Human resource information intelligent allocation system and method based on data mining

Abstract

The invention relates to the field of enterprise management, and provides a human resource information intelligent allocation system and method based on data mining, and the system carries out the semantic understanding of the professional skill information, work project information and performance information of employees applying for an exchange department, so as to understand the work capability of the employees applying for the exchange department. And meanwhile, semantic understanding is performed on post requirements of vacant posts so as to understand skills, capabilities and qualities required by the vacant posts, and then a relationship between the working capabilities of the employees applying for the exchange department and the skills, capabilities and qualities required by the vacant posts is established in the same feature space, so that whether the employees applying for the exchange department are matched with the vacant posts or not is judged. In this way, the matching degree of the employees and the posts can be evaluated more scientifically, and therefore optimal configuration of human resources is achieved.

Description

Technical Field

[0001] The present application relates to the field of enterprise management, and more specifically, to a human resources information intelligent allocation system and method based on data mining. Background Art

[0002] The essence of corporate competition is talent competition. One of the main ways to reduce the cost of human resource management is to improve the efficiency of human resource allocation. Allocation management means that when a job vacancy occurs, the business department reports the information to the human resources department and publishes the relevant information of the job in the corporate office system. The system allocates the human resource information according to scientific and standardized processes and management methods to meet the company's employment needs. In traditional human resource allocation, it often relies on manual judgment to determine the matching degree between the employees of the department applying for transfer and the vacant positions, which is prone to incomplete and inaccurate information. In addition, this manual judgment process mainly includes manual review and interviews. This process is time-consuming, and manual evaluation may be affected by personal preferences, experience and other factors.

[0003] Therefore, an optimized intelligent allocation solution for human resource information is needed. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present application provides a human resources information intelligent allocation system and method based on data mining.

[0005] According to one aspect of the present application, a human resources information intelligent allocation system based on data mining is provided, which includes:

[0006] The employee transfer-related data collection module is used to obtain the professional skills information, work project information, performance information and job requirements of vacant positions of employees applying for transfer;

[0007] An employee deployment-related data encoding module is used to encode the professional skill information, work project information, performance information of the employee applying for transfer and the job requirements of the vacant position to obtain a semantic understanding feature vector of the employee's work ability and a semantic understanding feature vector of the job requirements;

[0008] An employee work ability semantic understanding feature optimization module, used for performing cumulative presentation optimization on the employee work ability semantic understanding feature vector based on the position requirement semantic understanding feature vector based on fine-grained difference processing to obtain an optimized employee work ability semantic understanding feature vector;

[0009] An employee job matching feature generation module, used for performing position point multiplication on the optimized employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector to obtain an employee work ability-job requirement matching feature vector;

[0010] The employee position matching result generating module is used to obtain the matching result between the employee of the department applying for transfer and the vacant position based on the employee work ability-position requirement matching feature vector.

[0011] According to another aspect of the present application, a method for intelligent allocation of human resources information based on data mining is provided, which includes:

[0012] Obtain professional skills information, work project information, performance information and job requirements of vacant positions of employees applying for transfer;

[0013] Data encoding is performed on the professional skill information, work project information, performance information of the employee of the department applying for transfer and the job requirements of the vacant position to obtain a semantic understanding feature vector of the employee's work ability and a semantic understanding feature vector of the job requirements;

[0014] Based on the position requirement semantic understanding feature vector, the employee work ability semantic understanding feature vector is optimized by cumulative presentation based on fine-grained difference processing to obtain an optimized employee work ability semantic understanding feature vector;

[0015] Multiplying the optimized employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector by position point to obtain an employee work ability-job requirement matching feature vector;

[0016] Based on the employee work ability-position requirement matching feature vector, a matching result between the employee applying for transfer and the vacant position is obtained.

[0017] This application has significant technical effects due to the adoption of the above technical solutions:

[0018] The human resource information intelligent allocation system and method based on data mining provided by the present application understands the working ability of the employees in the department applying for transfer by semantically understanding the professional skill information, work project information, and performance information of the employees in the department applying for transfer, and understands the skills, abilities, and qualities required for the vacant positions by semantically understanding the job requirements of the vacant positions, and then establishes the relationship between the working ability of the employees in the department applying for transfer and the skills, abilities, and qualities required for the vacant positions in the same feature space, so as to judge whether the employees in the department applying for transfer match the vacant positions. In this way, the matching degree between employees and positions can be evaluated more scientifically, thereby realizing the optimal allocation of human resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] 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.

