Talent evaluation system based on big data

Abstract

The invention discloses a talent evaluation system based on big data, relates to the technical field of talent evaluation, and aims to clearly show the working effect and task quality of employees by quantifying evaluation indexes, provide more accurate personnel management and decision basis for enterprises, and provide more accurate decision basis for enterprises through a historical cross-collaboration module. The system can analyze cooperation conditions of employees among different departments and calculate an experience score Jypf and an applicability score Sypf, which is helpful for an enterprise to comprehensively understand performance of the employees in cross-department cooperation and identify talents with efficient cooperation potential. According to the scores, the enterprise can more effectively carry out the selection of the cross-department cooperation talents, thereby promoting the smooth operation of the cross-department cooperation. Besides, through the output of the promotion index Txzs, an enterprise can select relatively suitable talents in a cross-department project, conflicts and resource mismatching in cross-department cooperation are reduced, and the project execution efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of talent evaluation, and in particular to a talent evaluation system based on big data. Background Art

[0002] With the rapid development of information technology, big data has gradually penetrated various industries, becoming a vital force driving innovation and optimizing decision-making. In particular, in the field of enterprise management, the application of big data has enabled the optimization of employee management, performance evaluation, and cross-departmental collaboration. In the specific area of ​​talent management, talent assessment, as a key tool for corporate decision-making, is being empowered by big data technology to achieve more accurate and comprehensive employee evaluations. In particular, from the perspective of enhancing cross-departmental collaboration, talent assessment systems based on big data can provide different departments with a unified and efficient collaboration platform, helping them better share talent information and assessment results, thereby promoting resource sharing and information flow.

[0003] Although big data technology offers tremendous potential for cross-departmental collaboration, existing management models still suffer from shortcomings that hinder the effectiveness of the system. Currently, most companies rely on independent performance evaluations and staffing scheduling within each department when conducting employee evaluations and cross-departmental collaboration, lacking a unified, comprehensive perspective. This traditional approach fragments talent information between departments, leading to insufficient understanding of employees' abilities, personalities, and other information during cross-departmental collaboration. This impacts the smoothness and efficiency of collaboration and leads to deviations in talent scheduling. Furthermore, companies face issues with information redundancy and delayed decision-making in talent selection and scheduling. Traditional talent assessment methods often rely too heavily on subjective judgment and fail to consider an employee's comprehensive performance across multiple dimensions, which creates a degree of uncertainty in the selection and scheduling of talent. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the present invention provides a talent assessment system based on big data, which solves the problems in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a talent assessment system based on big data, including a big data acquisition module, a data preprocessing module, a department analysis module, a historical cross-collaboration module and a collaborative recommendation module; The big data acquisition module is used to collect the work status, performance status and cross-departmental collaboration of each employee in the enterprise based on big data technology to obtain relevant position status data information, relevant performance data information and relevant historical collaboration data information; The data preprocessing module is used to preprocess the relevant position status data information, relevant performance data information and relevant historical collaboration data information, remove noise and dimensionless processing, and then summarize and generate the enterprise talent data set; The department analysis module is used to analyze the task completion status of each employee in the department according to the relevant position status data information in the enterprise talent data set, calculate the work attitude factor Gtyz, and analyze the quality of each employee's task handling in the department in combination with the relevant performance data information to calculate the historical work quality evaluation coefficient , and preliminarily screen out the talent list to participate in cross-collaborative selection; The historical cross-collaboration module is used to analyze the cross-departmental collaboration of each employee in the historical period based on the talent list and the relevant historical collaboration data information in the enterprise talent data set, so as to calculate the experience score Jypf and the suitability score Sypf; The collaborative recommendation module is used to build a talent evaluation model using deep learning technology, and integrate the experience score Jypf, the applicability score Sypf and the historical work quality evaluation coefficient They are all input into the talent assessment model. After training and dimensionless processing, the recommendation index Txzs is fitted and output. Based on the value of the recommendation index Txzs, corporate employees who are relatively suitable for scheduling are selected.

[0006] Preferably, the big data acquisition module includes a department data collection unit and an inter-department collaboration collection unit; The department data collection unit is used to extract the work status and performance status of each employee from the human resources management department, the financial department and the task management system based on big data technology and combined with ETL technology to generate relevant position status data information and relevant performance data information, wherein the relevant position status data information includes the work completion rate in the historical period , completion efficiency in the historical period , the frequency of task timeouts in the historical period , the work completion rate in the current period, the completion efficiency in the current period and the task overtime frequency in the current period; the relevant performance data information includes the salary performance of each monitoring period in the historical period and the wage performance of each monitoring segment during the current period; The inter-departmental collaboration collection unit is used to collect the inter-departmental collaboration status of each employee in the enterprise using the task management system to generate relevant historical collaboration data information, which includes the frequency of inter-departmental communication. , Number of cross-departmental project collaborations , collaboration efficiency and the number of successful cross-departmental projects ; Among them, task management systems include Jira, Trello and Asana.

[0007] Preferably, the data preprocessing module includes a processing unit and a storage unit; The processing unit is used to perform data conversion operations on the relevant position status data information, relevant performance data information and relevant historical collaboration data information using ETL technology, and the data conversion operations include data cleaning, format conversion, data standardization, data aggregation and data mapping; The storage unit is used to use ETL technology to perform data loading operations on relevant job status data information, relevant performance data information and relevant historical collaboration data information, so as to load the relevant job status data information, relevant performance data information and relevant historical collaboration data information processed by the processing unit into the NoSQL database in the cloud platform for storage, and summarize and generate an enterprise talent data set.

