Human resource management method and system based on big data

Abstract

The invention discloses a human resource management method and system based on big data, and the method comprises the steps: obtaining and analyzing the historical multi-dimensional information data of an employee, and determining an influence factor related to the employee demission; extracting data features of the influence factors, and constructing a demission risk prediction model based on the data features and a preset model for prediction to obtain a demission risk prediction value; analyzing and evaluating the historical multi-dimensional information data of the employees based on preset value indexes to obtain comprehensive evaluation values of the preset value indexes, and performing comprehensive analysis and calculation on the overall values of the employees based on the comprehensive evaluation values to obtain value coefficient values of the employees; and classifying the employees based on the two values to obtain a plurality of employee types, and setting corresponding management strategies for the employee types for management. According to the invention, by combining the remission risk prediction value with the value coefficient, the employees are accurately classified, visualization and refinement of employee inventory are realized, personalized management strategies are provided for different employees, and the management efficiency of the whole talent team is improved.

Description

Technical Field

[0001] This invention relates to the field of data analysis, and in particular to a human resource management method and system based on big data. Background Technology

[0002] In today's fierce global competition and digital economy wave, talent has become the most core strategic asset of organizations. However, traditional human resource management largely relies on lagging indicators and subjective experience-based judgments. This "hindsight" management model is ill-equipped to cope with the increasingly fierce competition for talent and the complex and ever-changing business environment. Problems are only realized when top talents submit their resignations. Furthermore, there is a lack of scientific basis for accurately identifying high-potential employees and for allocating limited incentive resources to the most critical talents. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a human resource management method and system based on big data, comprising:

[0004] Obtain historical multidimensional information data of employees in the company, and analyze the historical multidimensional information data to identify the influencing factors related to employee turnover;

[0005] Extract the data features of influencing factors, and construct a turnover risk prediction model based on the data features and a preset neural network model to predict the future turnover risk of employees.

[0006] Based on preset value indicators, the historical multidimensional information data of employees are analyzed and evaluated to obtain the comprehensive evaluation value of each preset value indicator.

[0007] Based on the comprehensive evaluation value of each preset value indicator, the overall value of employees is comprehensively analyzed and calculated to obtain the value coefficient value of each employee.

[0008] Employees are categorized based on their predicted turnover risk and value coefficient, resulting in multiple employee types. Corresponding management strategies are then set for each employee type.

[0009] Furthermore, the process involves acquiring historical multidimensional information data of employees within the company, analyzing this data, and identifying influencing factors related to employee turnover, including:

[0010] Acquire historical multidimensional information data of employees in the company, and divide the historical multidimensional information data into multiple information data groups according to information type;

[0011] Identify employee turnover data and calculate the correlation between each data group and the turnover data.

[0012] Information data groups with a relevance greater than a preset threshold are identified as influencing factors related to employee turnover.

[0013] Furthermore, the extraction of data features of influencing factors, and the construction of a turnover risk prediction model based on these data features and a preset neural network model to predict future turnover risk values ​​for employees, includes:

[0014] Extract the data features of the information data group corresponding to the influencing factors, construct a dataset based on the historical information data group and the corresponding data features, and input the dataset into the preset neural network model to construct the initial model for predicting the risk of turnover.

[0015] The dataset is divided into training and testing sets according to a preset ratio, and the training and testing sets are input into the initial model for predicting turnover risk.

[0016] The initial model for predicting employee turnover risk is trained and tested until it meets the preset convergence conditions, thus obtaining the employee turnover risk prediction model.

[0017] The system acquires current multidimensional information data of employees and inputs it into the turnover risk prediction model to make predictions and obtain the predicted value of employee turnover risk.

[0018] Furthermore, the analysis and evaluation of employees' historical multidimensional information data based on preset value indicators yields a comprehensive evaluation value for each preset value indicator, including:

[0019] Determine the preset value indicators and identify the corresponding data from the employee's historical multidimensional information data;

[0020] The data is input into the corresponding evaluation model, and the evaluation model outputs the evaluation values ​​of each preset value indicator.