[0020] Figure 1 The present invention is a flowchart of a method for intelligently allocating human resources information based on data mining according to an embodiment of the present application.

[0021] Figure 2 In the method for intelligent allocation of human resources information based on data mining according to an embodiment of the present application, data encoding is performed on the professional skill information, work project information, performance information of the employees of the department applying for transfer, and the job requirements of the vacant positions to obtain a flowchart of the semantic understanding feature vector of the employee's work ability and the semantic understanding feature vector of the job requirement.

[0022] Figure 3 The present invention is a flowchart of an intelligent allocation method of human resource information based on data mining according to an embodiment of the present application, which extracts semantic features from the professional skill information, work project information, and performance information of the employees of the department applying for transfer to obtain a semantic understanding feature vector of the employee's work ability.

[0023] Figure 4 The system block diagram of the human resources information intelligent deployment system based on data mining according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. 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 exemplary embodiments described here.

[0025] In today's highly competitive business environment, competition between enterprises is essentially a competition for talent. Having an excellent team of talents can bring innovation vitality, efficient execution and sustainable development momentum to enterprises. Reducing the cost of human resource management has become one of the keys to improving the competitiveness of enterprises. Among them, improving the efficiency of human resource deployment is undoubtedly an effective method.

[0026] Human resource allocation management is an important part of enterprise human resource management. When a job vacancy occurs, the business department will promptly report the relevant information to the human resources department, and then publish the detailed information of the position in the enterprise office system, including job responsibilities, job requirements, work location, etc. The system allocates the human resource information within the enterprise based on scientific and standardized processes and management methods to quickly and accurately meet the enterprise's employment needs.

[0027] However, in the traditional human resource allocation model, the matching degree between the employees applying for transfer and the vacant positions is often manually judged. This process is prone to incomplete and inaccurate information. For example, manual judgment may only be based on the employee's past work experience, while ignoring the employee's potential ability and development willingness. At the same time, this manual judgment process mainly includes manual review and interview, which is not only time-consuming, but also may be affected by personal preferences, experience and other factors, resulting in the evaluation results being not objective and fair. For example, the interviewer may give a higher evaluation because he has a good impression of a certain type of employee; or because of the limitations of his own experience, he cannot accurately judge whether the employee is suitable for a specific position. Therefore, a human resource information intelligent allocation solution based on data mining is expected, which can judge whether the employees applying for transfer match the vacant positions by conducting in-depth mining and analysis of the information of the employees applying for transfer and the information of the vacant positions.

[0028] In recent years, deep learning and neural networks have been widely used in computer vision, natural language processing, text signal processing and other fields. In addition, deep learning and neural networks have also demonstrated a level close to or even beyond that of humans in image classification, object detection, semantic segmentation, text translation and other fields. The development of deep learning and neural networks has provided new solutions and solutions for the intelligent allocation of human resources information.

[0029] Figure 1 FIG. 1 is a flow chart of a method for intelligently allocating human resources information based on data mining according to an embodiment of the present application. Figure 1As shown, according to the embodiment of the present application, the intelligent allocation method of human resource information based on data mining includes: S110, obtaining the professional skill information, work project information, performance information and job requirements of the vacant positions of the employees of the department applying for transfer; S120, respectively encoding the professional skill information, work project information, performance information and job requirements of the vacant positions of the employees of the department applying for transfer to obtain the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector; S130, based on the job requirement semantic understanding feature vector, performing cumulative presentation optimization based on fine-grained difference processing on the employee work ability semantic understanding feature vector to obtain the optimized employee work ability semantic understanding feature vector; S140, multiplying the optimized employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector by position point to obtain the employee work ability-job requirement matching feature vector; S150, based on the employee work ability-job requirement matching feature vector, obtaining the matching result between the employee of the department applying for transfer and the vacant position.

[0030] In step S110, the professional skills information, work project information, performance information and job requirements of the vacant positions of the employees in the department applying for transfer are obtained. It should be understood that the professional skills information, work project information and performance information of the employees in the department applying for transfer specifically include the professional skills and skill proficiency data mastered by the employees and the project roles, responsibilities, project results data and employee performance, reward data, etc. of the employees participating in the work projects. The job requirements of the vacant positions include the basic information of the job title and type. In addition, it also includes information such as the skill requirements, work experience requirements, and educational background requirements of the positions. In general, the professional skills information, work project information, performance information and job requirements of the employees in the department applying for transfer, which are obtained, are important data sources for judging whether the employees match the positions. By comprehensively considering the skill matching, experience matching and ability matching, the matching degree between the employees in the department applying for transfer and the vacant positions can be comprehensively evaluated, thereby providing a scientific and reasonable decision-making basis for human resource allocation.

