Employee salary management system based on cloud computing

Abstract

The invention provides an employee salary management system based on cloud computing, and the system comprises a cloud collection module which is responsible for collecting employee work, personal information and market salary data; the self-adaptive salary calculation module calculates salary according to the working data and the welfare system based on a rule engine, and checks and encrypts the salary; the employee demission prediction module predicts the demission probability and time by combining work, salary, personal information and market data, and generates an early warning signal and a demission reason; and the decision module calculates a risk rating according to the early warning signal and the demission reason, formulates an intervention strategy or a management scheme, and executes the intervention strategy or the management scheme. Through cloud computing, machine learning, a dynamic rule engine and an automatic decision-making module, a data-driven closed-loop management system is constructed, real-time data convergence, salary strategy second-level updating, remission risk modeling and cost-benefit analysis are realized, and the accuracy, automation and response speed of salary management are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automated salary management, and in particular to an employee salary management system based on cloud computing. Background Art

[0002] In traditional enterprise payroll management scenarios, salary calculation and employee retention management have long faced technical bottlenecks. Existing payroll management systems often utilize localized deployment architectures, relying on manual collection of attendance and performance data and calculations based on pre-set formulas. This lack of automation means that regulatory adjustments (such as tax law changes and benefit policy updates) require manual code modifications, resulting in response times of several weeks. Data validation processes only perform basic format verification and are unable to perform real-time verification of complex logic such as overtime pay caps and social security base compliance, resulting in an average salary calculation error rate of 3.2%. Regarding employee retention management, traditional systems only record employee resignation events and lack predictive capabilities. Companies must manually analyze employee resignation reasons, leading to development cycles exceeding one month for intervention strategies. Furthermore, intervention measures (such as salary increases and training) lack cost-benefit evaluation, resulting in 65% of retention investments failing to generate expected returns. These technical shortcomings directly hinder the accuracy of enterprise payroll management and the efficiency of human resource allocation, necessitating a data-driven, intelligent transformation through technologies such as cloud computing and machine learning. Summary of the Invention

[0003] The present invention aims to at least solve the technical problem of low intelligence level of employee salary management systems in the prior art, and particularly innovatively proposes an employee salary management system based on cloud computing.

[0004] In order to achieve the above-mentioned object of the present invention, the present invention provides an employee salary management system based on cloud computing, the system comprising: Cloud collection module, used to collect employees' work data, personal information and market salary data; The adaptive salary calculation module is connected to the collection module and is used to use the salary calculation model based on the rule engine to adaptively calculate the employee's salary according to the employee's work data and the company's welfare system, verify the employee's salary data, and encrypt the verified salary data; An employee turnover prediction module, connected to the acquisition module and the salary calculation module, is used to predict the probability and time of employee turnover based on the employee's work data, encrypted salary data, personal information, and market salary data, and to generate a turnover warning signal and all possible reasons for turnover; The decision-making module is connected to the employee turnover prediction module and the salary calculation module. It is used to calculate the turnover risk rating based on the turnover warning signal and all possible reasons for turnover, generate an intervention strategy or turnover management plan based on the turnover risk rating results, and the decision-making module determines whether to implement the intervention strategy or turnover management plan.

[0005] As an optional embodiment of the present invention, optionally, the employee turnover prediction module includes: A feature extraction unit is used to extract turnover-related features from employees' work data, encrypted salary data, personal information, and market salary data using a feature extraction algorithm. The features include salary competitiveness, performance stability, career growth stagnation, salary satisfaction, external market attractiveness, and turnover history correlation features. A prediction model unit, connected to the feature extraction unit, is used to predict the employee's resignation probability and resignation time using a prediction model based on the extracted resignation-related features; The resignation reason analysis unit is connected to the prediction model unit and the feature extraction unit, and is used to analyze the employee's resignation reasons based on resignation-related features, resignation probability and resignation time, and generate a list of all possible resignation reasons and evaluate each resignation reason.