[0021] Determine the preset evaluation coefficients corresponding to each preset value indicator, and calculate the comprehensive evaluation value of each preset value indicator based on the preset evaluation coefficients and evaluation values ​​corresponding to each preset value indicator.

[0022] Furthermore, the formula for calculating the comprehensive evaluation value of the preset value index is as follows:

[0023] Di = fi * Pi,

[0024] Where Di is the comprehensive evaluation value of the i-th preset value indicator, fi is the preset evaluation coefficient of the i-th preset value indicator, and Pi is the evaluation value of the i-th preset value indicator.

[0025] Furthermore, the overall value of employees is comprehensively analyzed and calculated based on the comprehensive evaluation values ​​of each preset value indicator to obtain the value coefficient value of each employee, including:

[0026] Determine the preset value corresponding to each preset value indicator, and calculate the difference between the comprehensive evaluation value of each preset value indicator and the corresponding preset value to obtain the difference value of each preset value indicator.

[0027] A preset weight-difference interval correspondence is set in advance. For each difference interval, a corresponding preset weight is associated with it.

[0028] Obtain the difference value of each preset value indicator, and based on the mapping relationship between the difference value interval to which the difference value belongs and the preset weight-difference value interval correspondence, select the preset weight corresponding to the difference value interval to determine the weight of each preset value indicator.

[0029] The value coefficient of each employee is calculated based on the weights and comprehensive evaluation values ​​corresponding to each preset value indicator.

[0030] Furthermore, the formula for calculating the employee's value coefficient is as follows:

[0031]

[0032] Where Z is the difference, k is the preset conversion coefficient, αi is the weight of the i-th preset value indicator, and Di is the comprehensive evaluation value of the i-th preset value indicator.

[0033] Furthermore, based on each employee's predicted turnover risk value and value coefficient value, employees are categorized into multiple employee types, including:

[0034] A matrix is ​​constructed based on the predicted turnover risk value and value coefficient value of each employee. The horizontal axis of the matrix is ​​the value coefficient value from low to high, and the vertical axis is the predicted turnover risk value from low to high.

[0035] Determine the median value between the predicted turnover risk value and the value coefficient value for each employee. The median value is the sum of the highest and lowest values, divided by 2.

[0036] The matrix is ​​divided into four quadrants based on the median value of the predicted turnover risk and the median value of the value coefficient. Employees in the same quadrant are classified into the same employee type, resulting in multiple employee types, including high-value low turnover risk, high-value high turnover risk, low-value low turnover risk, and low-value high turnover risk.

[0037] Furthermore, the management of each employee type by setting corresponding management strategies includes:

[0038] For employees in the high-value, low-turnover-risk category, the management strategy is to provide market-leading compensation, bonuses, and long-term equity incentives, grant them strategic projects, innovative challenges, and decision-making autonomy, designate them as successors for key positions, and provide mentorship from senior executives.

[0039] For employees with high value and high turnover risk, the management strategy is to have a meeting between the supervisor and HR to provide personalized solutions for the main reasons for leaving, and to activate a fast track for compensation or give clear promotion / project commitments.

[0040] For employees in the low-value, low-turnover-risk category, the management strategy is to provide targeted training to bridge skills gaps, set more challenging goals, strengthen process coaching and feedback, and maintain their work motivation through timely praise and small incentives.

[0041] For employees in the low-value, high-turnover-risk category, the management strategy is to implement time-limited performance improvement plans, clarify improvement goals and support, conduct frequent checks and coaching, and document the progress. If there is no significant improvement during the improvement period, the optimization or negotiated resignation process will be initiated decisively.

[0042] The management strategy based on the employee type is used to manage employees in the same way in each employee type.

[0043] This invention also provides a big data-based human resource management system, comprising:

[0044] The acquisition module is used to acquire historical multidimensional information data of employees in the company, and to analyze the historical multidimensional information data to determine the influencing factors related to employee turnover.

[0045] The prediction module is used to extract the data features of influencing factors and build a turnover risk prediction model based on the data features and a preset neural network model to make predictions and obtain the predicted value of the employee's future turnover risk.

