Enterprise data analysis platform based on cloud computing

Abstract

The invention relates to the technical field of enterprise data analysis, and discloses an enterprise data analysis platform based on cloud computing, which comprises an employee behavior data acquisition unit, an employee emotion data acquisition unit and an employee loss early warning unit. According to the enterprise data analysis platform based on cloud computing, an employee behavior abnormity index, a salary deviation index and a negative vocabulary density are calculated, an employee loss probability value is calculated according to the data and an employee monthly task completion rate, the employee loss probability value generated through calculation is compared with a preset threshold value, whether an employee is about to be separated or not is judged, and the employee loss probability value is calculated. The HR department is arranged to contact with the directly subordinate leadership in time, targeted assistance or appointment is provided for abnormal employees in time, reasons that the employees want to demission are mastered, targeted solutions are provided, the employee demission probability is reduced, and therefore the employee loss probability is analyzed in time, the talent loss risk is reduced, and the employee demission efficiency is improved. The stability of a core team is improved, and recruitment cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise data analysis, and specifically to an enterprise data analysis platform based on cloud computing. Background Art

[0002] Enterprise data analysis platforms are the core infrastructure for enterprises to achieve data-driven decision-making. Their role is to efficiently integrate, clean, and manage scattered and heterogeneous internal and external data (such as business systems, Internet of Things devices, market intelligence, etc.), build a unified data asset system, and then discover the laws and values behind the data through multi-dimensional analysis, visual exploration, and AI models, ultimately enabling business optimization and strategic upgrading. The platform breaks down data silos between departments, builds a standardized data warehouse or data lake, supports real-time stream processing and offline batch computing, and meets multi-level requirements from basic reports to predictive analysis. For example, in the retail scenario, the platform can integrate online and offline sales, user behavior, and supply chain data, identify the characteristics of popular products and the reasons for slow-moving items through association analysis, use machine learning to predict regional demand, and guide precise inventory replenishment and promotion strategies; in manufacturing, combine equipment sensor data to monitor the production line status in real time, and use anomaly detection algorithms to warn of potential failures and reduce downtime losses.

[0003] Currently, enterprise data analysis platforms usually focus on the combination of front-end sales and back-end production, conduct integrated analysis of various data. During the analysis process of each item of data, the probability of employee turnover is often ignored, increasing the risk of talent loss, thereby causing a decline in the stability of the core team and an increase in recruitment costs. Summary of the Invention

[0004] (1) Technical Problems to be Solved

[0005] Aiming at the deficiencies of the prior art, the present invention provides an enterprise data analysis platform based on cloud computing. It is capable of calculating the employee behavior anomaly index, salary deviation index, and negative word density, and calculating the employee turnover probability value by combining these data with the employee's monthly task completion rate. Comparing the calculated employee turnover probability value with a preset threshold to determine whether an employee is about to leave, and promptly arranging for the HR department to contact their direct supervisor to provide targeted assistance or interviews to employees with anomalies in a timely manner, understand the reasons why employees want to leave, propose targeted solutions, reduce the probability of employee turnover, thereby realizing the timely analysis of the employee turnover probability, reducing the risk of talent loss, improving the stability of the core team, and reducing recruitment costs, thus solving the above problems.

[0006] (2) Technical Solutions

[0007] To achieve the above object, the present invention provides the following technical solutions: An enterprise data analysis platform based on cloud computing, including an employee behavior data collection unit, an employee salary data collection unit, an employee task efficiency collection unit, an employee emotion data collection unit, an employee data analysis unit, and an employee turnover warning unit;

[0008] The employee behavior data collection unit is used to collect the monthly late arrival times and leave times of each employee, and send the collected data to the employee data analysis unit. The employee data analysis unit calculates the employee behavior anomaly index according to the monthly late arrival times and leave times of each employee;

[0009] The employee salary data collection unit is used to collect the monthly salary of each employee and the average salary of employees in the same industry and the same position, and send the collected data to the employee data analysis unit. The employee data analysis unit calculates the salary deviation index according to the monthly salary of each employee and the average salary of employees in the same industry and the same position;

[0010] The employee task efficiency collection unit is used to collect the monthly task completion rate of employees, and send the collected monthly task completion rate of employees to the employee data analysis unit;