[0031] In step S120, the professional skills information, work project information, performance information of the employee applying for transfer and the job requirements of the vacant position are respectively encoded to obtain the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector. Specifically, Figure 2 In the method for intelligently allocating human resources information based on data mining according to an embodiment of the present application, the professional skill information, work project information, performance information of the employee applying for transfer and the job requirements of the vacant position are respectively encoded to obtain a semantic understanding feature vector of the employee's work ability and a semantic understanding feature vector of the job requirements. Figure 2As shown, the professional skill information, work project information, performance information of the employees of the department applying for transfer and the job requirements of the vacant positions are respectively data encoded to obtain a semantic understanding feature vector of the employee's work ability and a semantic understanding feature vector of the job requirements, including: S121, performing semantic feature extraction on the professional skill information, work project information, and performance information of the employees of the department applying for transfer to obtain the semantic understanding feature vector of the employee's work ability; S122, performing semantic feature extraction on the job requirements of the vacant positions to obtain the semantic understanding feature vector of the job requirements.

[0032] More specifically, Figure 3 The flowchart is a method for intelligently allocating human resources information based on data mining according to an embodiment of the present application, in which semantic features are extracted from the professional skills information, work project information, and performance information of the employee applying for transfer to obtain the semantic understanding feature vector of the employee's work ability. Figure 3 As shown, semantic features are extracted from the professional skill information, work project information, and performance information of the employee of the department applying for transfer to obtain a semantic understanding feature vector of the employee's work ability, including: S1211, preprocessing the professional skill information, work project information, and performance information of the employee of the department applying for transfer to obtain important concern information of the employee applying for transfer; S1212, segmenting the important concern information of the employee applying for transfer and passing it through a medium- and short-distance work ability semantic understander to obtain a medium- and short-distance employee work ability semantic understanding feature vector; S1213, segmenting the important concern information of the employee applying for transfer and passing it through a long-distance work ability semantic understander to obtain a long-distance employee work ability semantic understanding feature vector; S1214, fusing the medium- and short-distance employee work ability semantic understanding feature vector and the long-distance employee work ability semantic understanding feature vector to obtain the employee work ability semantic understanding feature vector.

[0033] Considering that the professional skills information, work project information, and performance information of the employees in the department applying for transfer come from different sources, with different formats and specifications, and even errors and missing information, these problems will reduce the quality of the data and affect the reliability of the employee data analysis results. In addition, the original data may contain a large amount of information that is irrelevant to the job matching analysis. This information is noise for the model and will interfere with the model's judgment. Therefore, in the technical solution of this application, it is necessary to pre-process the professional skills information, work project information, and performance information of the employees in the department applying for transfer to obtain important information of concern to the employees applying for transfer. After pre-processing, the quality of the relevant information of the employees applying for transfer will be improved, reducing the interference of errors and noise in the original data, and improving the accuracy of employee data analysis.

[0034] In order to help the human resource deployment model better understand the semantic information of the important information of the employees applying for job transfer, it is first necessary to perform word segmentation on the important information of the employees applying for job transfer. Through word segmentation, the keyword vocabulary in the original employee data can be obtained, and then the semantic relationship between the vocabulary is analyzed, and the overall meaning of the employee information text can be well understood. That is, the important information of the employees applying for job transfer after word segmentation is passed through the medium and short-distance work ability semantic understander to obtain the medium and short-distance employee work ability semantic understanding feature vector. Through the medium and short-distance work ability semantic understander, it is possible to identify in which fields the employees have professional skills and the degree of mastery of professional skills, and it is also possible to extract the employees' work experience characteristics and ability tendency characteristics. By extracting these features, the employees' abilities can be comprehensively and deeply evaluated, so as to more accurately judge whether the employees are suitable for the target positions. The medium and short-distance work ability semantic understander described in this application is a bidirectional long short-term memory neural network model including a word embedding layer. The word embedding layer can map the vocabulary of the important information of the employees applying for job transfer to the same space, so that similar vocabulary is close in distance in space, and this representation method can capture the semantic relationship between the original employee text vocabulary. The long short-term memory neural network model solves the problem of gradient vanishing or gradient exploding that is prone to occur in traditional recurrent neural networks when processing long sequences by introducing a "gate" mechanism (forget gate, input gate, output gate), allowing the model to better capture distance dependencies. The bidirectional long short-term memory neural network model combines LSTM in two directions, further improving the model's modeling capabilities for text sequences. After being processed by the medium- and short-distance work ability semantic understander, it can deeply understand the semantic content of the important information of employees applying for job transfers, accurately capture the characteristics of employees' professional skills, work experience, and performance, and construct a comprehensive and in-depth semantic understanding feature vector of employees' work ability.

