An intelligent optimization method and system for human resource management based on data mining

Abstract

The present invention discloses an intelligent optimization method and system for human resource management based on data mining, which relates to the field of human resource management optimization. The present invention optimizes the management of employee information in the human resources of an enterprise from two aspects: the selection of the storage method of employee information in the human resources department and the reduction of the number of times of employee information loss. Among them, by constructing a final storage method mapping equation, the storage method prediction data of the current enterprise is obtained, and then the storage method prediction data is used to correct and adjust the storage method of employee information in the current enterprise. By constructing a final mapping model for the number of times of employee information loss, the prediction data of the number of times of employee information loss of the current enterprise is obtained, and then the storage cluster of employees in the current enterprise is adjusted subsequently to ensure that the number of times of employee information loss is minimized, thereby reducing the risk of employee information loss in the enterprise and ensuring the information security and integrity of the enterprise.

Description

Technical Field

[0001] The present invention belongs to the field of human resource management optimization. Specifically, it particularly relates to an intelligent optimization method and system for human resource management based on data mining. Background Art

[0002] A Chinese patent with the publication number CN111401841B discloses a human resource management system and its method. Through the information access unit and the salary and welfare module, information sharing management is implemented. When an applicant needs to understand the enterprise's salary and welfare, they can submit an access application to the information access unit and directly access the detailed information of the salary and welfare module after obtaining the approval of the system administrator.

[0003] In the human resources department of an enterprise, the management of enterprise employee information is an important task. Employee information is divided into sensitive information and non-sensitive information. Therefore, it is necessary to select the storage method for these two types of information, either separate storage or centralized storage. If the appropriate storage method is not selected, it will not only lead to security risks for employees' information but also reduce the access efficiency of enterprise employee information. Secondly, if the characteristic parameters of the cluster for storing employee information cannot be set appropriately, due to various reasons such as cluster downtime or power failure, the risk of loss of stored employee information will increase. Summary of the Invention

[0004] In view of the problems in the related art, the present invention proposes an intelligent optimization method and system for human resource management based on data mining to overcome the above-mentioned technical problems existing in the existing related technologies.

[0005] To solve the above technical problems, the present invention is realized through the following technical solutions:

[0006] The present invention provides an intelligent optimization method for human resource management based on data mining, including the following steps:

[0007] S1. Collect employee data, storage method data, storage method influencing factor data, characteristic data of the cluster for storing employee information, and the number of times of employee information loss within the statistical period saved by multiple enterprises to obtain an enterprise employee data matrix, an enterprise storage method data set, an enterprise storage method influencing factor data matrix, an enterprise cluster characteristic data matrix, and an enterprise employee information loss number data set.

[0008] S2. Use the enterprise employee data matrix, the enterprise storage method data set, and the enterprise storage method influencing factor data matrix to construct a final storage method mapping equation; use the enterprise cluster characteristic data matrix and the enterprise employee information loss number data set to construct a final employee information loss number mapping model.

[0009] S3. Map the employee data and storage method influencing factor data of the current enterprise using the final storage method mapping equation to obtain the current storage method data; map the characteristic data of the employee information cluster stored by the current enterprise using the final employee information loss times mapping model to obtain the current enterprise employee information loss times data.

[0010] S4. Store the employee information of the current enterprise according to the current storage method data; optimize the current cluster characteristic data set according to the current enterprise employee information loss times data to obtain the current final cluster characteristic data set.

[0011] By collecting the employee data, storage method data, storage method influencing factor data, characteristic data of the employee information cluster stored, and the number of employee information losses within the statistical period of multiple enterprises, it provides data support for the subsequent construction of the final storage method mapping equation and the final employee information loss times mapping model; by constructing the final storage method mapping equation, it provides a mapping tool for the subsequent mapping of the employee data and storage method influencing factor data of the currently collected enterprise, thereby obtaining the storage method prediction data of the current enterprise, and then using this storage method prediction data to correct and adjust the storage method of the employee information of the current enterprise; by constructing the final employee information loss times mapping model, it provides a mapping tool for the subsequent mapping of the characteristic data of the employee information cluster stored by the current enterprise, thereby obtaining the employee information loss times prediction data of the current enterprise, and then for the subsequent adjustment of the employee storage cluster of the current enterprise to ensure that the number of employee information losses is minimized, thereby reducing the risk of enterprise employee information loss and ensuring the information security and integrity of the enterprise. END

[0012] Preferably, set multiple ordinary information types and sensitive information types of employees to obtain the employee ordinary information type set and the employee sensitive information type set , a 1i , a 2i respectively represent the i-th type of ordinary employee information and sensitive employee information set, , respectively represent the total number of the set ordinary employee information types and sensitive employee information types.

[0013] Set the employee information collection type set , the employee information storage method set and the storage method influencing factor type set ; , , respectively represent the number of ordinary information types in the employee information, the number of sensitive information types in the employee information, and the number of employees; , respectively represent that the general information and sensitive information of employees are stored separately and the general information and sensitive information of employees are stored centrally, which are represented by 0 and 1 respectively; represents setting the influencing factor type of the i-th storage method, represents the total number of the influencing factor types of the set storage methods.

[0014] S12. Cooperate with the set of general employee information types and the set of sensitive employee information types, and collect the employee data, storage method data, and storage method influencing factor data saved by multiple enterprises according to the employee information collection type set, the employee information storage method set, and the storage method influencing factor type set, to obtain the enterprise employee data matrix , the enterprise storage method data set and the enterprise storage method influencing factor data matrix ; represents the storage method data of the employee data of the i-th enterprise collected, and b represents the total number of enterprises collected; , are respectively as follows,

[0015] ; ;

[0016] Among them, , , respectively represent the quantity data of the general information types, the quantity data of the sensitive information types, and the employee quantity data in the employee information of the i-th enterprise collected; represents the storage method influencing factor data of the j-th type in the i-th enterprise collected.