[0046] The evaluation module is used to analyze and evaluate employees' historical multidimensional information data based on preset value indicators, and obtain the comprehensive evaluation value of each preset value indicator.

[0047] The calculation module is used to perform a comprehensive analysis and calculation of the overall value of employees based on the comprehensive evaluation value of each preset value indicator, and to obtain the value coefficient value of each employee.

[0048] The management module is used to classify employees based on their predicted turnover risk and value coefficient, resulting in multiple employee types, and to set corresponding management strategies for each employee type.

[0049] Compared with existing technologies, the human resource management method and system based on big data proposed in this invention have the following advantages:

[0050] This invention uses a neural network model to perform in-depth analysis of hundreds of multidimensional features, which can quantify the turnover probability of each employee, enabling management to shift from passively responding to resignation applications to providing early warnings months in advance, and providing a scientific basis for targeted retention measures, thereby significantly reducing the turnover rate of key talents and directly protecting the organization's intellectual assets.

[0051] This invention breaks away from the previous subjective and fragmented evaluation methods. Through preset multi-dimensional value indicators, it conducts standardized and comparable comprehensive value assessments of employees. This not only realizes the "visualization" of talent assets, allowing managers to clearly understand the distribution of team members, but also provides a fair and transparent decision-making basis for resource allocation, ensuring that the best resources are invested in the employees who can create the most value.

[0052] This invention achieves personalized and optimized management strategies through precise matching of talent classification and strategies. It integrates employees' "predicted turnover risk value" and "value coefficient" into a unified matrix, automatically identifies four types of people, and triggers preset and differentiated management strategies for each type of people. This allows limited management energy and resources to be allocated most effectively, thereby maximizing the overall talent management efficiency and the return on organizational talent capital. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the process structure of the big data-based human resource management method in an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of the composition of a big data-based human resource management system in an embodiment of the present invention. Detailed Implementation

[0055] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0056] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the platform or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0057] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0058] like Figure 1 As shown in the embodiments of this application, a human resource management method based on big data is provided, including: S100: acquiring historical multidimensional information data of employees in the company, and analyzing the historical multidimensional information data to determine the influencing factors related to employee turnover; S200: extracting data features of the influencing factors, and constructing a turnover risk prediction model based on the data features and a preset neural network model to predict the future turnover risk of employees; S300: analyzing and evaluating the historical multidimensional information data of employees based on preset value indicators to obtain a comprehensive evaluation value of each preset value indicator; S400: comprehensively analyzing and calculating the overall value of employees based on the comprehensive evaluation value of each preset value indicator to obtain the value coefficient value of each employee; S500: classifying employees based on the turnover risk prediction value and value coefficient value of each employee to obtain multiple employee types, and setting corresponding management strategies for each employee type.

[0059] Furthermore, this invention utilizes a neural network model to conduct in-depth analysis of hundreds of multidimensional features, quantifying the turnover probability of each employee. This transforms management from passively responding to resignation requests to providing advance warnings months in advance, offering a scientific basis for targeted retention measures. Consequently, it significantly reduces the turnover rate of key personnel, directly protecting the organization's intellectual assets. This invention breaks away from previous subjective and fragmented evaluation methods. Through pre-set multi-dimensional value indicators, it provides standardized and comparable comprehensive value assessments of employees. This not only achieves "visualization" of talent assets, allowing managers to clearly understand the distribution of team personnel, but also provides a fair and transparent decision-making basis for resource allocation, ensuring that the best resources are invested in the employees who create the most value. Through precise matching of talent classification and strategies, this invention achieves personalized and optimized management strategies. It integrates employees' "turnover risk prediction value" and "value coefficient" into a unified matrix, automatically identifying four groups and triggering pre-set, differentiated management strategies for each group. This allows limited management effort and resources to be allocated most effectively, thereby maximizing overall talent management efficiency and the return on organizational human capital.