[0035] The evaluation of employee work ability is not limited to the surface information of the important information of the employee applying for transfer, but also needs to deeply understand the deep semantic features hidden in the surface information. In order to capture the long-distance dependency in the text of the important information of the employee applying for transfer, it is necessary to segment the important information of the employee applying for transfer and pass it through the long-distance work ability semantic understander to obtain the long-distance employee work ability semantic understanding feature vector. The long-distance work ability semantic understander can deeply understand the semantic relationship in the text through deep learning technology, capture the long-distance dependency that is difficult to capture by traditional methods, and help to understand the employee's work ability more accurately and improve the accuracy of the evaluation. The long-distance work ability semantic understander described in this application is a semantic encoder based on a converter including a word embedding layer. The converter can directly calculate the dependency between any two positions in the sequence through the self-attention mechanism, thereby effectively capturing the long-distance dependency, and then understanding the employee's true work ability and potential. And the converter model can process the entire sequence in parallel during the training process, which greatly improves the calculation efficiency, which is conducive to processing large-scale human resources data.

[0036] The medium- and short-distance employee work ability semantic understanding feature vector mainly focuses on the medium- and short-distance dependencies between employee data features, while the long-distance employee work ability semantic understanding feature vector focuses on capturing the long-distance dependencies between employee data features. In order to obtain more comprehensive and in-depth employee information, the medium- and short-distance employee work ability semantic understanding feature vector and the long-distance employee work ability semantic understanding feature vector are integrated in the technical solution of this application to obtain the employee work ability semantic understanding feature vector. Through integration, the employee work ability features are made more three-dimensional and multi-dimensional, which helps to more accurately portray the comprehensive capabilities of employees.

[0037] In order to cut the job requirements of the vacant positions into meaningful words, it is necessary to perform word segmentation on the job requirements of the vacant positions. Compared with the entire paragraph, a single word requires less time and resources when processing, which helps to speed up the entire processing flow and word segmentation helps to better capture the semantic information in the job requirements text. Then the job requirements text after word segmentation is passed through the job requirements semantic understander to obtain the job requirements semantic understanding feature vector. Through the processing of the job requirements semantic understander, it is possible to obtain but not limited to job skill requirements, job education background requirements, work experience requirements, etc. These extracted features have an important impact on job matching judgments because they are directly related to whether the employee has the ability and quality to meet the job requirements. The job requirements semantic understander described in this application is a converter-based semantic encoder including a word embedding layer. The word embedding layer can convert words into vector representations, and these vectors can capture the semantic similarities and differences between words. The encoder based on the converter can further utilize these vector representations to deeply understand the deep semantic information of the job requirements through mechanisms such as self-attention mechanisms.

[0038] In step S130, based on the job requirement semantic understanding feature vector, the employee work ability semantic understanding feature vector is subjected to cumulative presentation optimization based on fine-grained difference processing to obtain an optimized employee work ability semantic understanding feature vector. Accordingly, considering that different preprocessing steps may be used when preprocessing employee information and job requirements, and in actual applications, employee work project information and performance information may have data missing or sparsity problems, while job requirements may be relatively complete, data sparsity will affect the quality and fine-grained inconsistency of feature distribution of the final generated employee work ability semantic understanding feature vector and job requirement semantic understanding feature vector. Since there is a fine-grained inconsistency in feature distribution between the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector, this will cause the fused employee work ability-job requirement matching feature vector to have abnormal stacking on the boundary of the feature manifold. Based on this, in the technical solution of the present application, before fusing the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector to obtain the employee work ability-job requirement matching feature vector, the employee work ability semantic understanding feature vector is subjected to cumulative presentation optimization based on fine-grained difference processing based on the job requirement semantic understanding feature vector.