[0017] S13. Set the set of employee information storage cluster feature types , represents the i-th employee information storage cluster feature type in the set of the set employee information storage cluster feature types, represents the total number of the employee information storage cluster feature types in the set of the set employee information storage cluster feature types; the set of the employee information storage cluster feature types includes the number of servers, performance level, cost performance level, scalability level, availability level, load balancing ability level, error recovery ability level, transparency level, manageability level, and programmability level of the employee information storage cluster.

[0018] Set the statistical period; collect the feature data of the employee information storage clusters of multiple enterprises and the number of times of employee information loss during the statistical period according to the set of the employee information storage cluster feature types, to obtain the enterprise cluster feature data matrix and the dataset of the number of times of enterprise employee information loss , represents the data of the number of times of employee information loss of the i-th enterprise collected; as follows,

[0019] ;

[0020] Among them, represents the characteristic data of the employee information storage cluster of the j-th type of the i-th enterprise collected.

[0021] The set of ordinary employee information types includes personal basic information, educational background, work experience, job title and professional title, health status, emergency contacts, family member information, training records, etc.; the set of sensitive employee information types includes biometric information such as fingerprints, facial recognition, iris scans, etc. for identifying personal identity, religious belief information, medical and health information, financial account information, whereabouts trajectory information, genetic information, blood type, and personal property information, etc.; since the proportion of sensitive information types in the employee information collected by the enterprise human resources department and the employee information data will have a greater impact on the enterprise's decision-making on the storage method of employee information, for example, separating sensitive information from ordinary information for storage will provide better protection for the security of sensitive information, but will additionally increase the storage cost and reduce the efficiency of accessing employee information; while storing sensitive information and ordinary information together increases the security risk of employee sensitive information, but reduces the storage cost and improves the efficiency of accessing employee information; the set of storage method influencing factor types includes compliance, user experience, cost, technical ability, and internal policies and culture, etc.

[0022] Preferably, the S2 includes the following steps:

[0023] S21. Construct an initial storage method mapping equation; as follows,

[0024] ;

[0025] Among them, represents the dependent variable of the initial storage method mapping equation, representing the employee information storage method data; represents the i-th independent variable of the initial storage method mapping equation, representing the i-th type of enterprise employee data, represents the independent variable coefficient of; represents the (i + 3)-th independent variable of the initial storage method mapping equation, representing the data of the i-th type of enterprise employee information storage method influencing factor, represents the independent variable coefficient of; represents the bias of the initial storage method mapping equation, and ceil represents the rounding function.

[0026] S22. Optimize the initial storage mode mapping equation by using the enterprise employee data matrix, the enterprise storage mode data set, and the enterprise storage mode influencing factor data matrix to obtain the final storage mode mapping equation.

[0027] S23. Construct an initial mapping model for the number of times of employee information loss; train and test the initial mapping model for the number of times of employee information loss by using the enterprise cluster feature data matrix and the enterprise employee information loss number data set; after the training and testing are completed, obtain the final mapping model for the number of times of employee information loss.

[0028] By optimizing the independent variable coefficients and bias terms in the initial storage mode mapping equation by using the enterprise employee data matrix, the enterprise storage mode data set, and the enterprise storage mode influencing factor data matrix, the obtained final storage mode mapping equation is more accurate for the mapping between enterprise employee data, enterprise storage mode influencing factor data, and enterprise storage mode; by training and testing the initial mapping model for the number of times of employee information loss, the obtained final mapping model for the number of times of employee information loss is more accurate for the mapping between enterprise cluster feature data and enterprise employee information loss number data.

[0029] Preferably, S22 includes the following steps:

[0030] S221. Input each row of data in the enterprise employee data matrix and the enterprise storage mode influencing factor data matrix into the initial storage mode mapping equation for mapping to obtain an initial storage mode mapping data set , c i represents the storage mode mapping data obtained by inputting the i-th row of data in the enterprise employee data matrix and the enterprise storage mode influencing factor data matrix into the initial storage mode mapping equation for mapping.

[0031] S222. When there is a difference between the storage mode mapping data in the initial storage mode mapping data set and the corresponding enterprise storage mode data in the enterprise storage mode data set, optimize the initial storage mode mapping equation until there is no difference between the storage mode mapping data in the initial storage mode mapping data set and the corresponding enterprise storage mode data in the enterprise storage mode data set, and obtain the final storage mode mapping equation; otherwise, there is no need to optimize the initial storage mode mapping equation, and use the initial storage mode mapping equation as the final storage mode mapping equation.

[0032] By comparing the data mapped by the initial storage method with the data of the enterprise storage method obtained from actual collection, when there is inconsistent data with the actual data, it indicates that the mapping accuracy rate of the mapping equation of the initial storage method does not meet the requirements, and it is necessary to adjust and optimize the independent variable coefficients and bias amounts of the mapping equation of the initial storage method.

[0033] Preferably, the S3 includes the following steps:

[0034] S31. According to the set of general employee information types and the set of sensitive employee information types, and according to the set of employee information collection types and the set of storage method influencing factor types, collect the employee data and storage method influencing factor data of the enterprise that currently needs to optimize human resource management, and obtain the current employee data set and the current storage method influencing factor data set ; e 11 、e 12 、e 13 respectively represent the quantity data of general information types in the employee information, the quantity data of sensitive information types in the employee information, and the quantity data of employees in the enterprise that currently needs to optimize human resource management; e 3i represents the data of the i-th type of storage method influencing factor of the enterprise that currently needs to optimize human resource management.

[0035] S32. According to the set of storage employee information cluster characteristic types, collect the characteristic data of the storage employee information cluster of the enterprise that currently needs to optimize human resource management, and obtain the current cluster characteristic data set ,e 4i represents the characteristic data of the i-th type of employee information storage cluster of the enterprise that needs to optimize human resource management before.