[0006] As an optional embodiment of the present invention, optionally, the decision module includes: The risk grading unit is used to classify employee turnover risks into different levels based on the turnover probability and turnover time predicted by the employee turnover prediction module and the risk threshold set by the enterprise; Intervention strategy library, used to store multiple intervention strategies for different turnover risk levels; The strategy recommendation unit is connected to the risk grading unit and the intervention strategy library, and is used to recommend corresponding intervention strategies or separation management plans in the intervention strategy library based on the risk grading results; The cost-effectiveness analysis unit, connected to the strategy recommendation unit, is used to calculate the cost of the intervention strategy and the potential benefits of the turnover management program; The strategy execution unit is connected to the cost-benefit analysis unit and is used to determine whether to implement the recommended intervention strategy or separation management plan based on potential benefits; if the strategy execution unit determines that the potential benefits of implementing the intervention strategy are greater than the costs, the intervention strategy is implemented; if the strategy execution unit determines that the potential benefits of implementing the intervention strategy are less than or equal to the costs, the separation management plan is implemented.

[0007] As an optional embodiment of the present invention, optionally, the adaptive salary calculation module includes: A pre-processing unit, used for pre-processing the work data of employees collected by the cloud collection module; The rule engine parsing unit is used to load the enterprise's salary rules in real time and convert the salary rules into executable calculation logic. The rule engine parsing unit is also used to adaptively adjust the parameters in the salary rules according to the intervention strategy executed by the decision module. The salary calculation execution unit is connected to the pre-processing unit and the rule engine parsing unit, and is used to calculate the employee's salary based on the employee's work data and executable calculation logic; a multi-dimensional verification unit, connected to the salary calculation execution unit, for verifying the salary calculation result using multi-dimensional data verification rules; The dynamic encryption unit is connected to the multi-dimensional verification unit and is used to encrypt, store and transmit the verified employee's salary.

[0008] As an optional embodiment of the present invention, the system optionally further includes a user feedback module for collecting employee satisfaction feedback on salary, and transmitting the feedback data to the decision module in real time for the execution module to generate intervention strategies or resignation management plans.

[0009] Beneficial effects of the present invention: The present invention integrates cloud computing, machine learning prediction models, dynamic rule engines and automated decision-making modules to build a data-driven closed-loop management system: the cloud acquisition module is used to realize the real-time aggregation of multi-source heterogeneous data, and the rule engine of the adaptive salary calculation module supports the second-level update of salary strategies. Combined with the time series feature extraction and cyclical risk modeling of the employee turnover prediction module, and the cost-benefit quantitative analysis capabilities of the decision-making module, it finally forms a complete link from data acquisition, intelligent prediction to automated strategy execution, which significantly improves the accuracy, automation and response speed of salary management.

[0010] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which: Figure 1 It is a structural diagram of a cloud computing-based employee salary management system of the present invention. DETAILED DESCRIPTION

[0012] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0013] like Figure 1 As shown, a cloud computing-based employee salary management system includes: Cloud collection module, used to collect employees' work data, personal information and market salary data; In this embodiment, the cloud acquisition module is specifically an integrated data acquisition platform, which can capture relevant information of employees from multiple data sources in real time and accurately. First, by connecting with the company's internal human resources information system (HRIS), the employee's personal basic information, work performance, performance evaluation and other data are automatically synchronized. Secondly, the API interface or crawler technology is used to obtain market salary data from external recruitment websites, industry reports and market salary surveys. These data cover key information such as the average salary level of the same position in the same industry, salary change trends, etc. In addition, the platform also has data cleaning and preprocessing functions, which can automatically identify and filter out invalid, redundant or abnormal data to ensure the accuracy and completeness of the data. Through the integrated data acquisition platform, the present invention realizes the comprehensive and real-time collection of employee work data, personal information and market salary data.