[0039] Specifically, in an embodiment of the present application, based on the job requirement semantic understanding feature vector, the employee work ability semantic understanding feature vector is subjected to cumulative presentation optimization based on fine-grained difference processing to obtain an optimized employee work ability semantic understanding feature vector, including: calculating the positional difference between the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector to obtain an employee matching shifted positional difference representation vector; after cascading the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector into an employee matching joint feature vector, the employee matching joint feature vector is input into a one-dimensional convolutional layer to obtain an employee matching shifted convolutional coding representation vector; after linearly embedding and modulating the employee work ability semantic understanding feature vector using a first embedding matrix to obtain an employee work ability semantic understanding feature linear modulation vector, calculating Calculate the positional difference between the employee work ability semantic understanding feature linear modulation vector and the job requirement semantic understanding feature vector to obtain an employee matching shift modulation difference representation vector; use the employee matching shifted positional difference representation vector as the value vector, the employee matching shifted convolutional coding representation vector and the employee matching shift modulation difference representation vector as the query vector and key vector, input the employee matching shifted positional difference representation vector, the employee matching shifted convolutional coding representation vector and the employee matching shift modulation difference representation vector into a shift corrector based on an imitation transformer structure to obtain an employee matching multi-scale hierarchical fusion shift modulation weight vector; calculate the positional point multiplication between the employee matching multi-scale hierarchical fusion shift modulation weight vector and the employee work ability semantic understanding feature vector to obtain the optimized employee work ability semantic understanding feature vector.

[0040] In an embodiment of the present application, based on the position requirement semantic understanding feature vector, the employee work ability semantic understanding feature vector is subjected to cumulative presentation optimization based on fine-grained difference processing to obtain an optimized employee work ability semantic understanding feature vector, including: processing the employee work ability semantic understanding feature vector with the position requirement semantic understanding feature vector according to the following formula to obtain an optimized employee work ability semantic understanding feature vector; wherein the formula is:

[0041]

[0042] V 2d =Conv 1D ([V1; V2])

[0043]

[0044] V1′=V1⊙V d

[0045] Among them, V1 represents the semantic understanding feature vector of the employee's work ability, V2 represents the semantic understanding feature vector of the job requirement, Indicates point-by-point subtraction, V 1d Indicates that the employee matching shift is represented by position difference vector, [;] indicates cascade operation, Conv 1D (·) represents one-dimensional convolutional coding, V 2d represents the employee matching shift convolutional coding representation vector, W1 represents the first embedding matrix, B1 represents the first modulation bias vector, V 3d represents the staff matched shift modulated differential representation vector, represents vector multiplication, Softmax represents the normalized exponential function, T represents the transposed vector of the feature vector, d represents the length of the feature vector, Relu represents the activation function, V d represents the employee matching multi-scale hierarchical fusion shift modulation weight vector, ⊙ represents the position point multiplication, and V1' represents the optimized employee work ability semantic understanding feature vector.

[0046] In the technical solution of the present application, in the process of fusing the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector, a cumulative presentation optimization method based on fine-grained difference processing is proposed to address the problem of fine-grained inconsistency in feature distribution that arises in the process of fusing the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector. The method aims to reduce the abnormal stacking phenomenon of the fused feature vector on the boundary of the feature manifold, so as to improve the robustness and classification performance of the model.

[0047] Specifically, the method first calculates the element-by-element difference between the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector to obtain the employee matching shift position difference representation vector, thereby capturing the difference between the two feature vectors. Then, the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector are cascaded into a joint feature vector and input into a one-dimensional convolution layer to generate an employee matching shift convolution coding representation vector. That is, the deep-level shift implicit information between the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector is captured by one-dimensional convolution coding. At the same time, the employee work ability semantic understanding feature vector is linearly embedded and modulated using the first embedding matrix to be encoded into an employee work ability semantic understanding feature linear modulation vector, and the element-by-element difference between the employee work ability semantic understanding feature linear modulation vector and the job requirement semantic understanding feature vector is calculated to obtain an employee matching shift modulation difference representation vector. The employee matching shift modulation difference representation vector is used to represent the position expression of the shift information of the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector in the linear space. That is, in the technical solution of the present application, the employee matching shift position differential representation vector, the employee matching shift convolutional coding representation vector and the employee matching shift modulation differential representation vector respectively represent the explicit expression of the fine-grained distribution difference shift information between the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector in different feature spaces.

[0048] Next, the employee matching shifted positional differential representation vector is used as the value vector, the employee matching shifted convolutional coding representation vector and the employee matching shifted modulated differential representation vector are used as the query vector and the key vector respectively, and these vectors are input into the shift corrector based on the simulated transformer structure to generate the employee matching multi-scale hierarchical fusion shift modulation weight vector. In this way, the attention mechanism is used to evaluate the importance of features at different scales, that is, the attention mechanism is used to simulate the hierarchical precipitation expression of the fine-grained difference feature distribution between the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector in different feature spaces. Finally, the element-by-element product of the employee matching multi-scale hierarchical fusion shift modulation weight vector and the employee work ability semantic understanding feature vector is calculated to obtain the optimized employee work ability semantic understanding feature vector, that is, the weighted operation is used to assign weights to each element in the employee work ability semantic understanding feature vector, which reduces the influence of local disturbances and improves the quality of the feature vector.