[0036] S33. Input the current employee data set, the current storage method data, and the current storage method influencing factor data set into the final storage method mapping equation for mapping to obtain the current storage method data e2;

[0037] Input the current cluster characteristic data set into the final employee information loss times mapping model for mapping to obtain the current enterprise employee information loss times data e5;

[0038] By mapping the employee data, storage method data, and storage method influencing factor data of the current enterprise, the corresponding storage method data can be obtained. This storage method data comprehensively considers various requirements, including the number of employees, the proportion of sensitive information, and storage costs, etc. Thus, this storage method data can be used as a reference method for storing the corresponding employee information of the current enterprise in the future; By mapping the characteristic data of the employee information storage cluster of the current enterprise, the corresponding number of employee information losses can be obtained, providing a basis for determining whether the cluster needs to be adjusted in the future.

[0039] Preferably, S4 includes the following steps:

[0040] S41. Store the employee information of the current enterprise according to the current storage method data e2.

[0041] S42. Set the information loss times threshold within the statistical period; when the current enterprise employee information loss times data e5 is greater than or equal to the information loss times threshold, adjust the current cluster characteristic data set until the current enterprise employee information loss times data e5 is less than the information loss times threshold, and obtain the current final cluster characteristic data set; otherwise, there is no need to adjust the current cluster characteristic data set, and use the current cluster characteristic data set as the current final cluster characteristic data set;

[0042] By setting the information loss times threshold, a quantitative determination basis is provided for determining whether the current enterprise employee information loss times data meets the requirements, making the determination more objective and accurate.

[0043] Preferably, the adjustment of the current cluster characteristic data set in S42 until the current enterprise employee information loss times data e5 is less than the information loss times threshold to obtain the current final cluster characteristic data set includes the following steps:

[0044] S421. Construct a second dragonfly population , represents the i-th dragonfly in the second dragonfly population, represents the scale of the second dragonfly population; set the maximum number of iterations of the second dragonfly population as and the current number of iterations as , which are respectively denoted as the second maximum number of iterations and the second current number of iterations; the search space dimension of the second dragonfly population is .

[0045] S422. Set the value range of the data corresponding to each employee information storage cluster characteristic type in the employee information storage cluster characteristic type set, and obtain the cluster characteristic value range set d3; as follows,

[0046] ;

[0047] Among them, and respectively represent the lower limit and the upper limit of the value of the data corresponding to the i-th employee information storage cluster feature type in the set of employee information storage cluster feature types;

[0048] Set the initial position of each dragonfly in the second dragonfly population according to the set of cluster feature value intervals to obtain the second initial position matrix ; as follows,

[0049] ;

[0050] Among them, represents the component of the initial position of the k-th dragonfly in the second dragonfly population in the i-th dimension; the calculation formula is as follows,

[0051] ;

[0052] In the formula, rand 2ki represents a random number between 0 and 1 generated for ;

[0053] S423. Construct the fitness function of the second dragonfly population according to the current enterprise employee information loss times data e5 ; as follows,

[0054]

[0055] S424. Start iteration. Before iteration, set the second current iteration count to 1; in the first round of iteration, use the fitness function of the second dragonfly population and cooperate with the final employee information loss times mapping model to calculate the fitness value of the initial position of each dragonfly in the second initial position matrix to obtain the third fitness value set; take the maximum fitness value in the third fitness value set and the corresponding initial position of the dragonfly as the third global best fitness and the third global best position respectively; update the initial position of each dragonfly in the second initial position matrix according to the third global best fitness and the third global best position; after the update is completed, increment the second current iteration count by 1 and enter the next round of iteration;

[0056] In each subsequent round of iteration, use the fitness function of the second dragonfly population And cooperate with the final employee information loss times mapping model to calculate the fitness value of the position of each dragonfly in the second dragonfly population updated in the previous iteration process, obtaining the fourth fitness value set; take the maximum fitness value in the fourth fitness value set and the position of the corresponding dragonfly as the fourth global best fitness and the fourth global best position respectively; update the position of each dragonfly in the second dragonfly population updated in the previous iteration process again according to the fourth global best fitness and the fourth global best position; after the update is completed, add 1 to the second current iteration number and enter the next iteration.

[0057] S425. When holds, stop the iteration to obtain the second final global best position; otherwise, continue the iteration until holds; take the second final global best position as the optimized current cluster feature data set; input the optimized current cluster feature data set into the final employee information loss times mapping model for mapping to obtain the optimized current enterprise employee information loss times data;

[0058] When the optimized current enterprise employee information loss times data is less than the information loss times threshold, take the optimized current cluster feature data set as the current final cluster feature data set; otherwise, return to S424 to continue the iteration until the optimized current enterprise employee information loss times data is less than the information loss times threshold;

[0059] By using the dragonfly optimization algorithm to simultaneously perform multiple iterative optimizations on the characteristic data of the employee information storage cluster of the current enterprise, and using the employee information loss times mapped by the final employee information loss times mapping model as the fitness function; therefore, as the iteration progresses, the employee information loss times corresponding to the characteristic data of the employee information storage cluster obtained by the iteration become less and less, indicating that the current iteration direction is correct and finally meets the requirements for storing employee information.

[0060] An intelligent optimization system for human resource management based on data mining, including an employee information type setting module, an enterprise employee information storage method data collection module, an enterprise employee information storage method data collection module, a storage method mapping equation construction module, an employee information loss times mapping model construction module, a current enterprise employee storage data collection module, a first mapping module, a second mapping module, a storage method management module, and a cluster feature optimization module.