[0008] Preferably, the department analysis module includes a department quality analysis unit and a preliminary screening unit; The department quality analysis unit is used to extract the work completion rate in the historical period from the relevant position status data information in the enterprise talent data set. , completion efficiency in the historical period and the frequency of task timeouts in the historical period , in order to analyze the task completion of each employee in their department, and after dimensionless processing, calculate the work attitude factor , the work attitude factor Obtained by the following formula: ; Where, represents the work attitude factor in the historical period, represents the completion rate of work in the historical period, Indicates the completion efficiency in the historical period, Indicates the task timeout frequency in the historical period, 、 and All represent weight values, where 、 and The specific value is set by the user according to the situation.

[0009] Preferably, the salary performance of each monitoring period in the historical period is extracted from the relevant performance data information in the enterprise talent data set. , and combined with the work attitude factors in the historical period , in order to analyze the quality of each employee's task handling in their department, and after dimensionless processing, calculate the historical work quality evaluation coefficient , the historical work quality assessment coefficient Obtained by the following formula: ; Where, represents the work attitude factor of the i-th monitoring period in the historical period, represents the average work attitude factor in the historical period, represents the salary performance of the i-th monitoring period in the historical period, represents the average salary performance in the historical period, Indicates the monitoring period, i=1, 2, 3, ..., n, and All represent weight values, where and The specific value is set by the user according to the situation.

[0010] Preferably, the preliminary screening unit is used to select the talent data set of the enterprise and the historical work quality evaluation coefficient The acquisition method is to calculate the current work quality evaluation coefficient of each employee , where each employee’s current work quality evaluation coefficient is Obtained by the following formula: ; Where, represents the work attitude factor of the i-th monitoring segment in the current period, represents the average work attitude factor in the current period, represents the salary performance of the i-th monitoring segment in the current period, Indicates the average salary performance in the current period; By using the current work quality evaluation coefficient of each employee and historical work quality assessment coefficient Comparative analysis is performed to preliminarily screen out talent lists for cross-collaborative selection. The specific preliminary screening process is as follows: If each employee's current work quality evaluation coefficient Exceeds historical work quality assessment coefficient At this time, the corresponding employees will be included in the talent list participating in the cross-collaboration selection, and the talent list participating in the cross-collaboration selection will be uploaded to the historical cross-collaboration module; If each employee's current work quality evaluation coefficient Does not exceed the historical work quality assessment coefficient At this time, the corresponding employees will not be included in the talent list participating in cross-collaboration selection.

[0011] Preferably, the historical cross-collaboration module includes a first analysis unit and a second analysis unit; The first analysis unit is configured to receive the talent list participating in the cross-collaboration selection from the preliminary screening unit, and extract relevant historical collaboration data information of the employees involved in the talent list participating in the cross-collaboration selection from the enterprise talent data set based on the talent list participating in the cross-collaboration selection, so as to analyze the cross-departmental collaboration of each employee in the talent list participating in the cross-collaboration selection during the historical period, so as to calculate and obtain the experience score Jypf. The experience score Jypf is obtained by the following formula: ; Where, Indicates the success rate of cross-departmental projects, Indicates the number of cross-departmental project collaborations, Indicates collaboration efficiency, Indicates the frequency of cross-departmental communication, 、 and are all weight values, among which, 、 and The specific value is set by the user according to the situation.

[0012] Preferably, the second analysis unit is used to pre-acquire the corresponding cross-departmental skill requirements Jy, which is specifically expressed as follows: Jy= , and according to the talent list participating in the cross-collaboration selection, extract the skills Jn possessed by each employee in the talent list participating in the cross-collaboration selection. The specific expression is: Jn= ;in, represents the mth skill requirement within the corresponding cross-department, The pth skill of the corresponding employee in the talent list participating in the cross-collaboration selection is represented. Based on the corresponding cross-departmental skill requirements Jy and the skills Jn possessed by each employee in the talent list participating in the cross-collaboration selection, the suitability score Sypf is obtained. The suitability score Sypf is obtained by the following formula: ; Where, Indicator function Indicates the corresponding cross-departmental skill requirements. Indicates the first skill of the corresponding employee in the talent list participating in the cross-collaboration selection. Indicates the number of corresponding cross-departmental skill requirements, .

[0013] Preferably, the collaborative recommendation module includes a comprehensive analysis unit and a recommendation unit; The comprehensive analysis unit is used to use deep learning technology and combine the enterprise talent data set and the skills Jn of each employee to build an initial model. The initial model is trained and tested using the enterprise talent data set and the skills Jn of each employee. The trained initial model is used as a recognition model to obtain feature information within the recognition model. The recognition model is trained and tested using the obtained feature information. Combined with the talent list participating in the cross-collaborative selection that has been initially screened, the trained recognition model is used as a talent assessment model. After training and dimensionless processing, the selection index Txzs is fitted and output. The selection index Txzs is obtained by the following formula: ; Where, 、 and All represent weight values, where 、 and The specific value is set by the user according to the situation.

[0014] Preferably, the recommendation unit is used to obtain the recommendation index Txzs of each employee in the talent list participating in the cross-collaborative recommendation according to the method of obtaining the recommendation index Txzs in the comprehensive analysis unit, and after size comparison, extract the maximum value of the recommendation index Txzs, and use the employee corresponding to the maximum value of the recommendation index Txzs as the preferred representative for this cross-departmental collaboration to carry out this cross-departmental collaboration.

