Human resource information digital management method and system based on big data

Abstract

The invention relates to the technical field of information management, and discloses a human resource information digital management method and system based on big data. Firstly, human resource data index scores in the original human resource data matrix are calculated by using a principal component analysis method, and executive ability values of employees are calculated after principal components are screened; secondly, classifying the employee execution capability values by using an improved clustering algorithm, and introducing a particle swarm optimization algorithm to search an optimal initial clustering center so as to complete position distribution; training a BP neural network to obtain a BP neural network model, predicting employee performance, and outputting an employee performance prediction value; and finally, performing employee performance assessment and evaluation according to the employee performance predicted value, thereby realizing digital management of human resource information. By analyzing and processing the human resource data, the purpose of human resource information digital management is achieved, and the method is objective and accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of information management, and in particular to a method and system for digital management of human resource information based on big data. Background Art

[0002] Chinese patent CN117875921A discloses a human resource management method and system based on artificial intelligence, which specifically includes obtaining human resource demand information, screening job resumes, and obtaining candidate resumes; training a natural speech processing model to obtain a trained natural language processing model, obtaining a first keyword for job information, the first keyword including first semantic feature information, first part of speech information, and first position information; processing the work experience information of a second preset position through the trained natural language processing model to obtain a second keyword for job information, the second keyword including second semantic feature information, second part of speech information, and second position information; using the first keyword and the second keyword to score the candidate resumes, obtaining the work experience matching score of the candidate resumes, and realizing human resource management. This method does not realize digital management of human resources, and relies solely on resume screening, and the management method is inefficient.

[0003] The traditional human resource management method is to use written records and other methods to manage human resources by the human resources department. However, with the improvement of informatization, the traditional human resource management method has problems such as slow information processing speed and low level of management refinement. At the same time, without using digital management methods, there is a lack of transparency and accuracy in employee recruitment, performance appraisal and other aspects, which brings problems to the company's human resource management. Summary of the invention

[0004] In response to the problems in the related technology, the present invention provides a method and system for digital management of human resource information based on big data to overcome the above-mentioned technical problems existing in the existing related technology.

[0005] In order to solve the above technical problems, the present invention is achieved through the following technical solutions: The present invention is a digital management method for human resource information based on big data, comprising the following steps: S1. Obtain an original human resources data set, generate an original human resources data matrix, perform principal component analysis on the generated original human resources data matrix to obtain a processed human resources data matrix, and then calculate employee execution capability values ​​to obtain an employee execution capability value set; S2. Classifying the employee execution capability value set based on the improved clustering algorithm, and introducing the particle swarm optimization algorithm to find the optimal initial clustering center to obtain the employee capability value distribution set, and then performing employee position distribution according to the employee capability value distribution set to obtain the employee position distribution result; S3, training the BP neural network to obtain a BP neural network model, obtaining a final human resources data set according to the employee position allocation results, the employee execution capability value set and the processed human resources data matrix, and performing employee performance prediction, and outputting the employee performance prediction value; S4. Conduct employee performance appraisal based on the employee performance prediction value, and feed back the employee performance appraisal results to the company's human resources department to realize digital management of human resources information.

[0006] The invention obtains original human resource data to generate an original human resource data matrix, uses principal component analysis to calculate the scores of human resource data indicators in the original human resource data matrix, obtains a processed human resource data matrix, assigns weights to the processed human resource data matrix, and calculates the execution ability value of each employee; this method can effectively and objectively and scientifically evaluate human resource data by screening out principal components while retaining most of the information of the original data, and converting linearly related indicators into linearly independent indicators; secondly, an improved clustering algorithm is used to classify the employee execution ability value set, and in the classification process, the optimal initial clustering center is found by introducing a particle swarm optimization algorithm, and positions are assigned to different employee execution ability values ​​after classification; the clustering algorithm is simple to execute and efficient, and the optimal initial clustering center is found by iteration, which can greatly reduce the time consumption of classification and avoid the problem of premature convergence; the BP neural network is then trained to obtain a BP neural network model, and the human resource data set is used to predict employee performance and output the employee performance prediction value; the neural network has a good effect on solving nonlinear mapping problems and is highly adaptable, and can well realize performance prediction; finally, employee performance appraisal is evaluated based on the employee performance prediction value to realize digital management of human resource information.