[0061] The present invention has the following beneficial effects:

[0062] 1. The present invention optimizes the management of employee information in enterprise human resources from two aspects: the selection of the storage method of employee information in the human resources department and the reduction of the number of times of employee information loss. Among them, by constructing a final storage method mapping equation, obtaining the storage method prediction data of the current enterprise, and then using the storage method prediction data to correct and adjust the storage method of employee information in the current enterprise; by constructing a final mapping model of the number of times of employee information loss, obtaining the prediction data of the number of times of employee information loss in the current enterprise, and then adjusting the current enterprise's employee storage cluster subsequently to ensure that the number of times of employee information loss is minimized, thereby reducing the risk of employee information loss in the enterprise and ensuring the information security and integrity of the enterprise;

[0063] 2. In the present invention, by mapping the characteristic data of the employee information storage cluster of the current enterprise, the corresponding number of times of employee information loss is obtained, providing a judgment basis for determining whether the cluster needs to be adjusted subsequently;

[0064] 3. In the present invention, the dragonfly optimization algorithm is used to perform multiple iterative optimizations on each characteristic data of the employee information storage cluster of the current enterprise simultaneously, and the number of times of employee information loss mapped by the final mapping model of the number of times of employee information loss is used as the fitness function; Therefore, as the iteration progresses, the number of times of employee information loss corresponding to each characteristic data of the iteratively obtained employee information storage cluster becomes less and less, indicating that the current iteration direction is correct and finally meets the requirements for storing employee information.

[0065] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0067] Figure 1 It is a schematic flow chart of an intelligent optimization method for human resource management based on data mining according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] The technical solutions in the embodiments of the invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the invention. Obviously, the described embodiments are only some embodiments of the invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the invention without creative efforts fall within the scope of protection of the invention.

[0069] In the description of the present invention, it should be understood that the terms "open hole", "upper", "lower", "top", "middle", "inner", etc. indicating the orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, rather than indicating or implying that the components or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention.

[0070] Example 1

[0071] Please refer to Figure 1 , this embodiment is an intelligent optimization method for human resource management based on data mining, including the following steps:

[0072] S1. Collect the employee data, storage method data, storage method influencing factor data, characteristic data of the employee information cluster, and the number of times of employee information loss within the statistical period saved by multiple enterprises, and obtain the enterprise employee data matrix, enterprise storage method data set, enterprise storage method influencing factor data matrix, enterprise cluster characteristic data matrix, and enterprise employee information loss number data set;

[0073] Set various common information types and sensitive information types of employees to obtain the employee common information type set and the employee sensitive information type set , a 1i , a 2i respectively represent the i-th type of common employee information and sensitive employee information set, , respectively represent the total number of the set employee common information types and employee sensitive information types;

[0074] Set the employee information collection type set , the employee information storage method set and the storage method influencing factor type set ; , , respectively represent the number of common information types in the employee information, the number of sensitive information types in the employee information, and the number of employees; , respectively represent the separate storage of employee common information and sensitive information and the centralized storage of employee common information and sensitive information, which are represented by 0 and 1 respectively; represents the i-th storage method influencing factor type set, represents the total number of the set storage method influencing factor types.

[0075] S12. Combine with the set of ordinary employee information types and the set of sensitive employee information types, and collect employee data, storage method data, and storage method influencing factor data saved by multiple enterprises according to the employee information collection type set, the employee information storage method set, and the storage method influencing factor type set, to obtain the enterprise-employee data matrix , the enterprise storage method data set , and the enterprise storage method influencing factor data matrix ; represents the storage method data of the employee data of the i-th enterprise collected, and b represents the total number of enterprises collected; , are as follows respectively,

[0076] ; ;

[0077] Among them, , , respectively represent the quantity data of ordinary information types, the quantity data of sensitive information types, and the employee quantity data in the employee information of the i-th enterprise collected; represents the storage method influencing factor data of the j-th type of the i-th enterprise collected.

[0078] S13. Set the set of employee information storage cluster feature types , represents the i-th employee information storage cluster feature type in the set of the set of employee information storage cluster features set, represents the total number of employee information storage cluster feature types in the set of the set of employee information storage cluster features; the set of employee information storage cluster feature types includes the number of servers, performance level, cost performance level, scalability level, availability level, load balancing ability level, error recovery ability level, transparency level, manageability level, and programmability level of the employee information storage cluster;

[0079] Set the statistical period; collect the feature data of the employee information storage clusters of multiple enterprises and the number of times of employee information loss during the statistical period according to the set of employee information storage cluster feature types, to obtain the enterprise cluster feature data matrix and the enterprise employee information loss times data set , represents the employee information loss times data of the i-th enterprise collected; is as follows,

[0080] ;

[0081] Among them, It represents the characteristic data of the employee information storage cluster of the j-th type of the i-th enterprise collected.

[0082] S2. Construct the final storage method mapping equation by using the enterprise employee data matrix, the enterprise storage method data set, and the enterprise storage method influencing factor data matrix; construct the final employee information loss times mapping model by using the enterprise cluster characteristic data matrix and the enterprise employee information loss times data set;

[0083] The S2 includes the following steps:

[0084] S21. Construct the initial storage method mapping equation; as follows,

[0085] ;

[0086] Where, represents the dependent variable of the initial storage method mapping equation and represents the employee information storage method data; represents the i-th independent variable of the initial storage method mapping equation and represents the i-th type of enterprise employee data, represents the independent variable coefficient of; represents the (i + 3)-th independent variable of the initial storage method mapping equation and represents the i-th type of enterprise employee information storage method influencing factor data, represents the independent variable coefficient of; represents the bias of the initial storage method mapping equation, and ceil represents the ceiling function.

[0087] S22. Optimize the initial storage method mapping equation by using the enterprise employee data matrix, the enterprise storage method data set, and the enterprise storage method influencing factor data matrix to obtain the final storage method mapping equation.

[0088] The S22 includes the following steps:

[0089] S221. Input each row of data in the enterprise employee data matrix and the enterprise storage method influencing factor data matrix into the initial storage method mapping equation for mapping to obtain the initial storage method mapping data set , c i represents inputting the enterprise employee data matrix and the i-th row of data in the enterprise storage method influencing factor data matrix into the initial storage method mapping equation for mapping to obtain the storage method mapping data.