[0011] The employee emotion data collection unit is used to collect the character data of employee meeting recordings and the character data of internal company email communications, and send the data to the employee data analysis unit. The employee data analysis unit identifies the number of negative words in the employee meeting recordings and the internal company email communication data, and calculates the negative word density;

[0012] The employee data analysis unit calculates the employee turnover probability value based on the employee behavior anomaly index, the salary deviation index, the monthly task completion rate of employees, and the negative word density, compares the employee turnover probability value with a preset threshold, determines whether an employee needs to be interviewed, and sends the result to the employee turnover warning unit.

[0013] Preferably, the monthly late arrival times of the employee are represented as CD i , and the monthly leave times of the employee are represented as Qj i , where the subscript i represents the employee number corresponding to the employee.

[0014] Preferably, the calculation expression of the employee behavior anomaly index is as follows:

[0015]

[0016] In the formula, XW i represents the employee behavior anomaly index of the employee with the employee number i, CD i +Qj iDenotes the sum of the monthly late arrival times and monthly leave times of the employee with employee number i. Denotes the average value of the sum of the monthly late arrival times and monthly leave times of all employees in the department to which the employee belongs. Denotes the standard deviation of the sum of the monthly late arrival times and monthly leave times of the employee and the sum of the monthly late arrival times and monthly leave times of all employees in the department to which the employee belongs.

[0017] Preferably, the employee meeting recording data is denoted as Ly i , and the company internal email data is denoted as Yj i , the number of negative words in the employee meeting recording is denoted as FML i , and the number of negative words in the company internal email communication data is denoted as FMY i , where the subscript i represents the employee number corresponding to the employee.

[0018] Preferably, the calculation expression of the negative word density is as follows:

[0019]

[0020] In the formula, FMmd i Denotes the negative word density of the employee with employee number i, FMY i +FML i Denotes the total number of negative words in the employee meeting recording and the company internal email communication data of the employee with employee number i, Ly i +Yj i Denotes the total number of characters of the employee meeting recording data and the company internal email data of the employee with employee number i.

[0021] Preferably, the monthly salary of the employee is denoted as Xz i , and the average salary of employees in the same position in the same industry is denoted as Hyxz.

[0022] Preferably, the calculation expression of the salary deviation index is as follows:

[0023]

[0024] In the formula, PLzs i Denotes the salary deviation index of the employee with employee number i, Xz i -Hyxz denotes the difference between the monthly salary of the employee with employee number i and the average salary of employees in the same position in the same industry.

[0025] Preferably, the monthly task completion rate of the employee is denoted as DC i ;

[0026] The calculation expression of the employee turnover probability value is as follows:

[0027] LS i = Xs1 * XW i + Xs2 * FMmd i + Xs3 * PLzs i + Xs4 * DC i

[0028] In the formula, Xs1 represents the abnormal index coefficient, Xs2 represents the negative word density coefficient, Xs3 represents the salary deviation index coefficient, Xs4 represents the employee monthly task completion rate coefficient. Xs1, Xs2, Xs3, and Xs4 are constant terms, and Xs1 + Xs2 + Xs3 + Xs4 = 1. Ls i represents the employee turnover probability value of the employee with the job number i.

[0029] Preferably, the preset threshold consists of a safety threshold and an observation threshold. Among them, the safety threshold is denoted as Aq, and its value is 0.4. The observation threshold is denoted as Gc, and its value is 0.6.

[0030] Preferably, the process for determining whether the employee needs to be interviewed is as follows:

[0031] Compare the calculated employee turnover probability value LS i with the preset threshold. When the employee turnover probability value LS i < Aq, it indicates that the employee turnover probability is low, and routine monitoring is performed;

[0032] When Aq ≤ LS i < Gc, it indicates that there is an employee turnover probability, and the direct superior leader is required to pay attention to this employee;

[0033] When LS i ≥ Gc, it indicates that the employee has a high turnover probability, and the direct superior leader is required to immediately interview the employee.