[0015] The present invention provides a talent evaluation system based on big data, which has the following beneficial effects: (1) The system uses big data technology to collect multi-dimensional data such as employees' work status, performance status, and cross-departmental collaboration, helping companies to more accurately understand the comprehensive capabilities of employees. This data-driven evaluation method is more scientific and objective than traditional subjective evaluation, and can more accurately reflect employees' actual performance, especially in terms of their work attitude, task completion, and cross-departmental collaboration capabilities. Through quantitative evaluation indicators such as work attitude factor Gtyz and historical work quality evaluation coefficient This system clearly demonstrates employee performance and task quality, providing companies with more accurate personnel management and decision-making. Through the historical cross-collaboration module, the system analyzes employee collaboration across departments and calculates the experience score (Jypf) and suitability score (Sypf). This helps companies gain a comprehensive understanding of employee performance in cross-departmental collaboration and identify talent with the potential for effective collaboration. Based on these scores, companies can more effectively identify talent for cross-departmental collaboration, enabling departments to more accurately select employees with collaborative potential and strong adaptability, thereby promoting smooth cross-departmental collaboration. Furthermore, by outputting the selection index (Txzs), companies can select relatively suitable talent for cross-departmental projects, reducing conflicts and resource misallocation in cross-departmental collaboration and improving project execution efficiency. The system uses a deep learning model to train and fit employee assessment data to ultimately generate the selection index (Txzs). This result provides companies with a data-based personnel scheduling solution that avoids the blindness and inefficiency of traditional scheduling methods. This data-driven approach allows companies to accurately select employees who meet project needs and departmental coordination requirements, further improving the coordination and execution of cross-departmental projects. The output of the recommendation index enables enterprises to make relatively optimal talent deployment in a shorter period of time, improve the speed and quality of project execution, and avoid work delays or inefficiency caused by talent mismatch.

[0016] (2) The department quality analysis unit calculates the work attitude factor by analyzing the data of employees' work completion rate, efficiency and task overtime frequency. This factor is based on the comprehensive performance in the historical period and can reflect the employees' work attitude and execution ability in many aspects. Through dimensionless processing, the deviation caused by data differences is eliminated, making the work attitude factors of different employees comparable, ensuring the fairness and scientificity of the evaluation results. The system combines the employee's salary performance and work attitude factor to further calculate the historical work quality evaluation coefficient. As a comprehensive evaluation of the employee's task quality in the department, this coefficient not only takes into account the employee's work attitude, but also fully considers his performance. The dimensionless processing further improves the consistency and accuracy of the data, helps the company to have a more comprehensive understanding of the employee's historical performance, and provides reliable data support for decision-making. The preliminary screening unit compares the employee's current work quality evaluation coefficient with the historical work quality evaluation coefficient, and intelligently screens out qualified employees. If the employee's current work quality evaluation coefficient exceeds the historical coefficient, he will be included in the cross-departmental collaborative talent selection list. This process has a high degree of automation, effectively avoiding the subjective bias in human screening, making the cross-departmental collaborative employee selection more accurate and fair. At the same time, the system can also update the talent pool in a timely manner according to the comparison results, and provide enterprises with dynamically optimized talent resources. By accurately screening out qualified employees, the system helps enterprises select those employees with excellent work quality and collaboration capabilities during cross-departmental collaboration, significantly improving the efficiency and coordination of cross-departmental collaboration. Through the automated screening and selection process, the time consumption and communication costs caused by manual intervention are reduced, and the sharing and mobility of internal resources in the enterprise are enhanced, thereby promoting smooth teamwork and efficient execution of projects. In summary, the present invention can provide enterprises with scientific, reasonable and accurate cross-departmental collaborative talent selection through accurate work attitude analysis, comprehensive historical work quality assessment and intelligent employee screening process, thereby improving the overall human resource management efficiency of the enterprise, optimizing the cross-departmental collaboration effect, and promoting effective interaction and resource sharing among employees.

[0017] (3) The first analysis unit extracts and analyzes the historical collaboration data of employees in the talent list participating in cross-collaboration selection, and calculates the experience score Jypf to ensure that each employee's historical cross-departmental collaboration performance is comprehensively evaluated. Through this process, the company can identify employees with efficient collaboration capabilities in cross-departmental collaboration, thereby providing high-quality team members for subsequent cross-departmental collaboration tasks and improving the effectiveness of collaboration. The second analysis unit extracts the skills possessed by employees participating in cross-collaboration selection and compares them with the skill requirements of the corresponding cross-departmental collaboration to calculate the suitability score Sypf. The suitability score is evaluated based on the match between the skill requirements and the actual skills of the employees. The indicator function is used to quantify the degree of skill matching to ensure that each employee's skills meet the actual needs of cross-departmental collaboration. Through this process, the company can clearly understand the adaptability of each employee in a specific cross-departmental task, ensure the accurate skill configuration of the project team, avoid the collaboration bottleneck caused by skill mismatch, and improve the overall execution of the team. Through precise scoring from the first and second analysis units, the system comprehensively assesses employees' historical collaboration experience and skill compatibility. This precise assessment not only helps companies select experienced and well-suited employees for cross-departmental collaboration, but also ensures that team members can deliver optimal performance across diverse tasks. Through intelligent screening and precise assessment, the system reduces the subjective factors and human bias in traditional personnel selection, improves the efficiency of cross-departmental collaboration, and reduces communication costs and execution difficulties.