[0007] Preferably, the S1 comprises the following steps: S11, obtaining human resource data indicators, the human resource data indicators include the company's educational background distribution, employee age ratio, employee cost structure, etc., each human resource data indicator selects n human resource data to form an original human resource data set A={A1,A2,A3,...,A m}, where A m Represents the mth human resources data indicator, and generates the original human resources data matrix B as follows: ; Among them, b mnrepresents the nth human resource data of the mth human resource data indicator; The principal component analysis method is used to calculate the human resource data index scores in the original human resource data matrix to obtain a processed human resource data matrix. The specific steps are as follows: S111, select the first Human resource data indicators constitute the human resource data set to be processed ,in Indicates The nth human resource data of the human resource data indicator; calculating the mean value and covariance of the human resource data in the human resource data set to be processed, obtaining the mean value and covariance of the human resource data, and performing standardization processing on the human resource data set to be processed according to the mean value and covariance of the human resource data, the calculation formula is as follows: ; in, Indicates The standardized data of the nth human resource data of human resource data indicators, Indicates The average value of human resources data for human resources data indicators, Indicates The human resource data covariance of the human resource data indicators; After the human resources data set to be processed is standardized, a processed human resources data set is obtained. Other human resources data indicators in the original human resources data matrix are processed in sequence to obtain a processed human resources data matrix B1 as follows: ; in, The standardized data representing the nth human resource data of the mth human resource data indicator; S112, calculating the human resources correlation coefficient matrix according to the processed human resources data matrix, the calculation formula is as follows: ; Among them, B2 represents the human resources correlation coefficient matrix, represents the transpose of the processed human resource data matrix; Calculate the eigenvalues ​​and eigenvectors of the human resources correlation coefficient matrix to obtain the eigenvalue set C1={λ 1, λ2,λ3,...,λ m} and the set of eigenvectors C2={a1,a2,a3,...,a m}, where λ mrepresents the eigenvalue of the mth human resource data indicator in the human resource correlation coefficient matrix, a m Represents the eigenvector of the mth human resource data indicator in the human resource correlation coefficient matrix; selects the human resource data indicators corresponding to the eigenvalues ​​greater than 1 in the eigenvalue set to obtain the principal component human resource data indicators, The human resource contribution rate is calculated using the principal component human resource data indicators, and the contribution threshold is set as , selecting the principal component human resource data indicators corresponding to the human resource contribution rate being greater than the contribution threshold, and screening the eigenvectors corresponding to the principal component human resource data indicators in the eigenvector set to obtain the principal component eigenvector set; S113, according to the original human resources data matrix, the processed human resources data matrix and the principal component eigenvector set, the principal component score is calculated to obtain the human resources data index score, and the score threshold is set to ω. When the human resources data index score is less than the score threshold, the corresponding human resources data index is deleted, otherwise the corresponding human resources data index is saved, and the processed human resources data matrix B3 is obtained as follows: ; in, Indicates The nth human resource data of a human resource data indicator; S12, dividing the processed human resource data matrix according to employees, and assigning weights to the human resource data indicators in the processed human resource data matrix, setting the weight set to ,in Indicates The weight corresponding to each human resources data indicator is The calculation formula of employee execution capability value is as follows: ; in, Indicates the company’s Employee execution capability value, Indicates the company’s Employee's Human resource data of human resource data indicators; Obtain the execution capability values ​​of all employees in the company in turn to form a set of employee execution capability values .

[0008] The invention uses principal component analysis to calculate the scores of human resource data indicators in the original human resource data matrix to obtain a processed human resource data matrix. While retaining most of the information of the original data, the principal components are screened out and linearly related indicators are converted into linearly independent indicators, which can effectively and objectively and scientifically evaluate the human resource data. Weights are then assigned to the processed human resource data matrix to calculate the execution ability value of each employee.