[0049] In summary, in the method, the employee matching shift position differential representation vector, the employee matching shift convolutional coding representation vector and the employee matching shift modulation differential representation vector respectively represent the explicit expression of the fine-grained distribution difference shift information between the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector in different feature spaces. The generation of the employee matching multi-scale hierarchical fusion shift modulation weight vector uses the attention mechanism to evaluate the importance of features at different scales and provide weight information for feature shift correction. Ultimately, the generation of the optimized employee work ability semantic understanding feature vector reduces the impact of disturbances, improves the model's robustness to local disturbances in the input data, enhances the model's adaptability to noisy or variant data sets, improves the fusion quality of feature vectors, and improves the overall classification performance and training efficiency of the model.

[0050] In step S140, the optimized employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector are multiplied by position to obtain the employee work ability-job requirement matching feature vector. It should be understood that in order to reflect the degree of matching between the employees applying for replacement and the vacant positions, the optimized employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector need to be multiplied by position in the technical solution of the present application. Positional point multiplication allows each element in the feature vector to be directly multiplied by the element in its corresponding position. In the two feature vectors of optimized employee work ability and job requirements, elements in the same position usually represent similar abilities or requirements. Through this direct comparison, the degree of fit between employees and job requirements in various aspects can be quantified. The result of the point multiplication reflects the degree of matching between employee abilities and job requirements. If the employee's ability is very strong and the job requirements are also high in a certain feature dimension, then the product of the two feature values ​​will be very large, indicating that the employee and the job are highly matched in this dimension. On the contrary, if the employee's ability or job requirements are low in a certain dimension, the product will be small, indicating that the matching degree is low. That is, the employee work ability-job requirement matching feature vector obtained by multiplying the position points provides a quantitative standard for judging whether the employee matches the vacant job.

[0051] In step S150, based on the employee work ability-position requirement matching feature vector, the matching result between the employee of the department applying for transfer and the vacant position is obtained. Specifically, in the embodiment of the present application, based on the employee work ability-position requirement matching feature vector, the matching result between the employee of the department applying for transfer and the vacant position is obtained, including: passing the employee work ability-position requirement matching feature vector through a position matching result generator based on a classifier to obtain a classification result, and the classification result is to judge whether the employee of the department applying for transfer matches the vacant position. It should be understood that a classifier is a machine learning model that predicts output based on the input employee work ability-position requirement matching feature vector. The classifier learns patterns in a large amount of historical data through training, so that it can identify the category to which the new input data belongs. The classification result provides an objective decision-making basis for the enterprise, which helps to reduce the subjectivity and uncertainty of human judgment. The enterprise can evaluate the degree of matching between employees and positions based on the classification results, so as to make more scientific and reasonable deployment decisions. Through the intelligent deployment method, the enterprise can quickly identify the employees who are most suitable for vacant positions, thereby optimizing the allocation of human resources, thereby reducing the recruitment and training costs of the enterprise and improving the overall operational efficiency.

[0052] In summary, the method for intelligent allocation of human resources information based on data mining according to the embodiment of the present application is explained, which semantically understands the professional skill information, work project information, and performance information of the employees of the department applying for transfer to understand the work ability of the employees of the department applying for transfer, and semantically understands the job requirements of the vacant positions to understand the skills, abilities, and qualities required for the vacant positions, and then establishes the relationship between the work ability of the employees of the department applying for transfer and the skills, abilities, and qualities required for the vacant positions in the same feature space, so as to judge whether the employees of the department applying for transfer match the vacant positions. In this way, the matching degree between employees and positions can be evaluated more scientifically, thereby achieving the optimal allocation of human resources.