[0018] (4) The comprehensive analysis unit uses deep learning technology to build an initial model and trains and tests the model using the enterprise talent data set and employee skill information. This process not only improves the model's predictive ability, but also ensures that the model can fully learn and understand the characteristics and skill matching of each employee. Through this intelligent training, the system can mine potential patterns from historical data and provide a more accurate basis for talent selection. After the comprehensive analysis unit training is completed, the system optimizes the model through dimensionless processing and finally outputs the selection index Txzs. The selection index combines multiple factors such as employee historical performance, skill adaptability, and work quality evaluation, avoiding the subjective bias that may occur in traditional methods and providing a scientific selection basis for cross-departmental collaboration in the enterprise. The recommendation unit compares the selection index Txzs of each employee and selects the employee with the largest selection index as the preferred representative to participate in cross-departmental collaboration. Through this selection method based on the selection index, the system ensures the accurate matching of cross-departmental collaboration members and gives priority to those employees with cooperation potential and ability. This not only improves the efficiency of cross-departmental collaboration, but also enhances the collaborative understanding between team members, thereby ensuring the smooth progress of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1This is a block diagram of a talent assessment system based on big data in the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] Example 1 See also Figure 1 ,The present invention provides a talent assessment system based on big data, including a big data acquisition module, a data preprocessing module, a department analysis module, a historical cross-collaboration module and a collaborative recommendation module; The big data acquisition module is used to collect the work status, performance status and cross-departmental collaboration of each employee in the enterprise based on big data technology to obtain relevant position status data information, relevant performance data information and relevant historical collaboration data information; The data preprocessing module is used to preprocess the relevant position status data information, relevant performance data information and relevant historical collaboration data information, remove noise and dimensionless processing, and then summarize and generate the enterprise talent data set; The department analysis module is used to analyze the task completion status of each employee in the department according to the relevant position status data information in the enterprise talent data set, calculate the work attitude factor Gtyz, and analyze the quality of each employee's task handling in the department in combination with the relevant performance data information to calculate the historical work quality evaluation coefficient , and preliminarily screen out the talent list to participate in cross-collaborative selection; The historical cross-collaboration module is used to analyze the cross-departmental collaboration of each employee in the historical period based on the talent list and the relevant historical collaboration data information in the enterprise talent data set, so as to calculate the experience score Jypf and the suitability score Sypf; The collaborative recommendation module is used to build a talent evaluation model using deep learning technology, and integrate the experience score Jypf, the applicability score Sypf and the historical work quality evaluation coefficient They are all input into the talent assessment model. After training and dimensionless processing, the recommendation index Txzs is fitted and output. Based on the value of the recommendation index Txzs, corporate employees who are relatively suitable for scheduling are selected.

[0022] During operation, the system's Big Data Acquisition module collects comprehensive data on employees' work status, performance, and cross-departmental collaboration. This information, after noise removal and dimensionless processing in the Data Preprocessing module, forms a complete and accurate enterprise talent dataset. Based on this data, the system scientifically assesses employees' task completion, work attitude, and work quality within their respective departments, ensuring accurate and comprehensive talent assessments. The system's Historical Cross-Collaboration Module analyzes employee collaboration performance across departments to calculate experience and suitability scores. This enables companies to more accurately match employee capabilities with collaboration needs during cross-departmental collaboration, avoiding resource misallocation and reducing interdepartmental communication friction, thereby improving the efficiency and effectiveness of cross-departmental collaboration. The Collaboration Recommendation Module utilizes deep learning technology to construct a talent assessment model. By inputting various scores (such as experience, suitability, and historical work quality assessment coefficients) into the model, the model is trained and dimensionless to generate a recommendation index. This recommendation index, based on the comprehensive performance of employees across multiple dimensions, helps companies efficiently select employees who are most suitable for current tasks and future development, ensuring accurate and appropriate talent selection. This big data-based analysis approach not only enhances companies' understanding and prediction of talent but also effectively reduces the subjective bias that can occur in traditional assessments, ensuring more efficient employee scheduling and work allocation, thereby optimizing overall work efficiency and project success rates. By leveraging big data technology, all employee assessment results and historical data are objectively presented and transparent, enabling management to clearly understand each employee's comprehensive capabilities and development potential, while avoiding unfairness caused by human factors. Because the system collects and analyzes employee data in real time, particularly in cross-departmental collaboration, companies can promptly identify employee strengths and weaknesses and provide personalized training and development based on the assessment results. This dynamic adjustment mechanism helps employees continuously optimize their work performance, enhancing both individual and team efficiency. In summary, the big data-based talent assessment system not only optimizes traditional talent assessment and selection methods but also produces positive results in improving cross-departmental collaboration efficiency, enabling scientific talent selection, and optimizing decision support, thereby maximizing the long-term development and competitiveness of companies.

[0023] Example 2 Please refer to Figure 1 ,Specifically: the big data acquisition module includes a department data collection unit and an ,interdepartmental collaboration collection unit; The department data collection unit is used to extract the work status and performance status of each employee from the human resources management department, the financial department and the task management system based on big data technology and combined with ETL (Extract, Transform, Load) technology to generate relevant position status data information and relevant performance data information, wherein the relevant position status data information includes the work completion rate in the historical period , completion efficiency in the historical period , the frequency of task timeouts in the historical period , the work completion rate in the current period, the completion efficiency in the current period and the task overtime frequency in the current period; the relevant performance data information includes the salary performance of each monitoring period in the historical period and the wage performance of each monitoring segment during the current period; The inter-departmental collaboration collection unit is used to collect the inter-departmental collaboration status of each employee in the enterprise using the task management system to generate relevant historical collaboration data information, which includes the frequency of inter-departmental communication. , Number of cross-departmental project collaborations , collaboration efficiency and the number of successful cross-departmental projects ; Among them, task management systems include Jira, Trello and Asana.

[0024] The data preprocessing module includes a processing unit and a storage unit; The processing unit is used to perform data conversion operations on the relevant position status data information, relevant performance data information and relevant historical collaboration data information using ETL technology, and the data conversion operations include data cleaning, format conversion, data standardization, data aggregation and data mapping; Among them, data cleaning: remove duplicate data, fill missing values, handle outliers, etc. Format conversion: convert date format, numeric format, etc. into the format required by the NoSQL database. For example, the employee's work start time may be stored in string form and needs to be converted to a standard date format. Data standardization: standardize the same attributes from different data sources, for example, unify the "ID" fields of employees in different systems with the same name. Data aggregation: aggregate data as needed (such as sum, average, etc.). Data mapping: map fields in different source systems to corresponding fields in the target system.

[0025] The storage unit is used to use ETL technology to perform data loading operations on relevant job status data information, relevant performance data information and relevant historical collaboration data information, so as to load the relevant job status data information, relevant performance data information and relevant historical collaboration data information processed by the processing unit into the NoSQL database (such as MongoDB, Cassandra, etc.) in the cloud platform for storage, and summarize and generate an enterprise talent data set.