[0009] Preferably, S2 comprises the following steps: S21. Classify the employee execution capability value set based on the improved clustering algorithm, and find the optimal initial clustering center by introducing the particle swarm optimization algorithm. The specific steps are as follows: S211, set the number of particles in the particle swarm within the search range to , the number of particles in the particle swarm is equal to the number of employee execution capability values ​​in the employee execution capability value set, the dimension of each particle in the particle swarm is p, and the dimension of the employee execution capability value in the corresponding employee execution capability value set is also p; k samples are selected from the employee execution capability value set as cluster centers, and the k cluster centers correspond to the initial positions of k particles; the particle positions are continuously updated within the search range, and the speed of the i-th particle in the particle swarm at time t+1 is set to , the position of the i-th particle in the particle swarm at time t is , then the position of the i-th particle in the particle swarm at time t+1 is ; S212, when the particles in the particle swarm are iterating, the particle position is updated by calculating the fitness function value of each particle. The specific steps are as follows: S2121, regarding the employee execution capability value as a sample point, selecting k sample points from the employee execution capability value set as initial cluster centers, calculating the Euclidean distances from all sample points in the employee execution capability value set to the initial cluster centers, and obtaining a Euclidean distance set; adding the sample point corresponding to the minimum Euclidean distance in the Euclidean distance set to the corresponding initial cluster center to form a cluster; then calculating the mean of the sample points in the cluster, and using the mean of the sample points as a new cluster center, repeating S2121 until the cluster center no longer changes, and obtaining a number of clusters; S2122. Calculate the distance from all sample points in the cluster to the cluster center. Set the number of cluster centers to d and the number of sample points in the cluster to e. Record the distance from the d1th cluster center to the e1th sample point as D(d1,e1). Update the particle position continuously according to the fitness function value. The fitness function value calculation formula of the e1th sample point is as follows: ; Where f(e1) represents the fitness function value of the e1th sample point, d1=1,2,3,...,d, e1=1,2,3,...,e; S213, compare the fitness function values ​​of all particles in the particle swarm at the current position, and take the particle corresponding to the current maximum fitness function value as the optimal particle; set the current iteration number and the maximum iteration number, and continuously update the particle speed and position with continuous iteration until the current iteration number reaches the maximum iteration number, stop iteration, and obtain the particle position corresponding to the maximum fitness function value, which is recorded as the final particle position, and the final particle position is the optimal initial clustering center; S22, selecting the optimal initial cluster center from the employee execution capability value set, using a clustering algorithm to divide the employee execution capability value set, and outputting a clustering result, wherein the clustering result includes a plurality of employee capability value clusters, which constitute an employee capability value distribution set ,in Indicates Employee capability value clusters; Set employee position capability value set ,in Indicates An employee position ability value is obtained, and the employee ability value cluster in the employee ability value allocation set is matched with the employee position ability value set. When the employee ability value cluster is smaller than the employee position ability value, the employee corresponding to the employee ability value cluster can be allocated to the employee position corresponding to the employee position ability value; otherwise, the employee corresponding to the employee ability value cluster cannot be allocated to the employee position corresponding to the employee position ability value, and the employee position allocation result is obtained.

[0010] The invention classifies the employee execution ability value sets by using an improved clustering algorithm. The clustering algorithm is simple to execute and highly efficient. During the classification process, a particle swarm optimization algorithm is introduced to find the optimal initial clustering center, which greatly reduces the classification time and avoids the problem of premature convergence, making it easier to assign positions to employees with different execution ability values ​​after classification.

[0011] Preferably, S3 comprises the following steps: S31, obtain the human resources data set of previous years, perform principal component analysis to obtain the processed human resources data set of previous years, then calculate the employee execution ability value set of previous years, and obtain the employee position allocation results of previous years; combine the processed human resources data set of previous years, the employee execution ability value set of previous years and the employee position allocation results of previous years to obtain the processed human resources data set of previous years, divide the human resources data sample set of previous years according to employees, and obtain the human resources data sample set of previous years C7={E1,E2,E3,...,E g}, where E grepresents the sample set of human resource data of the g-th employee in previous years; the sample set of human resource data of previous years is used to train the BP (back propagation) neural network to obtain the BP neural network prediction model. The specific steps are as follows: S311, dividing the sample set of human resources data from previous years into a sample training set and a sample test set, setting the number of nodes in the BP neural network input layer to , the number of hidden layer nodes is , the number of output layer nodes is 1, the sample training set is input into the BP neural network, and it is continuously iterated until the BP neural network converges to obtain a trained BP neural network; S312, input the sample test set into the trained BP neural network, and set the accuracy threshold to , when the accuracy of the output result is greater than the accuracy threshold, the BP neural network prediction model is obtained, otherwise the weight is adjusted until the accuracy of the output result is greater than the accuracy threshold; S32. Quantify the employee position allocation results, and combine the employee execution capability value set and the processed human resource data matrix to obtain the final human resource data set. Divide the final human resource data set according to employees and input it into the BP neural network prediction model, output the employee performance prediction value, and complete the employee performance prediction.