[0053] Figure 4 : is a system block diagram of a human resources information intelligent deployment system based on data mining according to an embodiment of the present application. Figure 4As shown, according to the embodiment of the present application, the human resources information intelligent deployment system 100 based on data mining includes: an employee deployment related data acquisition module 110, which is used to obtain the professional skill information, work project information, performance information and job requirements of the vacant positions of the employees of the department applying for transfer; an employee deployment related data encoding module 120, which is used to respectively encode the professional skill information, work project information, performance information and job requirements of the vacant positions of the employees of the department applying for transfer to obtain the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector; an employee work ability semantic understanding feature optimization module 130, which is used to Based on the job requirement semantic understanding feature vector, the employee work ability semantic understanding feature vector is optimized by cumulative presentation based on fine-grained difference processing to obtain an optimized employee work ability semantic understanding feature vector; an employee job matching feature generation module 140 is used to multiply the optimized employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector by position point to obtain an employee work ability-job requirement matching feature vector; an employee job matching result generation module 150 is used to obtain a matching result between the employee of the department applying for transfer and the vacant job based on the employee work ability-job requirement matching feature vector.

[0054] Here, those skilled in the art can understand that the specific functions and operations of the various units and modules in the above-mentioned human resources information intelligent allocation system 100 based on data mining have been referred to above. Figures 1 to 3 The invention has been introduced in detail in the description of the intelligent allocation method of human resource information based on data mining, and therefore, its repeated description will be omitted.

[0055] In summary, the human resource information intelligent allocation system 100 based on data mining according to the embodiment of the present application is explained, which understands the working ability of the employees of the department applying for transfer by semantically understanding the professional skill information, work project information, and performance information of the employees of the department applying for transfer, and at the same time, understands the job requirements of the vacant positions semantically to understand the skills, abilities, and qualities required for the vacant positions, and then establishes the relationship between the working ability of the employees of the department applying for transfer and the skills, abilities, and qualities required for the vacant positions in the same feature space, so as to judge whether the employees of the department applying for transfer match the vacant positions. In this way, the matching degree between employees and positions can be evaluated more scientifically, thereby achieving the optimal allocation of human resources.

[0056] As described above, the human resources information intelligent allocation system 100 based on data mining according to the embodiment of the present application can be implemented in various wireless terminals, such as a server for intelligent allocation of human resources information based on data mining, etc. In one example, the human resources information intelligent allocation system 100 based on data mining according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the human resources information intelligent allocation system 100 based on data mining can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the human resources information intelligent allocation system 100 based on data mining can also be one of the many hardware modules of the wireless terminal.

[0057] Alternatively, in another example, the human resources information intelligent allocation system 100 based on data mining and the wireless terminal may also be separate devices, and the human resources information intelligent allocation system 100 based on data mining may be connected to the wireless terminal via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

Claims

1. A method for intelligent allocation of human resources information based on data mining, characterized in that: include: Obtain professional skills information, work project information, performance information and job requirements of vacant positions of employees applying for transfer; Data encoding is performed on the professional skill information, work project information, performance information of the employee of the department applying for transfer and the job requirements of the vacant position to obtain a semantic understanding feature vector of the employee's work ability and a semantic understanding feature vector of the job requirements; Based on the position requirement semantic understanding feature vector, the employee work ability semantic understanding feature vector is optimized by cumulative presentation based on fine-grained difference processing to obtain an optimized employee work ability semantic understanding feature vector; Multiplying the optimized employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector by position point to obtain an employee work ability-job requirement matching feature vector; Based on the employee work ability-position requirement matching feature vector, a matching result between the employee applying for transfer and the vacant position is obtained.

2. The method for intelligent allocation of human resources information based on data mining according to claim 1 is characterized in that: The professional skills information, work project information, performance information of the employee of the department applying for transfer and the job requirements of the vacant position are respectively encoded to obtain the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector, including: Extracting semantic features from the professional skills information, work project information, and performance information of the employee of the department applying for transfer to obtain a semantic understanding feature vector of the employee's work ability; Semantic features are extracted from the job requirements of the vacant positions to obtain a semantic understanding feature vector of the job requirements.

3. The method for intelligent allocation of human resources information based on data mining according to claim 2 is characterized in that: Semantic feature extraction is performed on the professional skill information, work project information, and performance information of the employee of the department applying for transfer to obtain a semantic understanding feature vector of the employee's work ability, including: Pre-process the professional skills information, work project information, and performance information of the employee applying for transfer to obtain important information of concern to the employee applying for transfer; After the important information of the employee applying for job transfer is segmented, it is passed through a medium-short distance work ability semantic understander to obtain a medium-short distance employee work ability semantic understanding feature vector; The important information of the employee applying for job transfer is segmented and passed through a long-distance work ability semantic understander to obtain a semantic understanding feature vector of long-distance employee work ability; The semantic understanding feature vector of the employee's work ability is obtained by fusing the semantic understanding feature vector of the medium- and short-distance employee's work ability with the semantic understanding feature vector of the long-distance employee's work ability.