[0034] Compared with the prior art, the present invention provides an enterprise data analysis platform based on cloud computing, having the following beneficial effects:

[0035] The present invention calculates the employee behavior abnormal index, salary deviation index, and negative word density, and calculates the employee turnover probability value by using these data and the employee monthly task completion rate. The calculated employee turnover probability value is compared with the preset threshold to determine whether the employee is about to leave, and the HR department is timely arranged to contact the direct leader of the employee, and targeted assistance or interviews are provided to the employees with abnormalities in a timely manner, to understand the reasons why the employees want to leave, propose targeted solutions, reduce the employee turnover probability, so as to realize the timely analysis of the employee turnover probability, reduce the risk of talent loss, improve the stability of the core team, and reduce the recruitment cost. Description of the Drawings

[0036] Figure 1 This is a schematic diagram of the system framework of the present invention. Specific embodiments

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0038] Since most existing enterprise data analysis platforms cannot analyze whether employees will leave based on various data of employees, resulting in the phenomenon of enterprise brain drain, a cloud computing-based enterprise data analysis platform is proposed. Please refer to Figure 1 , which is composed of an employee behavior data collection unit, an employee salary data collection unit, an employee task efficiency collection unit, an employee emotion data collection unit, an employee data analysis unit, and an employee turnover warning unit.

[0039] Among them, the employee behavior data collection unit is used to collect the monthly late arrival times and leave times of each employee. The monthly late arrival times of an employee are represented as CD i , and the monthly leave times of an employee are represented as Qj i , where the subscript i represents the employee number corresponding to the employee. The employee behavior data collection unit sends the collected data to the employee data analysis unit. The employee data analysis unit calculates the employee behavior anomaly index based on the monthly late arrival times and leave times of each employee. The specific calculation expression is as follows:

[0040]

[0041] In the formula, XW i represents the employee behavior anomaly index of the employee with employee number i, CD i +Qj i represents the sum of the monthly late arrival times and monthly leave times of the employee with employee number i, represents the average value of the sum of the monthly late arrival times and monthly leave times of all employees in the department to which the employee belongs, represents the standard deviation of the sum of the monthly late arrival times and monthly leave times of the employee and the sum of the monthly late arrival times and monthly leave times of all employees in the department to which the employee belongs.

[0042] By calculating the behavior anomaly index of each employee, the abnormal behavior of the employee in that month can be reflected through the anomaly index, so as to analyze whether each employee will leave based on the abnormal behavior.

[0043] The employee salary data collection unit is used to collect the monthly salaries of each employee and the average salary of employees in the same industry and the same position. The monthly salary of an employee is denoted as Xz i , the average salary of employees in the same industry and the same position is denoted as Hyxz. The employee salary data collection unit sends the collected data to the employee data analysis unit. The employee data analysis unit calculates the salary deviation index based on the monthly salary of each employee and the average salary of employees in the same industry and the same position. The specific calculation formula is as follows:

[0044]

[0045] In the formula, PLzs i represents the salary deviation index of the employee with work number i, and Xz i -Hyxz represents the difference between the monthly salary of the employee with work number i and the average salary of employees in the same industry and the same position.

[0046] By calculating the salary deviation index of each employee, the specific differences in the salaries of each employee and their counterparts in the same industry and the same position can be objectively reflected, and based on this difference, the probability of employee turnover due to salary issues can be speculated and estimated.

[0047] The employee task efficiency collection unit is used to collect the monthly task completion rate of employees. The monthly task completion rate of employees is denoted as DC i , and the employee task efficiency collection unit sends the collected monthly task completion rate of employees to the employee data analysis unit.

[0048] By collecting the monthly completion rate of employees, the task achievement rate of employees can be objectively reflected. At the same time, a poor task achievement rate indicates that the employee has abnormal conditions or is slacking off during the month, and it also indicates that the employee may leave. Therefore, this data can be used as a reference factor for analyzing the probability of employee turnover.