[0090] S222. When there are differences between the storage method mapping data in the initial storage method mapping dataset and the corresponding enterprise storage method data in the enterprise storage method dataset, optimize the initial storage method mapping equation until there are no such differences between the storage method mapping data in the initial storage method mapping dataset and the corresponding enterprise storage method data in the enterprise storage method dataset, and obtain the final storage method mapping equation; otherwise, there is no need to optimize the initial storage method mapping equation, and use the initial storage method mapping equation as the final storage method mapping equation.

[0091] The optimization of the initial storage method mapping equation in S222 includes the following steps:

[0092] S2221. Construct the first dragonfly population , denotes the i-th dragonfly in the first dragonfly population, denotes the scale of the first dragonfly population; set the maximum number of iterations of the first dragonfly population to and the current number of iterations to , denoted as the first maximum number of iterations and the first current number of iterations respectively; the search space dimension of the first dragonfly population is .

[0093] S2222. Set the value range of each sub-variable coefficient and bias in the initial storage method mapping equation to obtain the independent variable value range matrix d1 and the bias value range [d 21 , d 22 ; d 21 , d 22 respectively represent the lower limit and upper limit of the bias value in the initial storage method mapping equation; d1 is as follows,

[0094] ;

[0095] Among them, , , respectively represent the lower limits of the values of the 1st, 2nd, and 3rd independent variable coefficients in the initial storage method mapping equation, , , respectively represent the upper limits of the values of the 1st, 2nd, and 3rd independent variable coefficients in the initial storage method mapping equation; , respectively represent the lower limit and upper limit of the value of the (i + 3)-th independent variable coefficient in the initial storage method mapping equation;

[0096] Set the initial position of each dragonfly in the first dragonfly population according to the independent variable value interval matrix and the bias value interval, and obtain the first initial position matrix set , represents the initial position matrix of the k-th dragonfly in the first dragonfly population; as follows,

[0097] ;

[0098] wherein, , , respectively represent the components of the initial position of the k-th dragonfly in the first dragonfly population on the 1st, 2nd, and 3rd independent variable coefficient dimensions in the initial storage mode mapping equation, represents the component of the initial position of the k-th dragonfly in the first dragonfly population on the (i + 3)-th independent variable coefficient dimension in the initial storage mode mapping equation, represents the component of the initial position of the k-th dragonfly in the first dragonfly population on the bias dimension in the initial storage mode mapping equation; , , , , The calculation formulas of are as follows,

[0099] ; ;

[0100] ; ;

[0101] ;

[0102] In the formula, rand 1k11 , rand 1k12 , rand 1k13 , rand 1k2i , rand 1k3 respectively represent random numbers between 0 and 1 generated for , , , , .

[0103] S2223. According to the initial storage mode mapping data set c = {c1, c2,..., c i ,..., c b} and the enterprise storage mode data set Construct the fitness function of the first dragonfly population ; as follows,

[0104] ;

[0105] In the formula, β is a positive number representing a correction parameter.

[0106] S2224. Start iteration. Before iteration, set the first current iteration number to 1; in the first round of iteration, use the fitness function of the first dragonfly population to calculate the fitness values of the initial position matrices of each dragonfly in the first initial position matrix set, obtaining a first fitness value set; take the maximum fitness value in the first fitness value set and the corresponding initial position matrix of the dragonfly as the first global best fitness and the first global best position respectively; update the initial position matrices of each dragonfly in the first initial position matrix set according to the first global best fitness and the first global best position; after the update is completed, increment the first current iteration number by 1 and enter the next round of iteration.

[0107] In each subsequent round of iteration, use the fitness function of the first dragonfly population to calculate the fitness values of the position matrices of each dragonfly in the first dragonfly population updated in the previous round of iteration, obtaining a second fitness value set; take the maximum fitness value in the second fitness value set and the corresponding position matrix of the dragonfly as the second global best fitness and the second global best position respectively; update the position matrices of each dragonfly in the first dragonfly population updated in the previous round of iteration again according to the second global best fitness and the second global best position; after the update is completed, increment the first current iteration number by 1 and enter the next round of iteration.

[0108] S2225. When , stop iteration to obtain the first final global best position; otherwise, continue iteration until ; substitute the first final global best position into the initial storage mode mapping equation to obtain an optimized storage mode mapping equation; input each row of data in the enterprise employee data matrix and the enterprise storage mode influencing factor data matrix into the optimized storage mode mapping equation for mapping to obtain an optimized storage mode mapping data set.

[0109] When there is no difference between the storage method mapping data in the optimized storage method mapping dataset and the corresponding enterprise storage method data in the enterprise storage method dataset, the optimized storage method mapping equation is used as the final storage method mapping equation; otherwise, return to S2224 to continue the iteration until there is no difference between the storage method mapping data in the optimized storage method mapping dataset and the corresponding enterprise storage method data in the enterprise storage method dataset, and the final storage method mapping equation is obtained.

[0110] The dragonfly optimization algorithm can effectively conduct global exploration in the search space by simulating the dynamic and static clustering behaviors of dragonflies. It helps to jump out of the local optimal solution and find the global optimal solution. It has fewer parameters, which makes the adjustment and optimization process of the algorithm more concise and has good robustness. Based on the above advantages, the dragonfly optimization algorithm is used in this solution to iteratively optimize the coefficients of each independent variable and the bias of the initial storage method mapping equation multiple times, and the sum of the differences between the data mapped by the initial storage method mapping equation and the actual data is used as the fitness function. Therefore, as the iteration progresses, the difference between the data mapped by the initial storage method mapping equation and the actual data will become smaller and smaller, and finally meet the accuracy requirements.