4. The method for intelligent allocation of human resources information based on data mining according to claim 3 is characterized in that: The semantic features of the job requirements of the vacant positions are extracted to obtain the semantic understanding feature vector of the job requirements, including: segmenting the job requirements of the vacant positions and passing them through a job requirement semantic understander to obtain the semantic understanding feature vector of the job requirements.

5. The method for intelligent allocation of human resources information based on data mining according to claim 4 is characterized in that: The medium- and short-distance work ability semantic understander is a bidirectional long short-term memory neural network model including a word embedding layer, and the long-distance work ability semantic understander and the job requirement semantic understander are converter-based semantic encoders including a word embedding layer.

6. The method for intelligent allocation of human resources information based on data mining according to claim 5 is characterized in that: Based on the position requirement semantic understanding feature vector, the employee work ability semantic understanding feature vector is optimized by cumulative presentation based on fine-grained difference processing to obtain an optimized employee work ability semantic understanding feature vector, including: Calculating the positional difference between the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector to obtain an employee matching shift positional difference representation vector; After concatenating the employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector into an employee matching joint feature vector, the employee matching joint feature vector is input into a one-dimensional convolutional layer to obtain an employee matching shift convolutional coding representation vector; After performing linear embedding modulation encoding on the employee work ability semantic understanding feature vector using the first embedding matrix to obtain an employee work ability semantic understanding feature linear modulation vector, calculating the position difference between the employee work ability semantic understanding feature linear modulation vector and the job requirement semantic understanding feature vector to obtain an employee matching shift modulation difference representation vector; Taking the employee matching shifted position differential representation vector as a value vector, the employee matching shifted convolutional coding representation vector and the employee matching shifted modulation differential representation vector as a query vector and a key vector, the employee matching shifted position differential representation vector, the employee matching shifted convolutional coding representation vector and the employee matching shifted modulation differential representation vector are input into a shift corrector based on an imitation transformer structure to obtain an employee matching multi-scale hierarchical fusion shift modulation weight vector; The optimized employee work ability semantic understanding feature vector is obtained by calculating the point-by-point multiplication between the employee matching multi-scale hierarchical fusion shift modulation weight vector and the employee work ability semantic understanding feature vector.

7. The method for intelligent allocation of human resources information based on data mining according to claim 6 is characterized in that: Based on the employee work ability-job requirement matching feature vector, a matching result between the employee of the department applying for transfer and the vacant job is obtained, including: passing the employee work ability-job requirement matching feature vector through a classifier-based job matching result generator to obtain a classification result, and the classification result is used to determine whether the employee of the department applying for transfer matches the vacant job.

8. A human resources information intelligent deployment system based on data mining, characterized in that: include: The employee transfer-related data collection module is used to obtain the professional skills information, work project information, performance information and job requirements of vacant positions of employees applying for transfer; An employee deployment-related data encoding module is used to encode the professional skill information, work project information, performance information of the employee applying for transfer and the job requirements of the vacant position to obtain a semantic understanding feature vector of the employee's work ability and a semantic understanding feature vector of the job requirements; An employee work ability semantic understanding feature optimization module, used for performing cumulative presentation optimization on the employee work ability semantic understanding feature vector based on the position requirement semantic understanding feature vector based on fine-grained difference processing to obtain an optimized employee work ability semantic understanding feature vector; An employee job matching feature generation module, used for performing position point multiplication on the optimized employee work ability semantic understanding feature vector and the job requirement semantic understanding feature vector to obtain an employee work ability-job requirement matching feature vector; The employee position matching result generating module is used to obtain the matching result between the employee of the department applying for transfer and the vacant position based on the employee work ability-position requirement matching feature vector.

9. The human resources information intelligent deployment system based on data mining according to claim 8 is characterized in that: The employee deployment related data coding module includes: An employee work ability semantic understanding feature extraction unit is used to extract semantic features from the professional skill information, work project information, and performance information of the employee of the department applying for transfer to obtain a semantic understanding feature vector of the employee work ability; The job requirement semantic understanding feature extraction unit is used to extract semantic features of the job requirement of the vacant job to obtain the job requirement semantic understanding feature vector.

10. The human resources information intelligent deployment system based on data mining according to claim 9 is characterized in that: The employee position matching result generating module is used to obtain the matching result between the employee of the department applying for transfer and the vacant position based on the employee work ability-position requirement matching feature vector.

Images