[0060] In an embodiment of this application, a big data-based human resource management method is provided. The method involves acquiring historical multidimensional information data of employees in the company and analyzing the historical multidimensional information data to determine influencing factors related to employee turnover. This includes: acquiring historical multidimensional information data of employees in the company and dividing the historical multidimensional information data into multiple information data groups according to information type; determining employee turnover data and calculating the correlation between each information data group and the turnover data; and determining the information type corresponding to the information data group with a correlation greater than a preset threshold as influencing factors related to employee turnover.

[0061] Specifically, the system systematically integrates all-dimensional historical data of employees within the enterprise and scientifically divides it into multiple logical data groups according to information attributes—including basic information, performance, behavioral trajectory, development dynamics, and environmental perception. Statistical analysis methods are used to calculate the correlation strength between each data group and actual employee turnover records, quantifying the correlation coefficients between various types of information and turnover behavior. A preset threshold mechanism is employed for significance filtering, accurately identifying the information types represented by data groups with correlation coefficients exceeding the critical value as key turnover influencing factors. This step constructs an objective screening mechanism for factor discovery, replacing traditional subjective guesswork with data correlation analysis to ensure that the identified influencing factors are statistically significant. A multi-dimensional risk warning indicator system is established, revealing the differentiated impact weights of different information dimensions on turnover decisions, laying the feature engineering foundation for subsequent accurate modeling.

[0062] In the embodiments of this application, a human resource management method based on big data is provided. The step of extracting data features of influencing factors and constructing a turnover risk prediction model based on the data features and a preset neural network model to predict the future turnover risk of employees includes: extracting data features of information data groups corresponding to influencing factors; constructing a dataset based on historical information data groups and corresponding data features; inputting the dataset into a preset neural network model to construct an initial turnover risk prediction model; dividing the dataset into a training set and a test set according to a preset ratio; inputting the training set and the test set into the initial turnover risk prediction model; training and testing the initial turnover risk prediction model until it meets a preset convergence condition to obtain the turnover risk prediction model; acquiring the current multidimensional information data of employees and inputting the current multidimensional information data into the turnover risk prediction model for prediction to obtain the employee's turnover risk prediction value.

[0063] Specifically, key data features are extracted from the data set corresponding to factors influencing employee turnover, including numerical statistical features, time-series trends, and one-hot encoding of categorical variables, to construct a structured dataset. A model training foundation is established based on historical information data and its feature labels, and this data is divided into training and test sets according to a preset ratio. The training set is input into a preset neural network model for multiple rounds of iterative training, and the model's generalization ability is verified using the test set. When the model's loss function converges and metrics such as accuracy and recall reach preset thresholds, a mature employee turnover risk prediction model is obtained. Inputting real-time multidimensional employee data into this model outputs a quantified predicted value for employee turnover risk. This step constructs a dynamically optimized prediction engine that automatically captures complex nonlinear feature relationships through machine learning, significantly improving prediction accuracy. It forms an end-to-end automated evaluation process, achieving seamless conversion from raw data to risk values ​​and supporting large-scale real-time prediction. An iterative self-improvement mechanism is established, allowing the model to continuously optimize with new data, enabling predictive capabilities to evolve over time and providing continuous intelligent support for corporate talent retention strategies.

[0064] In the embodiments of this application, a human resource management method based on big data is provided. The method involves analyzing and evaluating the historical multidimensional information data of employees based on preset value indicators to obtain a comprehensive evaluation value for each preset value indicator. The method includes: determining preset value indicators and identifying data corresponding to the preset value indicators from the historical multidimensional information data of employees; inputting the data into the corresponding evaluation model and having the evaluation model perform evaluation output to obtain the evaluation value of each preset value indicator; determining the preset evaluation coefficient corresponding to each preset value indicator and calculating the comprehensive evaluation value of each preset value indicator based on the preset evaluation coefficient and evaluation value.