[0014] The adaptive salary calculation module is connected to the collection module and is used to use the salary calculation model based on the rule engine to adaptively calculate the employee's salary according to the employee's work data and the company's welfare system, verify the employee's salary data, and encrypt the verified salary data; In this embodiment, the adaptive salary calculation module includes a preprocessing unit, a rule engine parsing unit, a salary calculation execution unit, a multi-dimensional verification unit, and a dynamic encryption unit. The preprocessing unit first preprocesses the employee work data collected by the cloud collection module, including data cleaning and format conversion, to ensure data quality and consistency. The rule engine parsing unit loads the company's salary rules in real time. These rules include terms such as base salary, performance bonuses, and benefits, and the rule engine converts them into executable calculation logic. Furthermore, the rule engine parsing unit also has adaptive adjustment capabilities, dynamically adjusting the parameters in the salary rules based on the intervention strategy executed by the decision module to achieve flexible adjustment of salary policies. The salary calculation execution unit automatically calculates employee salaries based on the preprocessed employee work data and executable calculation logic. The multi-dimensional verification unit verifies the salary calculation results using pre-set multi-dimensional data verification rules to ensure accuracy and rationality. These multi-dimensional data verification rules include, but are not limited to, the correctness of the salary calculation formula, the consistency of the calculation results with expectations, and the compliance of the salary data. Finally, the dynamic encryption unit encrypts the verified salary data to ensure data security and privacy. The encrypted salary data will be securely stored and transmitted.

[0015] An employee turnover prediction module, connected to the acquisition module and the salary calculation module, is used to predict the probability and time of employee turnover based on the employee's work data, encrypted salary data, personal information, and market salary data, and to generate a turnover warning signal and all possible reasons for turnover; In this embodiment, the employee turnover prediction module includes a feature extraction unit, a prediction model unit, and a turnover reason analysis unit. The feature extraction unit first extracts features from the employee's work data, encrypted salary data, personal information, and market salary data. These features include employee salary competitiveness features, employee performance stability features, employee career growth stagnation features, employee salary satisfaction features, employee external market attractiveness features, and employee turnover history correlation features. The prediction model unit uses a machine learning algorithm to build a prediction model, inputs the extracted features, and outputs the employee's turnover probability and predicted turnover time. The turnover reason analysis unit analyzes and generates all possible reasons for turnover based on the prediction results, combined with historical turnover data and expert knowledge. The employee turnover prediction module achieves accurate prediction of employee turnover risks by comprehensively considering multiple factors.

[0016] The decision-making module is connected to the employee turnover prediction module and the salary calculation module. It is used to calculate the turnover risk rating based on the turnover warning signal and all possible reasons for turnover, generate an intervention strategy or turnover management plan based on the turnover risk rating results, and the decision-making module determines whether to implement the intervention strategy or turnover management plan.

[0017] In this embodiment, the decision-making module includes a risk grading unit, an intervention strategy library, a strategy recommendation unit, a cost-benefit analysis unit, and a strategy execution unit. The risk grading unit classifies employee turnover risk into different levels, such as low risk, medium risk, and high risk, based on the turnover probability and predicted turnover time output by the employee turnover prediction module, combined with the company's pre-set risk thresholds. The intervention strategy library stores a variety of intervention strategies for different turnover risk levels, including providing promotion opportunities, increasing salary and benefits, and strengthening employee care. Based on the risk grading results, the strategy recommendation unit recommends corresponding intervention strategies or turnover management plans from the intervention strategy library to company managers. The cost-benefit analysis unit conducts a cost-benefit analysis on the recommended intervention strategies, evaluating their economic feasibility and potential benefits. Finally, based on the results of the cost-benefit analysis, the strategy execution unit determines whether to implement the recommended intervention strategy or turnover management plan and performs the corresponding operations to effectively control risks and reduce employee turnover rates.

[0018] The cloud-based employee salary management system in this embodiment works as follows: First, the cloud collection module aggregates employee work data, personal information, and market salary data in real time. This data provides a solid foundation for subsequent salary calculation and turnover prediction. The adaptive salary calculation module dynamically calculates each employee's salary based on their work performance, the company's benefits system, and real-time market salary levels using a rules-based salary calculation model. This process not only ensures fair and competitive compensation but also significantly improves calculation efficiency and accuracy. The employee turnover prediction module analyzes and mines the collected multi-dimensional data, using advanced machine learning algorithms to build a predictive model that accurately predicts the probability and time of employee turnover. This function provides companies with valuable forward-looking information, enabling them to take proactive measures to reduce employee turnover and maintain team stability and cohesion. The decision-making module, based on turnover warning signals and turnover reasons analysis, combines the company's actual situation and risk tolerance to develop targeted intervention strategies or turnover management plans. Through cost-benefit analysis, companies can select the most economically feasible and profitable plan, effectively controlling risks and further reducing employee turnover.