[0026] In this embodiment, the department data collection unit of the big data acquisition module extracts employee work status and performance data from multiple systems (such as the human resources management department, the finance department, and the task management system), covering multiple key indicators for historical and current periods, thereby providing the enterprise with a comprehensive and accurate assessment of employee work status. These indicators can help the enterprise monitor employee work performance and performance in real time, identify potential work problems in advance, optimize resource allocation, and thus improve overall work efficiency and task completion quality. The inter-departmental collaboration collection unit collects employee cross-departmental collaboration data through the task management system, covering information such as the frequency of cross-departmental communication and the number of project collaborations, and then evaluates employee performance in cross-departmental collaboration. By quantifying this historical collaboration data, the enterprise can clearly understand the collaboration status of employees across different departments, especially the collaboration efficiency and project success rate. This provides the enterprise with a more accurate cross-departmental collaboration performance assessment, helping each department more accurately select employees with excellent cross-departmental collaboration capabilities when collaborating on projects, reducing communication barriers and resource waste in collaboration. The processing unit of the data preprocessing module uses ETL technology to clean, convert, and standardize the collected data to ensure data consistency, integrity, and quality. Data cleansing effectively removes redundant data, fills missing values, and corrects outliers to ensure data accuracy. Format conversion adapts data from different systems to the requirements of a NoSQL database. Data standardization and mapping standardize identical data across different data sources, ensuring data consistency across the entire system. This series of processes significantly improves data reliability and provides a strong foundation for subsequent analysis and decision-making. The data preprocessing module's storage unit uses ETL technology to load processed data into a NoSQL database, enabling efficient storage and management.

[0027] The high performance and scalability of NoSQL databases enable the system to process and store massive amounts of employee work status, performance, and collaboration data, ensuring efficient data query and analysis. By aggregating all employee-related data into an enterprise talent dataset, companies can quickly retrieve and analyze data, effectively supporting management tasks such as talent assessment, cross-departmental collaboration, and employee scheduling. All processed and stored data serves as input for deep learning models, helping the system more accurately assess employee capabilities and potential and providing intelligent decision-making support for talent selection. Through deep learning technology, the system can continuously optimize the selection index Txzs, providing more accurate data support and decision-making basis for corporate personnel scheduling, and improving the matching of employees and tasks.

[0028] Example 3 Please refer to Figure 1 ,Specifically: the department analysis module includes a department quality analysis unit and a ,preliminary screening unit; The department quality analysis unit is used to extract the work completion rate in the historical period from the relevant position status data information in the enterprise talent data set. , completion efficiency in the historical period and the frequency of task timeouts in the historical period , in order to analyze the task completion of each employee in their department, and after dimensionless processing, calculate the work attitude factor , the work attitude factor Obtained by the following formula: ; Where, represents the work attitude factor in the historical period, represents the completion rate of work in the historical period, Indicates the completion efficiency in the historical period, Indicates the task timeout frequency in the historical period, 、 and Both represent weight values, where 0 < <1,0< <1,0< <1, 、 and The specific value is set by the user according to the situation.

[0029] The above work completion rate =Number of completed tasks / total number of assigned tasks; Completion efficiency =Task completion time / target task time; According to the relevant performance data information in the enterprise talent data set, the salary performance of each monitoring period in the historical period is extracted. , and combined with the work attitude factors in the historical period , in order to analyze the quality of each employee's task handling in their department, and after dimensionless processing, calculate the historical work quality evaluation coefficient , the historical work quality assessment coefficient Obtained by the following formula: ; Where, represents the work attitude factor of the i-th monitoring period in the historical period, represents the average work attitude factor in the historical period, represents the salary performance of the i-th monitoring period in the historical period, represents the average salary performance in the historical period, Indicates the monitoring period, i=1, 2, 3, ..., n, and Both represent weight values, where 0 < <1,0< <1, and The specific value is set by the user according to the situation.

[0030] Salary Performance Obtain through the finance department; The preliminary screening unit is used to evaluate the quality of the enterprise talent data according to the historical work quality evaluation coefficient. The acquisition method is to calculate the current work quality evaluation coefficient of each employee , where each employee’s current work quality evaluation coefficient is Obtained by the following formula: ; Where, represents the work attitude factor of the i-th monitoring segment in the current period, represents the average work attitude factor in the current period, represents the salary performance of the i-th monitoring segment in the current period, Indicates the average salary performance in the current period; Among them, the work attitude factor of the i-th monitoring period in the current period is Obtained by the following formula: ; in, Indicates the completion rate of work in the current period. Indicates the completion efficiency in the current period. Indicates the task timeout frequency in the current period; By using the current work quality evaluation coefficient of each employee and historical work quality assessment coefficient Comparative analysis is performed to preliminarily screen out talent lists for cross-collaborative selection. The specific preliminary screening process is as follows: If each employee's current work quality evaluation coefficient Exceeds historical work quality assessment coefficient At this time, the corresponding employees will be included in the talent list participating in the cross-collaboration selection, and the talent list participating in the cross-collaboration selection will be uploaded to the historical cross-collaboration module; If each employee's current work quality evaluation coefficient Does not exceed the historical work quality assessment coefficient At this time, the corresponding employees will not be included in the talent list participating in cross-collaboration selection.