[0065] Specifically, the preset value indicators cover performance contribution, competence and qualifications, cultural fit, and strategic value dimensions. Structured and unstructured data corresponding to each indicator are automatically extracted from the employee's historical multidimensional information database and input into a specially designed evaluation model. These models use natural language processing technology to parse text evaluations, employ a rule engine to quantify achievements, and use collaborative filtering algorithms to assess team influence, ultimately outputting standardized evaluation values ​​for each value indicator. Differential weight coefficients are assigned to each value indicator based on organizational strategic goals, and a weighted aggregation algorithm merges the multidimensional evaluation values ​​into a unified value coefficient. This step constructs a multidimensional value quantification system, breaking through the traditional single performance perspective and comprehensively measuring the composite value of employees to the organization; it forms an intelligent evaluation workflow, achieving automatic conversion from raw data to indicator scores through multi-model collaborative processing, significantly improving evaluation efficiency; and it establishes a configurable strategic alignment mechanism, flexibly responding to changes in the organization's talent strategy by adjusting weight coefficients, ensuring that evaluation results always align with business priorities and providing accurate and reliable data support for talent decision-making.

[0066] In an embodiment of this application, a human resource management method based on big data is provided, wherein the formula for calculating the comprehensive evaluation value of the preset value indicator is as follows:

[0067] Di = fi * Pi,

[0068] Where Di is the comprehensive evaluation value of the i-th preset value indicator, fi is the preset evaluation coefficient of the i-th preset value indicator, and Pi is the evaluation value of the i-th preset value indicator.

[0069] In the embodiments of this application, a big data-based human resource management method is provided. The method involves comprehensively analyzing and calculating the overall value of employees based on the comprehensive evaluation values ​​of various preset value indicators to obtain the value coefficient value for each employee. This includes: determining the preset value corresponding to each preset value indicator, and calculating the difference between the comprehensive evaluation value of each preset value indicator and the corresponding preset value to obtain the difference value of each preset value indicator; pre-setting a preset weight-difference interval correspondence relationship, whereby each difference interval is associated with a corresponding preset weight; obtaining the difference value of each preset value indicator, and based on the mapping relationship between the difference interval to which the difference value belongs and the preset weight corresponding to the difference interval within the preset weight-difference interval correspondence relationship, selecting the preset weight corresponding to the difference interval as the weight of each preset value indicator; and calculating the value coefficient value of each employee based on the weight corresponding to each preset value indicator and the comprehensive evaluation value.

[0070] Specifically, a baseline preset value is set for each pre-defined value indicator. The difference between an employee's comprehensive evaluation value for each indicator and the corresponding preset value is calculated to obtain a quantified difference value. A dynamic mapping relationship between weights and difference value intervals is pre-constructed, associating different preset weight coefficients with different ranges of difference values. The interval range to which each indicator's difference value belongs is automatically identified, and corresponding dynamic weights are matched accordingly. Based on these dynamic weights, the evaluation values ​​of each indicator are weighted and synthesized to generate a precise quantified employee value coefficient. This step establishes a dynamic weight adjustment mechanism, breaking through the limitations of traditional fixed weights. It intelligently adjusts the importance of indicators based on the degree of deviation between the employee's actual performance and the standard, making the evaluation results more discriminative. A goal-oriented evaluation system is formed, which intuitively presents the employee's strengths and weaknesses in various dimensions through preset values, providing a clear direction for personalized development. An adaptive talent value measurement model is constructed, with weight allocation automatically adjusted according to performance differences, ensuring that the outstanding contributions of high-value employees are fully reflected, effectively improving the accuracy and fairness of talent evaluation, and providing a reliable basis for differentiated talent management strategies.

[0071] In an embodiment of this application, a human resource management method based on big data is provided, wherein the formula for calculating the employee's value coefficient is as follows:

[0072]

[0073] Where Z is the difference, k is the preset conversion coefficient, αi is the weight of the i-th preset value indicator, and Di is the comprehensive evaluation value of the i-th preset value indicator.

[0074] In an embodiment of this application, a big data-based human resource management method is provided. The method categorizes employees based on their predicted turnover risk and value coefficient values ​​to obtain multiple employee types. This includes: constructing a matrix based on each employee's predicted turnover risk and value coefficient values, where the horizontal axis of the matrix represents value coefficient values ​​from low to high, and the vertical axis represents predicted turnover risk values ​​from low to high; determining the median value between each employee's predicted turnover risk and value coefficient values, where the median value is the sum of the highest and lowest values, divided by 2; dividing the matrix into four quadrants based on the median value of the predicted turnover risk and the median value of the value coefficient values, and grouping employees within the same quadrant into the same employee type to obtain multiple employee types, including high-value, low-turnover-risk, high-value, high-turnover-risk, low-value, low-turnover-risk, and low-value, high-turnover-risk types.