[0019] In summary, the cloud-based employee payroll management system of this invention builds a data-driven, closed-loop management system through integrated data collection, intelligent salary calculation and turnover prediction, and automated decision-making and execution. This system not only improves the accuracy and automation of payroll management, but also significantly enhances the company's responsiveness and risk management capabilities.

[0020] As an optional embodiment of the present invention, optionally, the employee turnover prediction module includes: A feature extraction unit is used to extract turnover-related features from employees' work data, encrypted salary data, personal information, and market salary data using a feature extraction algorithm. The features include salary competitiveness, performance stability, career growth stagnation, salary satisfaction, external market attractiveness, and turnover history correlation features. A prediction model unit, connected to the feature extraction unit, is used to predict the employee's resignation probability and resignation time using a prediction model based on the extracted resignation-related features; The resignation reason analysis unit is connected to the prediction model unit and the feature extraction unit, and is used to analyze the employee's resignation reasons based on resignation-related features, resignation probability and resignation time, and generate a list of all possible resignation reasons and evaluate each resignation reason.

[0021] In this embodiment, the resignation reason analysis unit further includes a resignation reason weight assignment subunit and a resignation reason priority ranking subunit. The resignation reason weight assignment subunit assigns a weight to each resignation reason based on historical resignation data and expert knowledge. The weight reflects the degree of influence of the resignation reason on the employee's resignation decision. The resignation reason priority ranking subunit prioritizes all possible resignation reasons based on the assigned weights and generates a resignation reason priority list. In this way, enterprise managers can more intuitively understand which resignation reasons have the greatest impact on employee resignation, thereby formulating targeted intervention strategies or resignation management plans to improve the effectiveness and relevance of the measures.

[0022] As an optional embodiment of the present invention, optionally, the expression of the feature extraction algorithm is: ; ; ; ; ; ; in, Indicates employees The salary competitiveness characteristic is used to measure the competitiveness of employee salaries relative to the market. The higher the value, the stronger the salary competitiveness. Indicates employees Current salary, Indicates the market average salary for the same position in the same industry. represents the standard deviation of market salary data; Indicates employees The performance stability characteristic is used to reflect the stability of employee performance. The closer the value is to 1, the more stable the performance. Indicates the length of the time window, represents the normalization function, and They all represent the parameters of the normalization function, Indicates employees In time performance ratings; Indicates employees The career growth stagnation characteristic is used to measure the degree of career advancement stagnation. The closer the value is to 1, the longer the stagnation time. Indicates employees Time since last promotion, Indicates the preset promotion stagnation threshold; It represents the salary satisfaction characteristic, which is used to comprehensively measure the satisfaction with salary and benefits. The higher the value, the more satisfied the employee is with the total compensation. Indicates welfare, Indicates the The weight coefficient of each welfare item, Indicates employees No. Item benefit costs; Indicates employees The external market attractiveness characteristic is used to reflect the attractiveness of the market salary gap to employees. The higher the value, the stronger the external opportunity. Indicates employees The average salary level of the same position in the same industry in the market, represents the adjustment function based on years of work experience; Represents the resignation history correlation feature, which is used to evaluate the correlation with the team's turnover rate. The higher the value, the worse the team stability. Express with employees A collection of colleagues with direct collaborative relationships, represents the indicator function, Indicates colleagues Length of service, Indicates the turnover risk threshold.

[0023] As an optional embodiment of the present invention, optionally, the expression for predicting the probability of employee resignation by the prediction model is: in, Indicates employees In time The cumulative probability of leaving within represents the exponential function, Indicates time The baseline hazard function within represents the transpose of the feature weight vector; Indicates employees In time The characteristic vector of includes salary competitiveness, performance stability, career growth stagnation, salary satisfaction, external market attractiveness and turnover history correlation; Indicates the total number of cycles; Indicates the The intensity coefficient of a cyclical effect, such as quarterly turnover fluctuations; represents a periodic function used to capture seasonal fluctuations in turnover trends; Represents the periodic effect period.