[0111] S23. Construct an initial mapping model for the number of times of employee information loss; use the enterprise cluster feature data matrix and the enterprise employee information loss times dataset to train and test the initial mapping model for the number of times of employee information loss; after the training and testing are completed, the final mapping model for the number of times of employee information loss is obtained.

[0112] The S23 includes the following steps:

[0113] S231. Construct an initial SVM classification model and set the training data ratio; use the initial SVM classification model as the initial mapping model for the number of times of employee information loss; divide the enterprise cluster feature data matrix and the enterprise employee information loss times dataset according to the training data ratio to obtain an enterprise cluster feature training data matrix, an enterprise employee information loss times training dataset, an enterprise cluster feature test data matrix, and an enterprise employee information loss times test dataset.

[0114] S232. Set the training error threshold; use the enterprise cluster feature training data matrix and the enterprise employee information loss times training dataset as training data and training labels respectively and input them into the initial mapping model for the number of times of employee information loss for training; during the training process, when the training error is less than the training error threshold, stop the training and obtain the trained mapping model for the number of times of employee information loss; otherwise, continue the training until the training error is less than the training error threshold.

[0115] S233. Set the test accuracy threshold; use the enterprise cluster feature test data matrix and the enterprise employee information loss times test data set as test data and test labels respectively, and input them into the trained employee information loss times mapping model for testing; after the test is completed, obtain the test accuracy; when the test accuracy is greater than or equal to the test accuracy threshold, use the trained employee information loss times mapping model as the final employee information loss times mapping model; otherwise, return to S232 to continue training the trained employee information loss times mapping model until the test accuracy is greater than or equal to the test accuracy threshold.

[0116] By respectively using the enterprise cluster feature training data matrix, the enterprise employee information loss times training data set, the enterprise cluster feature test data matrix, and the enterprise employee information loss times test data set to train and test the initial employee information loss times mapping model, the obtained final employee information loss times mapping model has better mapping and classification capabilities between enterprise cluster feature data and enterprise employee information loss times data.

[0117] S3. Use the final storage method mapping equation to map the employee data and storage method influencing factor data of the current enterprise to obtain the current storage method data; use the final employee information loss times mapping model to map the feature data of the employee information cluster stored by the current enterprise to obtain the current enterprise employee information loss times data.

[0118] S3 includes the following steps:

[0119] S31. According to the employee general information type set and the employee sensitive information type set, and according to the employee information collection type set and the storage method influencing factor type set, collect the employee data and storage method influencing factor data of the enterprise that needs to optimize human resource management currently, and obtain the current employee data set e1 = {e 11 , e 12 , e 13} and the current storage method influencing factor data set ; e 11 , e 12 , e 13 respectively represent the quantity data of the general information type in the employee information, the quantity data of the sensitive information type in the employee information, and the employee quantity data of the enterprise that needs to optimize human resource management currently; e 3i represents the i-th type of storage method influencing factor data of the enterprise that needs to optimize human resource management currently.

[0120] S32. Collect the characteristic data of the storage employee information cluster of the enterprise that currently needs to optimize human resource management according to the set of characteristic types of the employee information storage cluster, and obtain the current cluster characteristic data set. , e 4i represents the characteristic data of the i-th type of employee information storage cluster of the enterprise that needs to optimize human resource management currently.

[0121] S33. Input the current employee data set, the current storage method data, and the current storage method influencing factor data set into the final storage method mapping equation for mapping, and obtain the current storage method data e2.

[0122] Input the current cluster characteristic data set into the final employee information loss times mapping model for mapping, and obtain the current enterprise employee information loss times data e5.

[0123] S4. Store the employee information of the current enterprise according to the current storage method data; optimize the current cluster characteristic data set according to the current enterprise employee information loss times data, and obtain the current final cluster characteristic data set.

[0124] The S4 includes the following steps:

[0125] S41. Store the employee information of the current enterprise according to the current storage method data e2.

[0126] S42. Set the information loss times threshold within the statistical period; when the current enterprise employee information loss times data e5 is greater than or equal to the information loss times threshold, adjust the current cluster characteristic data set until the current enterprise employee information loss times data e5 is less than the information loss times threshold, and obtain the current final cluster characteristic data set; otherwise, there is no need to adjust the current cluster characteristic data set, and use the current cluster characteristic data set as the current final cluster characteristic data set.

[0127] The steps of adjusting the current cluster characteristic data set in S42 until the current enterprise employee information loss times data e5 is less than the information loss times threshold and obtaining the current final cluster characteristic data set include the following steps:

[0128] S421. Construct the second dragonfly population , represents the i-th dragonfly in the second dragonfly population, represents the scale of the second dragonfly population; set the maximum number of iterations of the second dragonfly population to be and the current number of iterations to be , which are respectively denoted as the second maximum number of iterations and the second current number of iterations; the search space dimension of the second dragonfly population is .

[0129] S422. Set the value range of the data corresponding to each type of employee information storage cluster feature in the employee information storage cluster feature type set, and obtain the cluster feature value range set d3; as follows,

[0130] ;

[0131] Among them, , respectively represent the lower limit and the upper limit of the value of the data corresponding to the i-th type of employee information storage cluster feature in the employee information storage cluster feature type set;

[0132] Set the initial position of each dragonfly in the second dragonfly population according to the cluster feature value range set, and obtain the second initial position matrix ; as follows,

[0133] ;

[0134] Among them, represents the component of the initial position of the k-th dragonfly in the second dragonfly population on the i-th dimension; the calculation formula is as follows,

[0135] ;

[0136] In the formula, rand 2ki represents a random number between 0 and 1 generated for .