[0012] The invention classifies the employee execution ability value sets by using an improved clustering algorithm. The clustering algorithm is simple to execute and highly efficient. During the classification process, a particle swarm optimization algorithm is introduced to find the optimal initial clustering center, which greatly reduces the classification time and avoids the problem of premature convergence, making it easier to assign positions to employees with different execution ability values ​​after classification.

[0013] Preferably, the S4 comprises the following steps: S41. Setting performance threshold set ,in , and Indicates the performance threshold. When the employee's performance forecast value is greater than or equal to the performance threshold When the employee performance evaluation is excellent, the employee performance forecast value is less than the performance threshold and greater than or equal to the performance threshold When the employee performance evaluation is average, the employee performance forecast value is less than the performance threshold and greater than or equal to the performance threshold When the employee performance evaluation is qualified, the employee performance forecast value is less than the performance threshold At this time, the employee's performance appraisal is evaluated as unqualified; the employee's performance appraisal evaluation result is obtained; S42. Feedback the employee performance appraisal results to the company's human resources department. The company's human resources department conducts a human resources management assessment on the employees based on the employee performance appraisal results, optimizes management processes and strategies, and realizes digital management of human resources information.

[0014] The present invention also discloses a system for a digital management method of human resource information based on big data, which specifically includes: an original human resource data processing module, an employee position allocation module, an employee performance prediction module and an employee performance assessment module; The original human resource data processing module is used to perform principal component analysis on the original human resource data and calculate the employee execution capability value; The employee position allocation module is used to classify employee execution capability values ​​using an improved clustering algorithm and then allocate employee positions; The employee performance prediction module is used to train the BP neural network to obtain a BP neural network model and output the employee performance prediction value; The employee performance appraisal module is used to conduct employee performance appraisal based on employee performance prediction values.

[0015] The present invention has the following beneficial effects: 1. The present invention uses principal component analysis to calculate the scores of human resource data indicators in the original human resource data matrix to obtain a processed human resource data matrix. While retaining most of the information of the original data, the principal components are screened out and linearly related indicators are converted into linearly independent indicators, which can effectively and objectively and scientifically evaluate the human resource data.

[0016] 2. The invention classifies the employee execution capability value sets by using an improved clustering algorithm. The clustering algorithm is simple to execute and highly efficient, and can achieve accurate classification.

[0017] 3. The invention introduces a particle swarm optimization algorithm to find the optimal initial cluster center during the classification process, thereby greatly reducing the time consumption of classification, avoiding the problem of premature convergence, and facilitating job allocation according to the different execution ability values ​​of employees after classification.

[0018] 4. The invention obtains a BP neural network model by training the BP neural network. The BP neural network has good effect in solving nonlinear mapping problems and strong adaptability, outputs employee performance prediction values, and can well realize performance prediction.