[0075] Specifically, by constructing a two-dimensional matrix with the value coefficient as the horizontal axis and the turnover risk prediction value as the vertical axis, and using the calculated median value as the dividing point (i.e., taking half of the sum of the maximum and minimum values ​​of the value coefficient and turnover risk prediction value of all employees), the matrix is ​​precisely divided into four strategic quadrants. Each employee is mapped to the corresponding quadrant according to their quantitative indicators: high value low turnover risk type (high value / low risk), high value high turnover risk type (high value / high risk), low value low turnover risk type (low value / low risk), and low value high turnover risk type (low value / high risk), thereby achieving automated and precise classification of talent. This step establishes a visualized talent strategy map, transforming complex talent assessment results into intuitive graphical displays, enabling managers to quickly grasp the overall distribution of talent within the organization; it forms a dynamic and adaptive classification mechanism, ensuring that classification standards automatically adjust as the organization's talent structure changes through real-time updated intermediate thresholds, avoiding biases caused by subjectively set thresholds; and it constructs a precise foundation for talent management, providing clear target groups for the implementation of subsequent differentiated human resource management strategies through scientific matrix positioning, making resource allocation more focused and greatly improving the pertinence and effectiveness of talent management work.

[0076] In the embodiments of this application, a big data-based human resource management method is provided. The method involves setting corresponding management strategies for different employee types, including: for employees in the high-value, low-turnover-risk category, the management strategy includes providing market-leading compensation, bonuses, and long-term equity incentives; granting them strategic projects, innovative challenges, and decision-making autonomy; designating them as key successors; and providing senior management mentorship. For employees in the high-value, high-turnover-risk category, the management strategy includes having superiors conduct interviews with HR; providing personalized solutions for the main reasons for departure; and initiating a fast-track compensation process or offering clear promotion opportunities. Promotion / Project Commitment; For employees in the low-value, low-turnover-risk category, the management strategy is to provide targeted training to bridge skill gaps, set more challenging goals, and strengthen process coaching and feedback, maintaining their work motivation through timely praise and small incentives; For employees in the low-value, high-turnover-risk category, the management strategy is to implement time-limited performance improvement plans, clarify improvement goals and support, conduct high-frequency checks and coaching, and document the progress. If there is no significant improvement during the improvement period, the optimization or negotiated resignation process will be initiated decisively; Based on the management strategies set for each employee type, employees in each employee type will be managed uniformly.

[0077] like Figure 2As shown in the embodiments of this application, a big data-based human resource management system is provided, including: an acquisition module, used to acquire historical multidimensional information data of employees in the company, and analyze the historical multidimensional information data to determine the influencing factors related to employee turnover; a prediction module, used to extract data features of the influencing factors, and construct a turnover risk prediction model based on the data features and a preset neural network model to predict the future turnover risk of employees; an evaluation module, used to analyze and evaluate the historical multidimensional information data of employees based on preset value indicators to obtain a comprehensive evaluation value of each preset value indicator; a calculation module, used to comprehensively analyze and calculate the overall value of employees based on the comprehensive evaluation value of each preset value indicator to obtain the value coefficient value of each employee; and a management module, used to classify employees based on the turnover risk prediction value and value coefficient value of each employee to obtain multiple employee types, and set corresponding management strategies for each employee type.

[0078] In summary, this invention provides a big data-based human resource management method and system, comprising: acquiring and analyzing employees' historical multidimensional information data to determine influencing factors related to employee turnover; extracting data features of the influencing factors and constructing a turnover risk prediction model based on these features and a preset model to predict turnover risk values; analyzing and evaluating employees' historical multidimensional information data based on preset value indicators to obtain a comprehensive evaluation value for each preset value indicator, and comprehensively analyzing and calculating the overall value of employees based on this evaluation value to obtain a value coefficient value for each employee; classifying employees based on these two values ​​to obtain multiple employee types, and setting corresponding management strategies for each type. This invention, by combining turnover risk prediction values ​​with value coefficients, accurately classifies employees, achieving visualization and refinement of employee inventory, providing personalized management strategies for different employees, and improving the overall management efficiency of the talent pool.