[0137] S423. Construct the fitness function of the second dragonfly population according to the current number of enterprise employee information loss data e5 ; as follows,

[0138]

[0139] S424. Start the iteration. Before the iteration, set the second current number of iterations to 1; in the first round of iteration, use the fitness function of the second dragonfly population And cooperate with the final employee information loss times mapping model to calculate the fitness value of the initial position of each dragonfly in the second initial position matrix, obtaining the third fitness value set; take the maximum fitness value in the third fitness value set and the corresponding initial position of the dragonfly as the third global best fitness and the third global best position respectively; update the initial position of each dragonfly in the second initial position matrix according to the third global best fitness and the third global best position; after the update is completed, add 1 to the second current iteration number and enter the next round of iteration;

[0140] In each other round of iteration process, use the fitness function of the second dragonfly population and cooperate with the final employee information loss times mapping model to calculate the fitness value of the position of each dragonfly in the second dragonfly population updated in the previous round of iteration, obtaining the fourth fitness value set; take the maximum fitness value in the fourth fitness value set and the corresponding position of the dragonfly as the fourth global best fitness and the fourth global best position respectively; update the position of each dragonfly in the second dragonfly population updated in the previous round of iteration again according to the fourth global best fitness and the fourth global best position; after the update is completed, add 1 to the second current iteration number and enter the next round of iteration.

[0141] S425. When is satisfied, stop the iteration and obtain the second final global best position; otherwise, continue the iteration until is satisfied; take the second final global best position as the optimized current cluster feature dataset; input the optimized current cluster feature dataset into the final employee information loss times mapping model for mapping to obtain the optimized current enterprise employee information loss times data;

[0142] When the optimized current enterprise employee information loss times data is less than the information loss times threshold, take the optimized current cluster feature dataset as the current final cluster feature dataset; otherwise, return to S424 to continue the iteration until the optimized current enterprise employee information loss times data is less than the information loss times threshold.

[0143] Embodiment 2

[0144] This embodiment discloses an intelligent optimization system for human resource management based on data mining. The system can implement the method of the above embodiment, including an employee information type setting module, an enterprise employee information storage method data collection module, an enterprise employee information storage method data collection module, a storage method mapping equation construction module, an employee information loss times mapping model construction module, a current enterprise employee storage data collection module, a first mapping module, a second mapping module, a storage method management module, and a cluster feature optimization module;

[0145] The employee information type setting module is used to set various types of ordinary information and sensitive information of employees, obtaining an employee ordinary information type set and an employee sensitive information type set;

[0146] The enterprise employee information storage method data collection module is used to collect employee data, storage method data, and storage method influencing factor data saved by multiple enterprises in cooperation with the employee ordinary information type set and the employee sensitive information type set, obtaining an enterprise employee data matrix, an enterprise storage method data set, and an enterprise storage method influencing factor data matrix;

[0147] The enterprise employee information storage method data collection module is used to collect the characteristic data of the employee information cluster stored by multiple enterprises and the number of times of employee information loss within the statistical period, obtaining an enterprise cluster characteristic data matrix and an enterprise employee information loss times data set;

[0148] The storage method mapping equation construction module is used to construct a final storage method mapping equation by using the enterprise employee data matrix, the enterprise storage method data set, and the enterprise storage method influencing factor data matrix;

[0149] The employee information loss times mapping model construction module is used to construct a final employee information loss times mapping model by using the enterprise cluster characteristic data matrix and the enterprise employee information loss times data set;

[0150] The current enterprise employee storage data collection module is used to collect employee data, storage method influencing factor data, and the characteristic data of the employee information cluster stored by the enterprise that needs to optimize human resource management currently, obtaining a current employee data set, a current storage method influencing factor data set, and a current cluster characteristic data set;

[0151] The first mapping module is used to map the current employee data set and the current storage method influencing factor data set by using the final storage method mapping equation, obtaining the current storage method data;

[0152] The second mapping module is used to map the current cluster characteristic data set by using the final employee information loss times mapping model, obtaining the current enterprise employee information loss times data;

[0153] The storage mode management module is used to store the employee information of the current enterprise according to the current storage mode data;

[0154] The cluster feature optimization module is used to optimize the current cluster feature data set according to the data of the number of times of loss of employee information of the current enterprise, and obtain the current final cluster feature data set.

[0155] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0156] 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 embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification in order to better explain the principle and practical application of the invention, so that those skilled in the art in the relevant technical field can well understand and utilize the invention.

Claims

1. A human resource management intelligent optimization method based on data mining, characterized in that: The following steps are involved: S1. Collect employee data, storage method data, storage method influencing factor data, characteristic data of stored employee information clusters, and the number of times employee information is lost during the statistical period saved by multiple companies, and obtain enterprise employee data matrix, enterprise storage method data set, enterprise storage method influencing factor data matrix, enterprise cluster characteristic data matrix, and enterprise employee information loss number data set; S2. Construct an initial employee information loss times mapping model; use the enterprise employee data matrix, enterprise storage method data set, and enterprise storage method influencing factor data matrix to train and test the initial employee information loss times mapping model to obtain the final storage method mapping equation; The enterprise cluster characteristic data matrix and the enterprise employee information loss frequency data set are used to build the final employee information loss frequency mapping model; The initial employee information loss times mapping model adopts the SVM model; S3, using the final storage mode mapping equation to map the current enterprise's employee data and storage mode influencing factor data to obtain current storage mode data; The final employee information loss frequency mapping model is used to map the characteristic data of the current enterprise stored employee information cluster to obtain the current enterprise employee information loss frequency data; S4, storing the employee information of the current enterprise according to the current storage mode data; According to the current data on the number of times the information of the employees in the enterprise is lost, the current cluster feature data set is optimized by using the dragonfly optimization algorithm to obtain the current final cluster feature data set.