[0049] The employee emotion data collection unit is used to collect the character data of employee meeting recordings and the character data of internal company email communications. The character data of employee meeting recordings is denoted as Ly i , and the character data of the internal company email data is denoted as Yj i . The employee emotion data collection unit sends the collected character data of employee meeting recordings and the character data of internal company email data to the employee data analysis unit. The employee data analysis unit identifies the number of negative words in the employee meeting recordings and the internal company email communication data. The number of negative words in the employee meeting recordings is denoted as FML i , and the number of negative words in the internal company email communication data is denoted as FMY i, the Employee Data Analysis Unit calculates the negative word density based on this data. It should be noted that the negative words include negative words such as "high stress", "low salary", "tired", etc. The specific calculation formula is as follows:

[0050]

[0051] In the formula, FMmd i represents the negative word density of the employee with employee number i, FMY i +FML i represents the total number of negative words in the employee meeting recording of the employee with employee number i and the number of negative words in the internal company email communication data. Ly i +Yj i represents the sum of the characters in the employee meeting recording data and the internal company email data of the employee with employee number i.

[0052] By analyzing the negative emotions expressed by employees in the company, grasp the negative emotions brought to employees by work pressure or work environment, and use the negative emotions of employees as a reference factor for calculating the employee turnover probability.

[0053] The Employee Data Analysis Unit calculates the employee turnover probability value based on the employee behavior anomaly index, salary deviation index, employee monthly task completion rate, and negative word density. The specific calculation formula is as follows:

[0054] LS i =Xs1*XW i +Xs2*FMmd i +Xs3*PLzs i +Xs4*DC i

[0055] In the formula, Xs1 represents the anomaly index coefficient, Xs2 represents the negative word density coefficient, Xs3 represents the salary deviation index coefficient, Xs4 represents the employee monthly task completion rate coefficient. Xs1, Xs2, Xs3, Xs4 are constant terms, and Xs1 + Xs2 + Xs3 + Xs4 = 1. LS i represents the employee turnover probability value of the employee with employee number i.

[0056] By calculating the employee turnover probability value, it can reflect the probability of each employee leaving under the influence of various factors.

[0057] The Employee Data Analysis Unit compares the employee turnover probability value with a preset threshold to determine whether an employee needs to be interviewed and sends the result to the Employee Turnover Early Warning Unit.

[0058] Among them, the preset threshold consists of a safety threshold and an observation threshold. The safety threshold is denoted as Aq, and its value is 0.4. The observation threshold is denoted as Gc, and its value is 0.6.

[0059] The process of determining whether an employee needs to be interviewed is as follows:

[0060] Compare the calculated employee turnover probability value LS i with the preset threshold. When the employee turnover probability value LS i < Aq, it indicates that the employee turnover probability is low, and regular monitoring is performed;

[0061] When Aq ≤ LS i < Gc, it indicates that there is an employee turnover probability, and the direct superior leader is required to pay attention to this employee;

[0062] When LS i ≥ Gc, it indicates that the employee has a high turnover probability, and the direct superior leader is required to conduct an interview immediately.

[0063] By comparing the calculated employee turnover probability value with the preset threshold, it is determined whether an employee is about to leave, and the HR department is timely arranged to contact their direct leader. For employees with abnormalities, targeted assistance or interviews are provided in a timely manner, the reasons why employees want to leave are grasped, targeted solutions are proposed, the employee turnover probability is reduced, so as to realize the timely analysis of the employee turnover probability, reduce the risk of brain drain, improve the stability of the core team, and reduce the recruitment cost.

[0064] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An enterprise data analysis platform based on cloud computing, characterized in that: It includes an employee behavior data collection unit, an employee salary data collection unit, an employee task efficiency collection unit, an employee emotion data collection unit, an employee data analysis unit, and an employee turnover warning unit; The employee behavior data collection unit is used to collect the monthly late arrival times and leave times of each employee, and send the collected data to the employee data analysis unit. The employee data analysis unit calculates the employee behavior anomaly index based on the monthly late arrival times and leave times of each employee; The employee salary data collection unit is used to collect the monthly salary of each employee and the average salary of employees in the same industry and the same position, and send the collected data to the employee data analysis unit. The employee data analysis unit calculates the salary deviation index based on the monthly salary of each employee and the average salary of employees in the same industry and the same position; The employee task efficiency collection unit is used to collect the monthly task completion rate of employees, and send the collected monthly task completion rate of employees to the employee data analysis unit; The employee emotion data collection unit is used to collect the character data of employee meeting recordings and the character data of internal company email communications, and send the data to the employee data analysis unit. The employee data analysis unit identifies the number of negative words in the employee meeting recordings and internal company email communications, and calculates the negative word density; The employee data analysis unit calculates the employee turnover probability value based on the employee behavior anomaly index, the salary deviation index, the monthly task completion rate of employees, and the negative word density, compares the employee turnover probability value with a preset threshold to determine whether an employee needs to be interviewed, and sends the result to the employee turnover warning unit.