[0079] Finally, it should be noted that those skilled in the art can obviously make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims and their equivalents, this invention also intends to include these modifications and variations.

[0080] The above description is merely one embodiment of the present invention, and should not be construed as limiting the scope of the invention. Any structural changes made based on the present invention, as long as they do not depart from the essence of the invention, should be considered as falling within the protection scope of the present invention and subject to its restrictions. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the platform described above can be referred to the corresponding processes in the foregoing platform embodiments, and will not be repeated here.

[0081] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, platform, article, or device / platform that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to those processes, platforms, articles, or devices / platforms.

[0082] The technical solutions of the present invention have been described in conjunction with the accompanying drawings and further embodiments. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to closely related technical features, and the technical solutions resulting from such changes or substitutions will all fall within the scope of protection of the present invention.

[0083] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. A human resource management method based on big data, characterized in that, include: Obtain historical multidimensional information data of employees in the company, and analyze the historical multidimensional information data to identify the influencing factors related to employee turnover; Extract the data features of influencing factors, and build a turnover risk prediction model based on the data features and a preset neural network model to predict the future turnover risk of employees. Based on preset value indicators, the historical multidimensional information data of employees are analyzed and evaluated to obtain the comprehensive evaluation value of each preset value indicator. Based on the comprehensive evaluation value of each preset value indicator, the overall value of employees is comprehensively analyzed and calculated to obtain the value coefficient value of each employee. Employees are categorized based on their predicted turnover risk and value coefficient, resulting in multiple employee types. Corresponding management strategies are then set for each employee type.

2. The human resource management method based on big data according to claim 1, characterized in that, The process involves acquiring historical multidimensional information data about employees within the company, analyzing this data, and identifying influencing factors related to employee turnover, including: Acquire historical multidimensional information data of employees in the company, and divide the historical multidimensional information data into multiple information data groups according to information type; Identify employee turnover data and calculate the correlation between each data group and the turnover data. Information data groups with a relevance greater than a preset threshold are identified as influencing factors related to employee turnover.

3. The human resource management method based on big data according to claim 2, characterized in that, The process involves extracting data features of influencing factors and constructing a turnover risk prediction model based on these data features and a pre-defined neural network model to predict future turnover risk values ​​for employees. This includes: Extract the data features of the information data group corresponding to the influencing factors, construct a dataset based on the historical information data group and the corresponding data features, and input the dataset into the preset neural network model to construct the initial model for predicting the risk of turnover. The dataset is divided into training and testing sets according to a preset ratio, and the training and testing sets are input into the initial model for predicting turnover risk. The initial model for predicting employee turnover risk is trained and tested until it meets the preset convergence conditions, thus obtaining the employee turnover risk prediction model. The system acquires current multidimensional information data of employees and inputs it into the turnover risk prediction model to make predictions and obtain the predicted value of employee turnover risk.

4. The human resource management method based on big data according to claim 3, characterized in that, The analysis and evaluation of employees' historical multidimensional information data based on preset value indicators yields a comprehensive evaluation value for each preset value indicator, including: Determine the preset value indicators and identify the corresponding data from the employee's historical multidimensional information data; The data is input into the corresponding evaluation model, and the evaluation model outputs the evaluation values ​​of each preset value indicator. Determine the preset evaluation coefficients corresponding to each preset value indicator, and calculate the comprehensive evaluation value of each preset value indicator based on the preset evaluation coefficients and evaluation values ​​corresponding to each preset value indicator.

5. The human resource management method based on big data according to claim 4, characterized in that, The formula for calculating the comprehensive evaluation value of the preset value index is as follows: Di = fi * Pi, Where Di is the comprehensive evaluation value of the i-th preset value indicator, fi is the preset evaluation coefficient of the i-th preset value indicator, and Pi is the evaluation value of the i-th preset value indicator.