[0019] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0021] Figure 1 A schematic diagram of the process of digital management of human resources information by a human resources information digital management system based on big data provided by the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0024] Example 1 Please refer to Figure 1 The present invention is a method for digital management of human resource information based on big data, comprising the following steps: S1. Obtain an original human resources data set, generate an original human resources data matrix, perform principal component analysis on the generated original human resources data matrix to obtain a processed human resources data matrix, and then calculate employee execution capability values ​​to obtain an employee execution capability value set; The S1 comprises the following steps: S11, obtaining human resource data indicators, the human resource data indicators include the company's educational background distribution, employee age ratio, employee cost structure, etc., each human resource data indicator selects n human resource data to form an original human resource data set A={A1,A2,A3,...,A m}, where A m Represents the mth human resources data indicator, and generates the original human resources data matrix B as follows: ; in, represents the nth human resource data of the mth human resource data indicator; The principal component analysis method is used to calculate the human resource data index scores in the original human resource data matrix to obtain a processed human resource data matrix. The specific steps are as follows: S111, select the first Human resource data indicators constitute the human resource data set to be processed ,in Indicates The nth human resource data of the human resource data indicator; calculating the mean value and covariance of the human resource data in the human resource data set to be processed, obtaining the mean value and covariance of the human resource data, and performing standardization processing on the human resource data set to be processed according to the mean value and covariance of the human resource data, the calculation formula is as follows: ; in, Indicates The standardized data of the nth human resource data of human resource data indicators, Indicates The average value of human resources data for human resources data indicators, Indicates The human resource data covariance of the human resource data indicators; After the human resources data set to be processed is standardized, a processed human resources data set is obtained. Other human resources data indicators in the original human resources data matrix are processed in sequence to obtain a processed human resources data matrix B1 as follows: ; in, The standardized data representing the nth human resource data of the mth human resource data indicator; S112, calculating the human resources correlation coefficient matrix according to the processed human resources data matrix, the calculation formula is as follows: ; Among them, B2 represents the human resources correlation coefficient matrix, represents the transpose of the processed human resource data matrix; Calculate the eigenvalues ​​and eigenvectors of the human resources correlation coefficient matrix to obtain the eigenvalue set C1={λ 1, λ2,λ3,...,λ m} and the set of eigenvectors C2={a1,a2,a3,...,a m}, where λ m represents the eigenvalue of the mth human resource data indicator in the human resource correlation coefficient matrix, am Represents the eigenvector of the mth human resource data indicator in the human resource correlation coefficient matrix; selects the human resource data indicators corresponding to the eigenvalues ​​greater than 1 in the eigenvalue set to obtain the principal component human resource data indicators, The human resource contribution rate is calculated using the principal component human resource data indicators, and the contribution threshold is set as , selecting the principal component human resource data indicators corresponding to the human resource contribution rate being greater than the contribution threshold, and screening the eigenvectors corresponding to the principal component human resource data indicators in the eigenvector set to obtain the principal component eigenvector set; S113, according to the original human resources data matrix, the processed human resources data matrix and the principal component eigenvector set, the principal component score is calculated to obtain the human resources data index score, and the score threshold is set to ω. When the human resources data index score is less than the score threshold, the corresponding human resources data index is deleted, otherwise the corresponding human resources data index is saved, and the processed human resources data matrix B3 is obtained as follows: ; in, Indicates The nth human resource data of a human resource data indicator; S12, dividing the processed human resource data matrix according to employees, and assigning weights to the human resource data indicators in the processed human resource data matrix, setting the weight set to ,in Indicates The weight corresponding to each human resources data indicator is The calculation formula of employee execution capability value is as follows: ; in, Indicates the company’s Employee execution capability value, Indicates the company’s Employee's Human resource data of human resource data indicators; Obtain the execution capability values ​​of all employees in the company in turn to form a set of employee execution capability values ; S2. Classifying the employee execution capability value set based on the improved clustering algorithm, and introducing the particle swarm optimization algorithm to find the optimal initial clustering center to obtain the employee capability value distribution set, and then performing employee position distribution according to the employee capability value distribution set to obtain the employee position distribution result; The S2 comprises the following steps: S21. Classify the employee execution capability value set based on the improved clustering algorithm, and find the optimal initial clustering center by introducing the particle swarm optimization algorithm. The specific steps are as follows: S211, set the number of particles in the particle swarm within the search range to , the number of particles in the particle swarm is equal to the number of employee execution capability values ​​in the employee execution capability value set, the dimension of each particle in the particle swarm is p, and the dimension of the employee execution capability value in the corresponding employee execution capability value set is also p; k samples are selected from the employee execution capability value set as cluster centers, and the k cluster centers correspond to the initial positions of k particles; the particle positions are continuously updated within the search range, and the speed of the i-th particle in the particle swarm at time t+1 is set to , the position of the i-th particle in the particle swarm at time t is , then the position of the i-th particle in the particle swarm at time t+1 is ; S212, when the particles in the particle swarm are iterating, the particle position is updated by calculating the fitness