context semantic association fusion feature vector to obtain the employee personnel information semantic association fusion multi-scale representation feature vector; Wherein, the access permission management module includes: A feature optimization unit, used for optimizing the feature distribution of the employee personnel information semantic association fusion multi-scale representation feature vector to obtain an optimized employee personnel information semantic association fusion multi-scale representation feature vector; An access rights generating unit, used for fusing the optimized employee personnel information semantic association with the multi-scale representation feature vector through a classifier-based access rights manager to obtain a management result, wherein the management result is used to represent the access rights label of the employee to be processed; Wherein, the feature optimization unit is used to: Performing a reverse error analysis based on backward scaling optimization on the semantic association fusion multi-scale representation feature vector of the employee personnel information to obtain the optimized semantic association fusion multi-scale representation feature vector of the employee personnel information; Wherein, the feature optimization unit comprises: A probabilistic activation subunit, used for performing probabilistic activation on the employee personnel information semantic association fusion multi-scale representation feature vector based on a Sigmoid function to obtain a probabilistic employee personnel information semantic association fusion multi-scale representation feature vector; A pre-classification subunit, used for inputting the employee personnel information semantic association fusion multi-scale representation feature vector into the classifier to obtain an initial class label probability value; An error measurement subunit, used to calculate the absolute value of the difference between the feature value of each position in the probabilistic employee personnel information semantic association fusion multi-scale representation feature vector and the initial class label probability value to obtain a backward machine learning fine-grained error measurement vector; An attention optimization subunit is used to use the backward machine learning fine-grained error metric vector as a key vector, the employee personnel information semantic association fusion multi-scale representation feature vector as a query vector and a value vector, and input the key vector, the query vector and the value vector into a converter structure to obtain a backward learning type attention optimization employee personnel information semantic association fusion multi-scale representation feature vector; A scaling factor calculation subunit, used to calculate the ratio between the sum of the backward machine learning fine-grained error metric vector and the length of the backward machine learning fine-grained error metric vector, and calculate a natural exponential function value with the ratio as a power as an optimization scaling factor; A scale scaling optimization subunit, used for calculating the product between the optimization scaling factor and the backward learning type attention optimization employee personnel information semantic association fusion multi-scale representation feature vector to obtain the optimized employee personnel information semantic association fusion multi-scale representation feature vector; Among them, the following feature optimization formula is used to perform reverse error analysis based on backward scaling optimization on the employee personnel information semantic association fusion multi-scale representation feature vector to obtain the optimized employee personnel information semantic association fusion multi-scale representation feature vector, wherein the feature optimization formula is: ; in, represents the probabilistic employee personnel information semantic association fusion multi-scale representation feature vector obtained after the employee personnel information semantic association fusion multi-scale representation feature vector is probabilistically activated, represents the initial class label probability value, It means subtracting by position. represents the fine-grained error metric vector for backward machine learning, represents the first The eigenvalues at the positions, The value of is the length of the backward machine learning fine-grained error metric vector, represents the transpose of a vector, represents vector multiplication, represents the normalized exponential function, It means point multiplication by position. Represents the optimized employee personnel information semantic association fusion multi-scale representation feature vector.
2. According to claim 1, the intelligent management system for personnel files is characterized in that: The personnel information semantic understanding module is used to: Each information item in the employee personnel information is semantically encoded by an information item semantic encoder based on the Bert model to obtain a sequence of semantic encoding feature vectors of the employee personnel information items.
3. The intelligent management system for personnel files according to claim 2 is characterized in that: The first-scale semantic association fusion unit is used to: The sequence of semantically encoded feature vectors of the employee personnel information items is input into an employee personnel information context semantic association encoder based on a BiLSTM model to obtain the first-scale employee personnel information context semantic association fusion feature vector.
4. The intelligent personnel file management system according to claim 3 is characterized in that: The access authority generating unit includes: A fully connected encoding subunit, configured to use the fully connected layer of the access permission manager to perform fully connected encoding on the optimized employee personnel information semantic association fusion multi-scale representation feature vector to obtain a fully connected encoding feature vector; A probabilization subunit, used for passing the fully connected encoded feature vector through the Softmax classification function of the access authority manager to obtain a probability value of the optimized employee personnel information semantic association fusion multi-scale representation feature vector belonging to each access authority label; The management result determination subunit is used to determine the access permission label corresponding to the largest probability value as the management result.
5. A personnel file intelligent management method, using the personnel file intelligent management system according to claim 1, characterized in that: include: Extracting the personnel files of the employees to be processed from the employee personnel file library, and extracting the personnel information of the employees from the personnel files of the employees to be processed, wherein the personnel information of the employees includes the name, department, position and responsibilities of the employees; Semantically encoding each information item in the employee personnel information to obtain a sequence of semantic encoding feature vectors of the employee personnel information items; Performing multi-scale semantic association analysis on the sequence of semantic encoding feature vectors of the employee personnel information items to obtain a multi-scale representation feature vector of the employee personnel information semantic association fusion; Based on the semantic association of the employee's personnel information and the fusion of the multi-scale representation feature vector, the access permission label of the employee to be processed is determined.
6. The intelligent personnel file management method according to claim 5 is characterized in that: Semantically encoding each information item in the employee personnel information to obtain a sequence of semantic encoding feature vectors of the employee personnel information items, including: Each information item in the employee personnel information is semantically encoded by an information item semantic encoder based on the Bert model to obtain a sequence of semantic encoding feature vectors of the employee personnel information items.
Citation Information
Patent Citations
Chinese named entity recognition method and device based on multi-view semantic feature fusion
CN114580416A
User authority unified management method and device for enterprise-level application system
CN117592035A