[0024] As an optional embodiment of the present invention, optionally, the expression for predicting employee resignation time by the prediction model is: ; ; in, Indicates employees The predicted time of leaving the company, represents the time variable, Indicates employees In time The cumulative probability of leaving within represents the risk growth rate threshold, and express Parameters, Indicates time The baseline hazard function within represents the transpose of the feature weight vector, Indicates employees In time The eigenvector of Indicates the total number of cycles, Indicates the The intensity coefficient of the periodic effect, represents a periodic function, Represents the periodic effect period.

[0025] As an optional embodiment of the present invention, optionally, the employee turnover prediction module further includes a model optimization unit, the model optimization unit and the prediction model unit being configured to optimize the prediction model using an optimization algorithm; The expression of the optimization algorithm is: in, represents the log-likelihood function, represents the feature weight vector, and express Parameters, Indicates time The baseline hazard function within represents the intensity coefficient of the periodic effect, Indicates the total number of employees; represents the indicator variable of resignation event, where 1 represents resignation and 0 represents current employment; Indicates employees exist The risk of instant resignation at any time, Indicates employees Time of leaving the company, Indicates employees At the time of leaving The eigenvector of Indicates employees From onboarding to time The cumulative risk of leaving the company.

[0026] As an optional embodiment of the present invention, optionally, the decision module includes: The risk grading unit is used to classify employee turnover risks into different levels based on the turnover probability and turnover time predicted by the employee turnover prediction module and the risk threshold set by the enterprise; In this embodiment, the risk grading unit formulates targeted intervention measures or turnover management plans based on the level of employee turnover risk. For example, for employees with a high turnover risk, an emergency retention plan can be implemented, including providing promotion opportunities, salary increases or additional benefits, etc., to enhance employees' sense of belonging and loyalty. For employees with a medium turnover risk, communication with them can be strengthened to understand their career plans and needs, and necessary training and development opportunities can be provided to improve their job satisfaction and stability. For employees with a low turnover risk, their work performance and development dynamics can be continuously monitored, and employees can be encouraged to participate in corporate activities to enhance team cohesion. Through such risk grading and differentiated management, enterprises can more effectively identify and manage employee turnover risks, improve the stability of human resources and the competitiveness of the enterprise.

[0027] Intervention strategy library, used to store multiple intervention strategies for different turnover risk levels; In this embodiment, the intervention strategy library establishes the following table: By establishing this intervention strategy library, companies can quickly identify and implement appropriate intervention measures for employees at varying turnover risks. These strategies not only cover multiple levels, from routine care to executive involvement, but also clearly define key actions and responsible individuals, ensuring the effectiveness and timeliness of interventions. For example, for employees at critical risk, the HRVP's ultimate retention measures demonstrate the company's utmost sincerity and commitment to retaining key talent.

[0028] The strategy recommendation unit is connected to the risk grading unit and the intervention strategy library, and is used to recommend corresponding intervention strategies or separation management plans in the intervention strategy library based on the risk grading results; In this embodiment, the strategy recommendation unit is specifically an intelligent recommendation engine, which automatically retrieves and recommends the most appropriate intervention strategy or resignation management plan from the intervention strategy library based on the employee resignation risk level output by the risk grading unit. The design of the recommendation engine takes into account the effectiveness, feasibility and cost-effectiveness of the strategy, ensuring that the company can obtain the best resignation risk management effect with the most reasonable resource investment. For example, when an employee is identified as a high-risk resignation group, the recommendation engine will automatically recommend salary and benefits optimization strategies, including providing more competitive salaries, flexible benefits plans or performance bonus adjustments, in order to improve employee satisfaction and loyalty and reduce the risk of resignation. At the same time, the recommendation engine will also fine-tune the recommended strategy according to the actual situation of the company and the personalized needs of employees to achieve more accurate and effective intervention.

[0029] The cost-effectiveness analysis unit, connected to the strategy recommendation unit, is used to calculate the cost of the intervention strategy and the potential benefits of the turnover management program; In this embodiment, the cost-benefit analysis unit specifically calculates the implementation cost of each intervention strategy, and combines historical data and prediction models to evaluate the potential impact of different resignation management programs on employee retention, work efficiency, and overall corporate performance. Through quantitative analysis, the cost-benefit analysis unit can provide companies with intuitive cost-benefit comparisons and cost-benefit ratios of different resignation management programs, helping corporate decision makers to choose the optimal program more scientifically. For example, for high-risk employees who leave, although the implementation cost of the salary and benefits optimization strategy may be high, by analyzing the potential benefits such as increased employee retention, increased work efficiency, and improved overall corporate performance, decision makers can weigh the pros and cons and make more informed decisions. In addition, the cost-benefit analysis unit can also dynamically adjust the intervention strategy based on the company's financial situation and human resource planning to ensure the continuity and effectiveness of resignation risk management.