[0031] In this embodiment, the quality analysis unit of the department calculates the work attitude factor by extracting and analyzing the employee's work completion rate, completion efficiency and task overtime frequency in the historical period. This work attitude factor can quantify the employee's task execution in the department and help managers comprehensively evaluate the employee's work attitude and work efficiency. Through dimensionless processing, the dimensional differences between different employees can be eliminated, making the calculation of the work attitude factor more objective and fair. The system combines the historical work attitude factor with performance data to calculate the historical work quality evaluation coefficient. Through a comprehensive analysis of the employee's salary performance and work attitude in the historical period, the generated work quality evaluation coefficient can more accurately reflect the employee's actual work performance. In this way, the company can comprehensively judge the employee's task completion quality based on actual performance and work attitude, further improve the talent selection standards, and ensure that the work quality of each employee is more in line with department needs. The initial screening unit automatically identifies promising employees for cross-departmental collaboration by comparing their current performance evaluation coefficient with their historical performance evaluation coefficient. Employees whose current performance evaluation coefficient exceeds their historical level are included in the talent selection list. This screening mechanism, based on real-time data analysis and historical data comparison, ensures that the most promising and high-performing employees are selected for cross-departmental collaboration. This process not only improves collaboration effectiveness but also optimizes employee dynamic management. Through scientific work attitude and quality assessment, the system accurately selects qualified employees, ensuring a more balanced allocation of human resources within the company's cross-departmental collaboration. This precise employee screening method helps reduce human resource waste, improve the efficiency and success rate of cross-departmental project execution, and optimize internal collaboration processes. In summary, the department analysis module, through precise work attitude assessment, historical performance analysis, and real-time screening, maximizes the performance quality of employees within their respective departments, laying a solid foundation for successful cross-departmental collaboration and promoting the optimization of internal collaboration efficiency and talent management.

[0032] Example 4 Please refer to Figure 1 ,Specifically: the historical cross-collaboration module includes a first analysis unit and a second analysis unit; The first analysis unit is configured to receive the talent list participating in the cross-collaboration selection from the preliminary screening unit, and extract relevant historical collaboration data information of the employees involved in the talent list participating in the cross-collaboration selection from the enterprise talent data set based on the talent list participating in the cross-collaboration selection, so as to analyze the cross-departmental collaboration of each employee in the talent list participating in the cross-collaboration selection during the historical period, so as to calculate and obtain the experience score Jypf. The experience score Jypf is obtained by the following formula: ; Where, Indicates the success rate of cross-departmental projects, Indicates the number of cross-departmental project collaborations, Indicates collaboration efficiency, Indicates the frequency of cross-departmental communication, 、 and are weight values, Represents the success rate of cross-departmental projects, where 0 < <1,0< <1,0< <1, 、 and The specific value is set by the user according to the situation.

[0033] Frequency of cross-departmental communication Acquire through camera; Collaboration efficiency Refers to the speed at which tasks are completed; The second analysis unit is used to pre-acquire the corresponding cross-departmental skill requirements Jy, which are specifically expressed as follows: Jy= , and according to the talent list participating in cross-collaboration selection, extract the skills Jn possessed by each employee in the talent list participating in cross-collaboration selection. The specific expression is: Jn= ;in, represents the mth skill requirement within the corresponding cross-department, The pth skill of the corresponding employee in the talent list participating in the cross-collaboration selection is represented. Based on the corresponding cross-departmental skill requirements Jy and the skills Jn possessed by each employee in the talent list participating in the cross-collaboration selection, the suitability score Sypf is obtained. The suitability score Sypf is obtained by the following formula: ; Where, represents the indicator function, Indicates the corresponding cross-departmental skill requirements. Indicates the first skill of the corresponding employee in the talent list participating in the cross-collaboration selection. Indicates the number of corresponding cross-departmental skill requirements, .

[0034] In this embodiment, the first analysis unit extracts employees' historical cross-departmental collaboration data from the enterprise talent dataset, analyzes key indicators such as the number of successful cross-departmental projects, number of collaborations, efficiency, and communication frequency, and calculates each employee's experience score, Jypf. This data not only comprehensively reflects the employee's performance in historical collaborations but also reveals their competence in multi-departmental collaboration and their communication and coordination skills. By setting appropriate weights, the system can accurately quantify each employee's collaboration experience, enabling talent screening to go beyond work performance and deeply reflect their ability to collaborate across departments. The second analysis unit obtains the corresponding cross-departmental skill requirements and matches them with the employee's skills to calculate a suitability score, Sypf. This score, based on an indicator function, automatically analyzes whether an employee's skills meet the requirements for cross-departmental collaboration. This process avoids the errors and subjectivity inherent in manual screening and improves the accuracy of cross-departmental talent selection. By matching skills with requirements, enterprises can ensure that selected employees possess true cross-departmental collaboration capabilities, thereby improving the effectiveness and efficiency of collaboration. By dually assessing employee experience and suitability, the system provides a more comprehensive understanding of each employee's performance and skill fit in cross-departmental collaboration. This ensures that every employee participating in a cross-departmental project possesses not only strong collaboration experience but also the appropriate professional skills to meet project requirements, thereby improving overall team collaboration efficiency and project success rates. This reduces project delays and resource waste caused by skill mismatches or poor collaboration. This module enables companies to select talent for cross-departmental collaboration based on data-driven intelligent analysis, avoiding the bias and errors that can arise from traditional manual decision-making. By accurately grading both experience and suitability, companies can more scientifically assess their employees' cross-departmental collaboration potential, optimize resource allocation, and enhance teamwork coordination. Ultimately, this helps drive efficient project implementation, strengthen the company's overall execution capabilities, and enhance its market competitiveness. In summary, the historical cross-departmental collaboration module, through precise experience and suitability scoring, improves the accuracy of employee cross-departmental collaboration assessments, enabling more efficient and accurate personnel scheduling and cross-departmental collaboration, and optimizing the overall effectiveness of cross-departmental collaboration.