function value of each particle. The specific steps are as follows: S2121, regarding the employee execution capability value as a sample point, selecting k sample points from the employee execution capability value set as initial cluster centers, calculating the Euclidean distances from all sample points in the employee execution capability value set to the initial cluster centers, and obtaining a Euclidean distance set; adding the sample point corresponding to the minimum Euclidean distance in the Euclidean distance set to the corresponding initial cluster center to form a cluster; then calculating the mean of the sample points in the cluster, and using the mean of the sample points as a new cluster center, repeating S2121 until the cluster center no longer changes, and obtaining a number of clusters; S2122. Calculate the distance from all sample points in the cluster to the cluster center. Set the number of cluster centers to d and the number of sample points in the cluster to e. Record the distance from the d1th cluster center to the e1th sample point as D(d1,e1). Update the particle position continuously according to the fitness function value. The fitness function value calculation formula of the e1th sample point is as follows: ; Where f(e1) represents the fitness function value of the e1th sample point, d1=1,2,3,...,d, e1=1,2,3,...,e; S213, compare the fitness function values ​​of all particles in the particle swarm at the current position, and take the particle corresponding to the current maximum fitness function value as the optimal particle; set the current iteration number and the maximum iteration number, and continuously update the particle speed and position with continuous iteration until the current iteration number reaches the maximum iteration number, stop iteration, and obtain the particle position corresponding to the maximum fitness function value, which is recorded as the final particle position, and the final particle position is the optimal initial clustering center; S22, selecting the optimal initial cluster center from the employee execution capability value set, using a clustering algorithm to divide the employee execution capability value set, and outputting a clustering result, wherein the clustering result includes a plurality of employee capability value clusters, which constitute an employee capability value distribution set ,in Indicates Employee capability value clusters; Set employee position capability value set ,in Indicates employee position competency values, matching the employee competency value cluster in the employee competency value allocation set with the employee position competency value set, when the employee competency value cluster is smaller than the employee position competency value, the employee corresponding to the employee competency value cluster can be allocated to the employee position corresponding to the employee position competency value; otherwise, the employee corresponding to the employee competency value cluster cannot be allocated to the employee position corresponding to the employee position competency value, and obtaining the employee position allocation result; S3, training the BP neural network to obtain a BP neural network model, obtaining a final human resources data set according to the employee position allocation results, the employee execution capability value set and the processed human resources data matrix, and performing employee performance prediction, and outputting the employee performance prediction value; The S3 comprises the following steps: S31, obtain the human resources data set of previous years, perform principal component analysis to obtain the processed human resources data set of previous years, then calculate the employee execution ability value set of previous years, and obtain the employee position allocation results of previous years; combine the processed human resources data set of previous years, the employee execution ability value set of previous years and the employee position allocation results of previous years to obtain the processed human resources data set of previous years, divide the human resources data sample set of previous years according to employees, and obtain the human resources data sample set of previous years C7={E1,E2,E3,...,E g}, where E g represents the sample set of human resource data of the g-th employee in previous years; the sample set of human resource data of previous years is used to train the BP neural network to obtain the BP neural network prediction model. The specific steps are as follows: S311, dividing the sample set of human resources data from previous years into a sample training set and a sample test set, setting the number of nodes in the BP neural network input layer to , the number of hidden layer nodes is , the number of output layer nodes is 1, the sample training set is input into the BP neural network, and it is continuously iterated until the BP neural network converges to obtain a trained BP neural network; S312, input the sample test set into the trained BP neural network, and set the accuracy threshold to , when the accuracy of the output result is greater than the accuracy threshold, the BP neural network prediction model is obtained, otherwise the weight is adjusted until the accuracy of the output result is greater than the accuracy threshold; S32, quantifying the employee position allocation result, and combining the employee execution capability value set and the processed human resource data matrix to obtain a final human resource data set, dividing the final human resource data set according to employees and inputting it into the BP neural network prediction model, outputting the employee performance prediction value, and completing the employee performance prediction; S4. Conduct employee performance appraisal and evaluation based on the employee performance prediction value, and feed back the employee performance appraisal and evaluation results to the company's human resources department to realize digital management of human resources information; The S4 comprises the following steps: S41. Setting performance threshold set ,in , and Indicates the performance threshold. When the employee's performance forecast value is greater than or equal to the performance threshold When the employee performance evaluation is excellent, the employee performance forecast value is less than the performance threshold and greater than or equal to the performance threshold When the employee performance evaluation is average, the employee performance forecast value is less than the performance threshold and greater than or equal to the performance threshold When the employee performance evaluation is qualified, the employee performance forecast value is less than the performance threshold At this time, the employee's performance appraisal is evaluated as unqualified; the employee's performance appraisal evaluation result is obtained; S42. Feedback the employee performance appraisal results to the company's human resources department. The company's human resources department conducts a human resources management assessment on the employees based on the employee performance appraisal results, optimizes management processes and strategies, and realizes digital management of human resources information.