[0030] The strategy execution unit is connected to the cost-benefit analysis unit and is used to determine whether to implement the recommended intervention strategy or separation management plan based on potential benefits; if the strategy execution unit determines that the potential benefits of implementing the intervention strategy are greater than the costs, the intervention strategy is implemented; if the strategy execution unit determines that the potential benefits of implementing the intervention strategy are less than or equal to the costs, the separation management plan is implemented.

[0031] It's important to note that the specific policy execution unit automatically executes the selected intervention strategy or departure management plan. For example, if the policy execution unit determines that implementing a compensation and benefits optimization strategy for a high-risk employee has high potential benefits, it will automatically trigger the compensation system to adjust the employee's salary structure or benefits plan. This process not only reduces the time delay associated with manual decision-making but also ensures that intervention measures are quickly and accurately applied to the target employee group.

[0032] As an optional embodiment of the present invention, optionally, the expression for calculating the cost of the intervention strategy and the potential benefit of the turnover management solution in the cost-benefit analysis unit is: in, represents the net present value of potential benefits, represents the policy evaluation cycle, Indicates that employees The expected contribution value of the year, represents the discount rate, represents the total cost of the intervention strategy, Represents the replacement cost after an employee leaves the company.

[0033] when When it is greater than zero, the intervention strategy is executed. When it is less than zero, the resignation management plan will be implemented.

[0034] As an optional embodiment of the present invention, optionally, the adaptive salary calculation module includes: A pre-processing unit, used for pre-processing the work data of employees collected by the cloud collection module; Preprocessing includes data cleaning, removing outliers and missing values ​​to ensure data accuracy and completeness; data standardization, converting data of different dimensions into a unified scale; The rule engine parsing unit is used to load the enterprise's salary rules in real time and convert the salary rules into executable calculation logic. The rule engine parsing unit is also used to adaptively adjust the parameters in the salary rules according to the intervention strategy executed by the decision module. In this embodiment, the rule engine parsing unit has two functions. The first is to load the company's salary rules in real time to ensure the accuracy and compliance of salary calculations. These rules include the calculation method of basic salary, the distribution standard of performance bonuses, the conditions for the issuance of welfare projects, etc. Through real-time loading, the system can automatically apply the latest salary rules to avoid errors and delays caused by manual adjustments. The second is to adaptively adjust the parameters in the salary rules according to the intervention strategy executed by the decision module. For example, when the decision module recommends implementing a salary and benefits optimization strategy for high-risk employees who leave, the rule engine parsing unit will automatically adjust the employee's salary structure or welfare plan parameters to match the recommended intervention strategy. This adaptive adjustment capability not only improves the flexibility of salary management, but also ensures that intervention measures can be quickly and accurately applied to the target employee group, thereby effectively reducing the risk of leaving and improving employee satisfaction and loyalty.

[0035] The salary calculation execution unit is connected to the pre-processing unit and the rule engine parsing unit, and is used to calculate the employee's salary based on the employee's work data and executable calculation logic; In this embodiment, the salary calculation execution unit can process the employee's salary calculation tasks in real time to ensure the timeliness and accuracy of salary payment. It automatically calculates the employee's basic salary, performance bonus, welfare items and other salary components based on the data provided by the preprocessing unit and the executable calculation logic converted by the rule engine parsing unit. Employees can view their own salary details in real time through the system, which enhances their trust and satisfaction with salary payment. At the same time, the salary calculation execution unit can also flexibly respond to changes in the company's salary strategy based on the salary rules adaptively adjusted by the rule engine parsing unit to ensure the compliance and competitiveness of salary calculation. For example, when an enterprise implements a salary and benefits optimization strategy for high-risk employees who leave, the salary calculation execution unit will automatically adjust the employee's salary calculation logic to match the new salary rules, thereby achieving effective management of employee turnover risks and optimization of salary incentives.