[0035] Example 5 Please refer to Figure 1 ,Specifically: the collaborative recommendation module includes a comprehensive analysis unit and a recommendation unit; The comprehensive analysis unit is used to use deep learning technology and combine the enterprise talent data set and the skills Jn of each employee to build an initial model. The initial model is trained and tested using the enterprise talent data set and the skills Jn of each employee. The trained initial model is used as a recognition model to obtain feature information within the recognition model. The recognition model is trained and tested using the obtained feature information. Combined with the talent list participating in the cross-collaborative selection that has been initially screened, the trained recognition model is used as a talent assessment model. After training and dimensionless processing, the selection index Txzs is fitted and output. The selection index Txzs is obtained by the following formula: ; Where, 、 and Both represent weight values, where 0 < <1,0< <1,0< <1, 、 and The specific value is set by the user according to the situation.

[0036] The recommendation unit is used to obtain the recommendation index Txzs of each employee in the talent list participating in the cross-collaborative recommendation according to the method of obtaining the recommendation index Txzs in the comprehensive analysis unit, extract the maximum value of the recommendation index Txzs after size comparison, and use the employee corresponding to the maximum value of the recommendation index Txzs as the preferred representative for this cross-departmental collaboration to carry out this cross-departmental collaboration.

[0037] In this embodiment, the comprehensive analysis unit utilizes deep learning technology to construct an initial model. This model, combined with the company's talent dataset and employee skill information, is trained and tested to produce a highly accurate recognition model. This model can deeply analyze an employee's skill match, historical performance, and other key factors, and automatically generates a recommendation index (Txzs). The recommendation index, based on each employee's potential and adaptability in cross-departmental collaboration, provides a quantitative, data-driven basis for talent selection. Through this process, companies can scientifically select employees who are relatively well-suited for cross-departmental collaboration without relying on subjective judgment. To ensure the fairness and scientific nature of the selection process, the comprehensive analysis unit performs dimensionless processing when deriving the recommendation index to eliminate bias caused by different data sources and dimensions. Furthermore, by combining the employee's historical data and skills, the recommendation index (Txzs) comprehensively assesses the employee's collaboration ability, adaptability, and historical performance. This processing effectively eliminates the dimensionality discrepancies common in traditional methods, ensuring the accuracy and stability of the evaluation results. By combining deep learning technology with a data-driven selection mechanism, the present invention provides enterprises with a precise personnel selection method, which can not only effectively improve the efficiency of cross-departmental collaboration, but also ensure the complementary skills of team members and enhance overall collaboration capabilities. Ultimately, enterprises can quickly achieve reasonable resource allocation in complex cross-departmental projects, improve work efficiency and project success rate, and reduce costs and risks caused by personnel mismatch or poor collaboration. In summary, the present invention, through the application of deep learning technology, accurately screens outstanding employees participating in cross-departmental collaboration, optimizes team configuration, improves the overall effect of cross-departmental collaboration, and promotes efficient operation and sustainable development of enterprises.

[0038] 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 variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A talent assessment system based on big data, characterized by: It includes big data acquisition module, data pre-processing module, department analysis module, historical cross-collaboration module and collaborative recommendation module; The big data acquisition module is used to collect the work status, performance status and cross-departmental collaboration of each employee in the enterprise based on big data technology to obtain relevant position status data information, relevant performance data information and relevant historical collaboration data information; The data preprocessing module is used to preprocess the relevant position status data information, relevant performance data information and relevant historical collaboration data information, remove noise and dimensionless processing, and then summarize and generate the enterprise talent data set; The department analysis module is used to analyze the task completion status of each employee in the department according to the relevant position status data information in the enterprise talent data set, calculate the work attitude factor Gtyz, and analyze the quality of each employee's task handling in the department in combination with the relevant performance data information to calculate the historical work quality evaluation coefficient , and preliminarily screen out the talent list to participate in cross-collaborative selection; The historical cross-collaboration module is used to analyze the cross-departmental collaboration of each employee in the historical period based on the talent list and the relevant historical collaboration data information in the enterprise talent data set, so as to calculate the experience score Jypf and the suitability score Sypf; The collaborative recommendation module is used to build a talent evaluation model using deep learning technology, and integrate the experience score Jypf, the applicability score Sypf and the historical work quality evaluation coefficient They are all input into the talent assessment model. After training and dimensionless processing, the recommendation index Txzs is fitted and output. Based on the value of the recommendation index Txzs, corporate employees who are relatively suitable for scheduling are selected.

2. The talent assessment system based on big data according to claim 1, characterized in that: The big data acquisition module includes a department data collection unit and an inter-department collaboration collection unit; The department data collection unit is used to extract the work status and performance status of each employee from the human resources management department, the financial department and the task management system based on big data technology and combined with ETL technology to generate relevant position status data information and relevant performance data information, wherein the relevant position status data information includes the work completion rate in the historical period , completion efficiency in the historical period , the frequency of task timeouts in the historical period , the work completion rate in the current period, the completion efficiency in the current period and the task overtime frequency in the current period; the relevant performance data information includes the salary performance of each monitoring period in the historical period and the wage performance of each monitoring segment during the current period; The inter-departmental collaboration collection unit is used to collect the inter-departmental collaboration status of each employee in the enterprise using the task management system to generate relevant historical collaboration data information, which includes the frequency of inter-departmental communication. , Number of cross-departmental project collaborations , collaboration efficiency and the number of successful cross-departmental projects ; Among them, task management systems include Jira, Trello and Asana.

3. The talent assessment system based on big data according to claim 2, characterized in that: The data preprocessing module includes a processing unit and a storage unit; The processing unit is used to perform data conversion operations on the relevant position status data information, relevant performance data information and relevant historical collaboration data information using ETL technology, and the data conversion operations include data cleaning, format conversion, data standardization, data aggregation and data mapping; The storage unit is used to use ETL technology to perform data loading operations on relevant job status data information, relevant performance data information and relevant historical collaboration data information, so as to load the relevant job status data information, relevant performance data information and relevant historical collaboration data information processed by the processing unit into the NoSQL database in the cloud platform for storage, and summarize and generate an enterprise talent data set.