2. According to claim 1, a method for intelligent optimization of human resource management based on data mining is characterized in that: Set common information types and sensitive information types of multiple employees to obtain common information type sets and sensitive information type sets of employees; set employee information collection type sets, employee information storage method sets, and storage method influencing factor type sets; S12, collecting employee data, storage method data, and storage method influencing factor data stored by multiple companies according to the employee general information type set and the employee sensitive information type set and the employee information collection type set, the employee information storage method set, and the storage method influencing factor type set, to obtain an enterprise employee data matrix, an enterprise storage method data set, and an enterprise storage method influencing factor data matrix; S13. Set a characteristic type set of an employee information storage cluster; the characteristic type set of the employee information storage cluster includes the number of servers, performance level, cost-effectiveness level, scalability level, availability level, load balancing capability level, error recovery capability level, transparency level, manageability level and programmability level of the employee information storage cluster; set a statistical period; collect characteristic data of employee information storage clusters of multiple enterprises and the number of times employee information is lost during the statistical period according to the characteristic type set of the employee information storage cluster, and obtain an enterprise cluster characteristic data matrix and an enterprise employee information loss frequency data set.

3. The method for intelligent optimization of human resource management based on data mining according to claim 2 is characterized in that: The S2 comprises the following steps: S21, constructing an initial storage mode mapping equation; S22, optimizing the initial storage mode mapping equation using the enterprise employee data matrix, the enterprise storage mode data set, and the enterprise storage mode influencing factor data matrix to obtain a final storage mode mapping equation; S23, constructing an initial employee information loss frequency mapping model; using the enterprise cluster feature data matrix and the enterprise employee information loss frequency data set to train and test the initial employee information loss frequency mapping model; after the training and testing are completed, obtaining the final employee information loss frequency mapping model.

4. The method for intelligent optimization of human resource management based on data mining according to claim 3 is characterized in that: The S22 comprises the following steps: S221, inputting each row of data in the enterprise employee data matrix and the enterprise storage mode influencing factor data matrix into the initial storage mode mapping equation for mapping, to obtain an initial storage mode mapping data set; S222. When the initial storage mode mapping data set contains storage mode mapping data that is different from the corresponding enterprise storage mode data in the enterprise storage mode data set, the initial storage mode mapping equation is optimized until the initial storage mode mapping data set contains storage mode mapping data that is different from the corresponding enterprise storage mode data in the enterprise storage mode data set, and a final storage mode mapping equation is obtained; otherwise, there is no need to optimize the initial storage mode mapping equation, and the initial storage mode mapping equation is used as the final storage mode mapping equation.

5. The method for intelligent optimization of human resource management based on data mining according to claim 4 is characterized by: In S222, the initial storage mode mapping equation is optimized using a dragonfly optimization algorithm.

6. The method for intelligent optimization of human resource management based on data mining according to claim 3 is characterized by: The initial employee information loss times mapping model described in S23 adopts the SVM model.

7. The method for intelligent optimization of human resource management based on data mining according to claim 6 is characterized in that: The S3 comprises the following steps: S31, collecting employee data and storage method influencing factor data of the enterprise currently requiring human resource management optimization according to the employee general information type set and the employee sensitive information type set and according to the employee information collection type set and the storage method influencing factor type set, to obtain a current employee data set and a current storage method influencing factor data set; S32, collecting characteristic data of the employee information storage cluster of the enterprise that currently needs to optimize human resource management according to the employee information storage cluster characteristic type set, to obtain a current cluster characteristic data set; S33, input the current employee data set, the current storage mode data and the current storage mode influencing factor data set into the final storage mode mapping equation for mapping, and obtain the current storage mode data; input the current cluster feature data set into the final employee information loss times mapping model for mapping, and obtain the current enterprise employee information loss times data.

8. The method for intelligent optimization of human resource management based on data mining according to claim 7 is characterized in that: The S4 comprises the following steps: S41, storing the employee information of the current enterprise according to the current storage mode data; S42. Set a threshold value for the number of information losses within the statistical period; when the current number of times the enterprise employee information is lost is greater than or equal to the threshold value for the number of information losses, adjust the current cluster feature data set until the current number of times the enterprise employee information is lost is less than the threshold value for the number of information losses, and obtain the current final cluster feature data set; otherwise, there is no need to adjust the current cluster feature data set, and the current cluster feature data set is used as the current final cluster feature data set.

9. The method for intelligent optimization of human resource management based on data mining according to claim 8 is characterized in that: In S42, the current cluster feature data set is adjusted until the number of times the current enterprise employee information is lost is less than the information loss number threshold, and obtaining the current final cluster feature data set includes the following steps: S421, construct a second dragonfly population; set the maximum number of iterations of the second dragonfly population to And the current iteration number is , respectively recorded as the second maximum number of iterations and the second current number of iterations; S422, setting a value interval of data corresponding to each employee information storage cluster feature type in the employee information storage cluster feature type set to obtain a cluster feature value interval set; setting an initial position of each dragonfly in the second dragonfly population according to the cluster feature value interval set to obtain a second initial position matrix; S423, constructing a fitness function of the second dragonfly population according to the current enterprise employee information loss frequency data; S424, start iteration, set the second current iteration number to 1 before iteration; in each round of iteration, use the fitness function of the second dragonfly population and cooperate with the final employee information loss number mapping model to calculate the fitness value of the position of each dragonfly in the second dragonfly population updated in the previous round of iteration, and update the position of each dragonfly in the second dragonfly population updated in the previous round of iteration; after the update is completed, add 1 to the second current iteration number and enter the next round of iteration; S425, when When , stop the iteration and get the second final global optimal position; otherwise, continue to iterate until until the second final global optimal position is used as the optimized current cluster feature data set; the optimized current cluster feature data set is input into the final employee information loss times mapping model for mapping, and the optimized current enterprise employee information loss times data is obtained; When the optimized current enterprise employee information loss times data is less than the information loss times threshold, the optimized current cluster feature data set is used as the current final cluster feature data set; otherwise, return to S424 to continue iterating until the optimized current enterprise employee information loss times data is less than the information loss times threshold.

10. A system for implementing the intelligent optimization method for human resource management based on data mining as described in any one of claims 1 to 9.

Images