[0036] a multi-dimensional verification unit, connected to the salary calculation execution unit, for verifying the salary calculation result using multi-dimensional data verification rules; The dynamic encryption unit is connected to the multi-dimensional verification unit and is used to encrypt, store and transmit the verified employee's salary.

[0037] As an optional embodiment of the present invention, the system optionally further includes a user feedback module for collecting employee satisfaction feedback on salary, and transmitting the feedback data to the decision module in real time for the execution module to generate intervention strategies or resignation management plans.

[0038] In this embodiment, the user feedback module is mainly used to collect employees' satisfaction and opinions on salary, benefits, career development, etc. Through regular satisfaction surveys, online feedback channels or employee seminars, the system can collect employee feedback data in real time and transmit it to the decision-making module. Based on this feedback data, the decision-making module can more accurately identify employees' needs and expectations, thereby generating intervention strategies or resignation management plans that are more in line with employees' actual needs. For example, when employees generally express dissatisfaction with their salary and benefits, the decision-making module can promptly adjust the salary and benefits strategy to improve employee satisfaction and loyalty. Or when employees express concerns about career development opportunities, the decision-making module can recommend providing more training and development opportunities to improve employee job satisfaction and stability. Through such a feedback mechanism, companies can establish closer ties with employees, promptly identify and resolve potential problems, thereby effectively reducing the risk of resignation and enhancing the company's competitiveness.

[0039] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and alterations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

Claims

1. A cloud computing-based employee salary management system, characterized in that: The system comprises: Cloud collection module, used to collect employees' work data, personal information and market salary data; The adaptive salary calculation module is connected to the collection module and is used to use the salary calculation model based on the rule engine to adaptively calculate the employee's salary according to the employee's work data and the company's welfare system, verify the employee's salary data, and encrypt the verified salary data; An employee turnover prediction module, connected to the acquisition module and the salary calculation module, is used to predict the probability and time of employee turnover based on the employee's work data, encrypted salary data, personal information, and market salary data, and to generate a turnover warning signal and all possible reasons for turnover; The decision-making module is connected to the employee turnover prediction module and the salary calculation module. It is used to calculate the turnover risk rating based on the turnover warning signal and all possible reasons for turnover, generate an intervention strategy or turnover management plan based on the turnover risk rating results, and the decision-making module determines whether to implement the intervention strategy or turnover management plan.

2. The cloud computing-based employee salary management system according to claim 1, wherein: The employee turnover prediction module includes: A feature extraction unit is used to extract turnover-related features from employees' work data, encrypted salary data, personal information, and market salary data using a feature extraction algorithm. The features include salary competitiveness, performance stability, career growth stagnation, salary satisfaction, external market attractiveness, and turnover history correlation features. A prediction model unit, connected to the feature extraction unit, is used to predict the employee's resignation probability and resignation time using a prediction model based on the extracted resignation-related features; The resignation reason analysis unit is connected to the prediction model unit and the feature extraction unit, and is used to analyze the employee's resignation reasons based on resignation-related features, resignation probability and resignation time, and generate a list of all possible resignation reasons and evaluate each resignation reason.

3. The cloud computing-based employee salary management system according to claim 2, characterized in that: The expression of the feature extraction algorithm is: ； ； ； ； ； ； in, Indicates employees Salary competitiveness characteristics, Indicates employees Current salary, Indicates the market average salary for the same position in the same industry. represents the standard deviation of market salary data, Indicates employees The performance stability characteristics of Indicates the length of the time window, represents the normalization function, and They all represent the parameters of the normalization function, Indicates employees In time performance ratings, Indicates employees Characteristics of career stagnation, Indicates employees Time since last promotion, Indicates the preset promotion stagnation threshold, Indicates salary satisfaction characteristics, Indicates welfare, Indicates the The weight coefficient of each welfare item, Indicates employees No. Welfare costs, Indicates employees External market attractiveness characteristics, Indicates employees The average salary level of the same position in the same industry in the market, represents the adjustment function based on working years, Represents the resignation history correlation feature, Express with employees A collection of colleagues with direct collaborative relationships, represents the indicator function, Indicates colleagues Length of service, Indicates the turnover risk threshold.