[0025] Example 2 The present invention also discloses a system for a digital management method of human resource information based on big data, which specifically includes: an original human resource data processing module, an employee position allocation module, an employee performance prediction module and an employee performance assessment module; The original human resource data processing module is used to perform principal component analysis on the original human resource data and calculate the employee execution capability value; The employee position allocation module is used to classify employee execution capability values ​​using an improved clustering algorithm and then allocate employee positions; The employee performance prediction module is used to train the BP neural network to obtain a BP neural network model and output the employee performance prediction value; The employee performance appraisal module is used to conduct employee performance appraisal based on employee performance prediction values.

[0026] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0027] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. A digital management method for human resources information based on big data, characterized in that: The steps include: S1. Obtain an original human resources data set, generate an original human resources data matrix, perform principal component analysis on the generated original human resources data matrix to obtain a processed human resources data matrix, and then calculate employee execution capability values ​​to obtain an employee execution capability value set; S2. Classifying the employee execution capability value set based on an improved clustering algorithm, and finding the optimal initial clustering center to obtain an employee capability value distribution set, and then performing employee position distribution according to the employee capability value distribution set to obtain an employee position distribution result; S3, training the BP neural network to obtain a BP neural network model, obtaining a final human resources data set according to the employee position allocation results, the employee execution capability value set and the processed human resources data matrix, and performing employee performance prediction, and outputting the employee performance prediction value; S4. Conduct employee performance appraisal based on the employee performance prediction value, and feed back the employee performance appraisal results to the company's human resources department to realize digital management of human resources information.

2. According to the method for digital management of human resources information based on big data according to claim 1, it is characterized in that: The S1 comprises the following steps: S11, obtaining human resource data indicators to form an original human resource data set and generate an original human resource data matrix; using principal component analysis to calculate the scores of the human resource data indicators in the original human resource data matrix to obtain a processed human resource data matrix; S12, dividing the processed human resource data matrix according to employees, assigning weights to human resource data indicators in the processed human resource data matrix, calculating employee execution capability values, and obtaining an employee execution capability value set.

3. According to claim 2, a method for digital management of human resources information based on big data is characterized in that: The S11 comprises the following steps: S111, selecting a human resources data set to be processed from the original human resources data matrix, calculating the mean value and covariance of the human resources data in the human resources data set to be processed, and then performing standardization processing on the human resources data set to be processed to obtain a processed human resources data matrix; S112, generating a human resources correlation coefficient matrix according to the processed human resources data matrix, Calculating the eigenvalues ​​and eigenvectors of the human resources correlation coefficient matrix, and screening the eigenvalues ​​of the human resources correlation coefficient matrix to obtain a set of principal component eigenvectors; S113, calculating the principal component score according to the original human resource data matrix, the processed human resource data matrix and the principal component eigenvector set to obtain the human resource data index score, and setting a threshold for screening to obtain the processed human resource data matrix.

4. According to the big data-based digital management method for human resources information of claim 3, it is characterized in that: The S2 comprises the following steps: S21, classifying the employee execution capability value set based on an improved clustering algorithm, and finding the optimal initial clustering center by introducing a particle swarm optimization algorithm; S22. Select the optimal initial cluster center from the employee execution capability value set, use a clustering algorithm to divide the employee execution capability value set, output the clustering results, form an employee capability value allocation set, and then perform employee position allocation to obtain an employee position allocation result.

5. The method for digital management of human resource information based on big data according to claim 4 is characterized in that: The S21 comprises the following steps: S211, set the number of particles in the particle swarm within the search range to , the number of particles in the particle swarm is equal to the number of employee execution capability values ​​in the employee execution capability value set, the dimension of each particle in the particle swarm is p, and the dimension of the employee execution capability value in the corresponding employee execution capability value set is also p; k samples are selected from the employee execution capability value set as cluster centers, and the k cluster centers correspond to the initial positions of k particles; the particle positions are continuously updated within the search range, and the speed of the i-th particle in the particle swarm at time t+1 is set to , the position of the i-th particle in the particle swarm at time t is , then the position of the i-th particle in the particle swarm at time t+1 is ; S212, when the particles in the particle swarm are iterating, the particle positions are updated by calculating the fitness function values; S213, comparing the fitness function values ​​of all particles in the particle swarm at the current position, and taking the particle corresponding to the current maximum fitness function value as the optimal particle; setting the current number of iterations and the maximum number of iterations, with continuous iterations, the particle speed and position are continuously updated until the current number of iterations reaches the maximum number of iterations, and then stopping the iteration to obtain the particle position corresponding to the maximum fitness function value, which is recorded as the final particle position, and the final particle position is the optimal initial clustering center.