2. The enterprise data analysis platform based on cloud computing according to claim 1, wherein: The monthly late arrival times of the employee are denoted as CD i , and the monthly leave times of the employee are denoted as Qj i , where the subscript i represents the employee number corresponding to the employee.

3. The enterprise data analysis platform based on cloud computing according to claim 2, wherein: The calculation expression of the employee behavior anomaly index is as follows: In the formula, XW i represents the employee behavior anomaly index of the employee with employee number i, CD i +Qj i represents the sum of the monthly late arrival times and monthly leave times of the employee with employee number i, represents the average value of the sum of the monthly late arrival times and monthly leave times of all employees in the department to which the employee belongs, represents the standard deviation of the sum of the monthly late arrival times and monthly leave times of the employee and the sum of the monthly late arrival times and monthly leave times of all employees in the department to which the employee belongs.

4. The enterprise data analysis platform based on cloud computing according to claim 3, wherein: The recorded data of the staff meeting is represented as Ly i , the internal company email data is represented as Yj i , the number of negative words in the recorded staff meeting is represented as FML i , the number of negative words in the internal company email communication data is represented as FMY i , where the subscript i represents the employee number corresponding to the employee.

5. The enterprise data analysis platform based on cloud computing according to claim 4, characterized in that: The calculation expression of the negative word density is as follows: In the formula, FMmd i represents the negative word density of the employee with employee number i, FMY i +FML i represents the sum of the number of negative words in the employee meeting recording of the employee with employee number i and the number of negative words in the internal company email communication data, Ly i +Yj i represents the sum of the employee meeting recording data characters and the internal company email data characters of the employee with employee number i.

6. The enterprise data analysis platform based on cloud computing according to claim 5, wherein: The monthly salary of the employee is expressed as Xz i , and the average salary of employees in the same position in the same industry is expressed as Hyxz.

7. The enterprise data analysis platform based on cloud computing according to claim 6, wherein: The calculation expression of the salary deviation index is as follows: In the formula, PLzs i represents the salary deviation index of the employee with the employee number i, Xz i -Hyxz represents the difference between the monthly salary of the employee with the employee number i and the average salary of employees in the same industry and the same position.

8. The enterprise data analysis platform based on cloud computing according to claim 7, wherein: The monthly task completion rate of the employee is expressed as DC i ; The calculation expression of the employee turnover probability value is as follows: LS i = Xs1 * XW i + Xs2 * FMmd i + Xs3 * PLzs i + Xs4 * DC i In the formula, Xs1 represents the abnormal index coefficient, Xs2 represents the negative word density coefficient, Xs3 represents the salary deviation index coefficient, Xs4 represents the monthly task completion rate coefficient of employees, Xs1, Xs2, Xs3, and Xs4 are constant terms, and Xs1 + Xs2 + Xs3 + Xs4 = 1, LS i represents the employee turnover probability value of the employee with the work number i.

9. The enterprise data analysis platform based on cloud computing according to claim 8, characterized in that: The preset threshold consists of a safety threshold and an observation threshold. Among them, the safety threshold is denoted as Aq, and its value is 0.

4. The observation threshold is denoted as Gc, and its value is 0.

6.

10. The enterprise data analysis platform based on cloud computing according to claim 9, characterized in that: The process of determining whether the employee needs to be interviewed is as follows: Compare the calculated employee turnover probability value LS i with the preset threshold. When the employee turnover probability value LS i < Aq, it indicates a low employee turnover probability and routine monitoring is performed; When Aq ≤ LS i <Gc indicates the existence of an employee turnover probability, and the immediate superior leader is required to pay attention to this employee; When LS i ≥ Gc, it means that the employee has a high probability of turnover, and the direct superior leader is immediately interviewed.

Images