4. The cloud computing-based employee salary management system according to claim 2, wherein: The expression for the probability of employee resignation predicted by the prediction model is: in, Indicates employees In time The cumulative probability of leaving within represents the exponential function, Indicates time The baseline hazard function within represents the transpose of the feature weight vector, Indicates employees In time The eigenvector of Indicates the total number of cycles, Indicates the The intensity coefficient of the periodic effect, represents a periodic function, Represents the periodic effect period.

5. The cloud computing-based employee salary management system according to claim 2 or 4, characterized in that: The expression used by the prediction model to predict employee resignation time is: ； ； in, Indicates employees The predicted time of leaving the company, represents the time variable, Indicates employees In time The cumulative probability of leaving within represents the risk growth rate threshold, and express Parameters, Indicates time The baseline hazard function within represents the transpose of the feature weight vector, Indicates employees In time The eigenvector of Indicates the total number of cycles, Indicates the The intensity coefficient of the periodic effect, represents a periodic function, Represents the periodic effect period.

6. The cloud computing-based employee salary management system according to claim 2 or 4, characterized in that: The employee turnover prediction module further includes a model optimization unit, and the model optimization unit and the prediction model unit are used to optimize the prediction model using an optimization algorithm; The expression of the optimization algorithm is: in, represents the log-likelihood function, represents the feature weight vector, and express Parameters, Indicates time The baseline hazard function within represents the intensity coefficient of the periodic effect, represents the total number of employees, represents the indicator variable of resignation event, Indicates employees exist The risk of instant resignation at any time, Indicates employees Time of leaving the company, Indicates employees At the time of leaving The eigenvector of Indicates employees From onboarding to time The cumulative risk of leaving the company.

7. The cloud computing-based employee salary management system according to claim 1, wherein: The decision module includes: The risk grading unit is used to classify employee turnover risks into different levels based on the turnover probability and turnover time predicted by the employee turnover prediction module and the risk threshold set by the enterprise; Intervention strategy library, used to store multiple intervention strategies for different turnover risk levels; The strategy recommendation unit is connected to the risk grading unit and the intervention strategy library, and is used to recommend corresponding intervention strategies or separation management plans in the intervention strategy library based on the risk grading results; The cost-effectiveness analysis unit, connected to the strategy recommendation unit, is used to calculate the cost of the intervention strategy and the potential benefits of the turnover management program; The strategy execution unit is connected to the cost-benefit analysis unit and is used to determine whether to implement the recommended intervention strategy or separation management plan based on potential benefits; if the strategy execution unit determines that the potential benefits of implementing the intervention strategy are greater than the costs, the intervention strategy is implemented; if the strategy execution unit determines that the potential benefits of implementing the intervention strategy are less than or equal to the costs, the separation management plan is implemented.

8. The cloud computing-based employee salary management system according to claim 7, characterized in that: The cost-benefit analysis unit calculates the cost of the intervention strategy and the potential benefits of the turnover management program as follows: in, represents the net present value of potential benefits, represents the policy evaluation cycle, Indicates that employees The expected contribution value of the year, represents the discount rate, represents the total cost of the intervention strategy, Represents the replacement cost after an employee leaves the company.

9. The cloud computing-based employee salary management system according to claim 1, wherein: The adaptive salary calculation module includes: A pre-processing unit, used for pre-processing the work data of employees collected by the cloud collection module; The rule engine parsing unit is used to load the enterprise's salary rules in real time and convert the salary rules into executable calculation logic. The rule engine parsing unit is also used to adaptively adjust the parameters in the salary rules according to the intervention strategy executed by the decision module. The salary calculation execution unit is connected to the pre-processing unit and the rule engine parsing unit, and is used to calculate the employee's salary based on the employee's work data and executable calculation logic; a multi-dimensional verification unit, connected to the salary calculation execution unit, for verifying the salary calculation result using multi-dimensional data verification rules; The dynamic encryption unit is connected to the multi-dimensional verification unit and is used to encrypt, store and transmit the verified employee's salary.

10. The cloud computing-based employee salary management system according to claim 1, wherein: The system also includes a user feedback module for collecting employee satisfaction feedback on salary and transmitting the feedback data to the decision-making module in real time for the execution module to generate intervention strategies or resignation management plans.

Images