6. The method for digital management of human resource information based on big data according to claim 5 is characterized in that: The S212 comprises the following steps: S2121, regarding the employee execution capability value as a sample point, selecting k sample points from the employee execution capability value set as initial cluster centers, calculating the Euclidean distances from all sample points in the employee execution capability value set to the initial cluster centers, and obtaining a Euclidean distance set; adding the sample point corresponding to the minimum Euclidean distance in the Euclidean distance set to the corresponding initial cluster center to form a cluster; then calculating the mean of the sample points in the cluster, and using the mean of the sample points as a new cluster center, repeating S2121 until the cluster center no longer changes, and obtaining a number of clusters; S2122. Calculate the distance from all sample points in the cluster to the cluster center. Set the number of cluster centers to d and the number of sample points in the cluster to e. Record the distance from the d1th cluster center to the e1th sample point as D(d1,e1). Update the particle position continuously according to the fitness function value. The fitness function value calculation formula of the e1th sample point is as follows: ； Among them, f(e1) represents the fitness function value of the e1th sample point, d1=1,2,3,...,d, e1=1,2,3,...,e.

7. The method for digital management of human resource information based on big data according to claim 6 is characterized in that: The S3 comprises the following steps: S31, obtaining a human resources data set from previous years, performing principal component analysis to obtain a processed human resources data set from previous years, and then calculating a set of employee execution capability values ​​from previous years to obtain employee position allocation results from previous years; combining the processed human resources data set from previous years, the set of employee execution capability values ​​from previous years, and the employee position allocation results from previous years to obtain a processed human resources data set from previous years, dividing the human resources data sample set from previous years according to employees to obtain a human resources data sample set from previous years; using the human resources data sample set from previous years to train a BP neural network to obtain a BP neural network prediction model; S32. Quantify the employee position allocation results, and combine the employee execution capability value set and the processed human resource data matrix to obtain the final human resource data set. Divide the final human resource data set according to employees and input it into the BP neural network prediction model, output the employee performance prediction value, and complete the employee performance prediction.

8. According to the method of digital management of human resource information based on big data according to claim 7, it is characterized in that: The method of using the human resources data sample set of previous years to train the BP neural network to obtain the BP neural network prediction model includes the following steps: The human resources data sample set of previous years is divided into a sample training set and a sample test set, and the number of nodes in the BP neural network input layer is set to , the number of hidden layer nodes is , the number of output layer nodes is 1, the sample training set is input into the BP neural network, and it is continuously iterated until the BP neural network converges to obtain a trained BP neural network; The sample test set is input into the trained BP neural network, and the accuracy threshold is set to , when the accuracy of the output result is greater than the accuracy threshold, the BP neural network prediction model is obtained, otherwise the weights are adjusted until the accuracy of the output result is greater than the accuracy threshold.

9. The method for digital management of human resource information based on big data according to claim 8 is characterized in that: The S4 comprises the following steps: S41. Set a performance threshold, conduct employee performance appraisal and evaluation by comparing the employee performance prediction value with the performance threshold, and obtain the employee performance appraisal and evaluation result; S42. Feedback the employee performance appraisal results to the company's human resources department. The company's human resources department conducts a human resources management assessment on the employees based on the employee performance appraisal results, optimizes management processes and strategies, and realizes digital management of human resources information.

10. A system for implementing the method for digital management of human resource information based on big data as claimed in any one of claims 1 to 9, characterized in that: Specifically include: Original human resource data processing module, employee position allocation module, employee performance prediction module and employee performance appraisal module; The original human resource data processing module is used to perform principal component analysis on the original human resource data and calculate the employee execution capability value; The employee position allocation module is used to classify employee execution capability values ​​using an improved clustering algorithm and then allocate employee positions; The employee performance prediction module is used to train the BP neural network to obtain a BP neural network model and output the employee performance prediction value; The employee performance appraisal module is used to conduct employee performance appraisal based on employee performance prediction values.

Images