6. The human resource management method based on big data according to claim 4, characterized in that, The overall value of employees is comprehensively analyzed and calculated based on the comprehensive evaluation values ​​of each preset value indicator, resulting in a value coefficient value for each employee, including: Determine the preset value corresponding to each preset value indicator, and calculate the difference between the comprehensive evaluation value of each preset value indicator and the corresponding preset value to obtain the difference value of each preset value indicator. A preset weight-difference interval correspondence is set in advance. For each difference interval, a corresponding preset weight is associated with it. Obtain the difference value of each preset value indicator, and based on the mapping relationship between the difference value interval to which the difference value belongs and the preset weight-difference value interval correspondence, select the preset weight corresponding to the difference value interval to determine the weight of each preset value indicator. The value coefficient of each employee is calculated based on the weights and comprehensive evaluation values ​​corresponding to each preset value indicator.

7. A human resource management method based on big data according to claim 6, characterized in that, The formula for calculating the employee's value coefficient is as follows: Where Z is the difference, k is the preset conversion coefficient, αi is the weight of the i-th preset value indicator, and Di is the comprehensive evaluation value of the i-th preset value indicator.

8. A human resource management method based on big data according to claim 6, characterized in that, The employees are categorized based on their predicted turnover risk and value coefficient, resulting in multiple employee types, including: A matrix is ​​constructed based on the predicted turnover risk value and value coefficient value of each employee. The horizontal axis of the matrix is ​​the value coefficient value from low to high, and the vertical axis is the predicted turnover risk value from low to high. Determine the median value between the predicted turnover risk value and the value coefficient value for each employee. The median value is the sum of the highest and lowest values, divided by 2. The matrix is ​​divided into four quadrants based on the median value of the predicted turnover risk and the median value of the value coefficient. Employees in the same quadrant are classified into the same employee type, resulting in multiple employee types, including high-value low turnover risk, high-value high turnover risk, low-value low turnover risk, and low-value high turnover risk.

9. A human resource management method based on big data according to claim 8, characterized in that, The management of employees by setting corresponding management strategies for each type includes: For employees in the high-value, low-turnover-risk category, the management strategy is to provide market-leading compensation, bonuses, and long-term equity incentives, grant them strategic projects, innovative challenges, and decision-making autonomy, designate them as successors for key positions, and provide mentorship from senior executives. For employees with high value and high turnover risk, the management strategy is to have a meeting between the supervisor and HR to provide personalized solutions for the main reasons for leaving, and to activate a fast track for compensation or give clear promotion / project commitments. For employees in the low-value, low-turnover-risk category, the management strategy is to provide targeted training to bridge skills gaps, set more challenging goals, strengthen process coaching and feedback, and maintain their work motivation through timely praise and small incentives. For employees in the low-value, high-turnover-risk category, the management strategy is to implement time-limited performance improvement plans, clarify improvement goals and support, conduct frequent checks and coaching, and document the progress. If there is no significant improvement during the improvement period, the optimization or negotiated resignation process will be initiated decisively. The management strategy based on the employee type is used to manage employees in the same way in each employee type.

10. A human resource management system based on big data, characterized in that, include: The acquisition module is used to acquire historical multidimensional information data of employees in the company, and to analyze the historical multidimensional information data to determine the influencing factors related to employee turnover. The prediction module is used to extract the data features of influencing factors and build a turnover risk prediction model based on the data features and a preset neural network model to make predictions and obtain the predicted value of the employee's future turnover risk. The evaluation module is used to analyze and evaluate employees' historical multidimensional information data based on preset value indicators, and obtain the comprehensive evaluation value of each preset value indicator. The calculation module is used to perform a comprehensive analysis and calculation of the overall value of employees based on the comprehensive evaluation value of each preset value indicator, and to obtain the value coefficient value of each employee. The management module is used to classify employees based on their predicted turnover risk and value coefficient, resulting in multiple employee types, and to set corresponding management strategies for each employee type.

Images