4. The talent assessment system based on big data according to claim 3, characterized in that: The department analysis module includes a department quality analysis unit and a preliminary screening unit; The department quality analysis unit is used to extract the work completion rate in the historical period from the relevant position status data information in the enterprise talent data set. , completion efficiency in the historical period and the frequency of task timeouts in the historical period , in order to analyze the task completion of each employee in their department, and after dimensionless processing, calculate the work attitude factor , the work attitude factor Obtained by the following formula: ； Where, represents the work attitude factor in the historical period, represents the completion rate of work in the historical period, Indicates the completion efficiency in the historical period, Indicates the task timeout frequency in the historical period, 、 and All represent weight values, where 、 and The specific value is set by the user according to the situation.

5. The talent assessment system based on big data according to claim 4, characterized in that: According to the relevant performance data information in the enterprise talent data set, the salary performance of each monitoring period in the historical period is extracted. , and combined with the work attitude factors in the historical period , in order to analyze the quality of each employee's task handling in their department, and after dimensionless processing, calculate the historical work quality evaluation coefficient , the historical work quality assessment coefficient Obtained by the following formula: ； Where, represents the work attitude factor of the i-th monitoring period in the historical period, represents the average work attitude factor in the historical period, represents the salary performance of the i-th monitoring period in the historical period, represents the average salary performance in the historical period, Indicates the monitoring period, i=1, 2, 3, ..., n, and All represent weight values, where and The specific value is set by the user according to the situation.

6. The talent assessment system based on big data according to claim 5, characterized in that: The preliminary screening unit is used to evaluate the quality of the enterprise talent data set and the historical work quality coefficient The acquisition method is to calculate the current work quality evaluation coefficient of each employee , where each employee’s current work quality evaluation coefficient is Obtained by the following formula: ； Where, represents the work attitude factor of the i-th monitoring segment in the current period, represents the average work attitude factor in the current period, represents the salary performance of the i-th monitoring segment in the current period, Indicates the average salary performance in the current period; By using the current work quality evaluation coefficient of each employee and historical work quality assessment coefficient Comparative analysis is performed to preliminarily screen out talent lists for cross-collaborative selection. The specific preliminary screening process is as follows: If each employee's current work quality evaluation coefficient Exceeds historical work quality assessment coefficient At this time, the corresponding employees will be included in the talent list participating in the cross-collaboration selection, and the talent list participating in the cross-collaboration selection will be uploaded to the historical cross-collaboration module; If each employee's current work quality evaluation coefficient Does not exceed the historical work quality assessment coefficient At this time, the corresponding employees will not be included in the talent list participating in cross-collaboration selection.

7. The talent assessment system based on big data according to claim 6, characterized in that: The historical cross-collaboration module includes a first analysis unit and a second analysis unit; The first analysis unit is configured to receive the talent list participating in the cross-collaboration selection from the preliminary screening unit, and extract relevant historical collaboration data information of the employees involved in the talent list participating in the cross-collaboration selection from the enterprise talent data set based on the talent list participating in the cross-collaboration selection, so as to analyze the cross-departmental collaboration of each employee in the talent list participating in the cross-collaboration selection during the historical period, so as to calculate and obtain the experience score Jypf. The experience score Jypf is obtained by the following formula: ； Where, Indicates the success rate of cross-departmental projects, Indicates the number of cross-departmental project collaborations, Indicates collaboration efficiency, Indicates the frequency of cross-departmental communication, 、 and are all weight values, among which, 、 and The specific value is set by the user according to the situation.

8. The talent assessment system based on big data according to claim 7, characterized in that: The second analysis unit is used to pre-acquire the corresponding cross-departmental skill requirements Jy, which are specifically expressed as follows: Jy= , and according to the talent list participating in cross-collaboration selection, extract the skills Jn possessed by each employee in the talent list participating in cross-collaboration selection. The specific expression is: Jn= ;in, represents the mth skill requirement within the corresponding cross-department, The pth skill of the corresponding employee in the talent list participating in the cross-collaboration selection is represented. Based on the corresponding cross-departmental skill requirements Jy and the skills Jn possessed by each employee in the talent list participating in the cross-collaboration selection, the suitability score Sypf is obtained. The suitability score Sypf is obtained by the following formula: ； Where, Indicator function Indicates the corresponding cross-departmental skill requirements. Indicates the first skill of the corresponding employee in the talent list participating in the cross-collaboration selection. Indicates the number of corresponding cross-departmental skill requirements, .

9. The talent assessment system based on big data according to claim 1, characterized in that: The collaborative recommendation module includes a comprehensive analysis unit and a recommendation unit; The comprehensive analysis unit is used to use deep learning technology and combine the enterprise talent data set and the skills Jn of each employee to build an initial model. The initial model is trained and tested using the enterprise talent data set and the skills Jn of each employee. The trained initial model is used as a recognition model to obtain feature information within the recognition model. The recognition model is trained and tested using the obtained feature information. Combined with the talent list participating in the cross-collaborative selection that has been initially screened, the trained recognition model is used as a talent assessment model. After training and dimensionless processing, the selection index Txzs is fitted and output. The selection index Txzs is obtained by the following formula: ； Where, 、 and All represent weight values, where 、 and The specific value is set by the user according to the situation.

10. The talent assessment system based on big data according to claim 9, characterized in that: The recommendation unit is used to obtain the recommendation index Txzs of each employee in the talent list participating in the cross-collaborative recommendation according to the method of obtaining the recommendation index Txzs in the comprehensive analysis unit, extract the maximum value of the recommendation index Txzs after size comparison, and use the employee corresponding to the maximum value of the recommendation index Txzs as the preferred representative for this cross-departmental collaboration to carry out this cross-departmental collaboration.

Images