Job skills assessment method and system

Abstract

The present invention relates to a job skill assessment system, comprising: a data collection module: responsible for comprehensively collecting employees' learning data, as well as multi-dimensional data acquisition, and updating data in a verification platform system in real time; a skill analysis module: responsible for performing skill analysis based on the collected learning data, and performing personalized adjustment of the skill analysis model according to individual differences and learning styles of employees, and calculating the learning coverage breadth and engineering application capabilities of employees at various knowledge points; a job knowledge graph module: responsible for constructing and maintaining a job knowledge graph, and adjusting the weight of knowledge points according to job demand information on a recruitment platform; in the present invention, an objective data analysis method is used to determine the weight distribution of knowledge points, and the job knowledge graph module dynamically adjusts the weight of knowledge points according to job demand information on the recruitment platform, thereby ensuring the timeliness of the knowledge graph, and this dynamic adjustment mechanism enables the system to keep up with industry development trends.

Description

Technical Field

[0001] The present invention relates to the technical field of job skill assessment, and in particular to a job skill assessment method and system. Background Art

[0002] In the certification platform system, job skills assessment plays a vital role. On the one hand, by assessing the job skills level of employees, the platform can provide employees with personalized learning suggestions and training plans to help them quickly master the required skills and knowledge. On the other hand, the platform can also issue corresponding certificates and qualifications to employees based on the assessment results as a strong proof of their professional ability.

[0003] After searching, the invention patent with Chinese patent number CN112101828A discloses a job skill assessment method, system, electronic device and storage medium, which belongs to the field of data processing technology. The method shown includes: collecting learning situation data; performing skill analysis based on the learning situation data to obtain skill analysis results; performing job skill assessment based on the skill analysis results and a pre-built job knowledge map to generate assessment results; and displaying the assessment results. According to the content of the present invention, by collecting employees' learning situation data, the comprehensive learning situation of employees can be truly, effectively and comprehensively reflected. Furthermore, skill analysis based on employees' learning situation data can truly and accurately reflect employees' skill levels. Therefore, skill assessment based on skill analysis results and job knowledge maps can provide employees with true, accurate and detailed assessment results.

[0004] However, in the process of using the above system, the setting of knowledge point weights may be affected by subjective factors, resulting in unreasonable weight distribution, which will affect the fairness and objectivity of the evaluation results. Therefore, it is particularly important to propose a job skill evaluation method and system. Summary of the invention

[0005] The purpose of the present invention is to provide a job skills assessment method and system.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] Job skills assessment system, including:

[0008] Data collection module: responsible for comprehensively collecting employees' learning data, including course learning status, task completion status, report scores, test scores, and obtaining data from multiple dimensions such as employee self-evaluation, teacher evaluation, and corporate feedback. At the same time, it updates the data in the certification platform system in real time to ensure that the skill analysis results reflect the current skill level of employees;

[0009] Skill analysis module: responsible for conducting skill analysis based on the collected learning data, making personalized adjustments to the skill analysis model based on the individual differences and learning styles of employees, and calculating the learning coverage and engineering application capabilities of employees at each knowledge point;

[0010] Job knowledge graph module: responsible for building and maintaining the job knowledge graph, and adjusting the weight of knowledge points according to job demand information on the recruitment platform;

[0011] Position analysis module: responsible for conducting position skill assessment and generating assessment results based on the skill analysis results and position knowledge graph. The assessment results may include assessment scores, position capability radar charts, and other forms;

[0012] Job skill assessment module: includes a user interaction submodule and a post-processing submodule. The user interaction submodule is responsible for interacting with the user, receiving the user's instructions and input, and displaying the assessment results. The post-processing submodule is responsible for processing the information transmitted by the user interaction submodule, including transmitting the learning situation data to the skill analysis module, receiving the skill assessment results transmitted by the skill analysis module, and transmitting the assessment results to the user interaction submodule for display. At the same time, the post-processing submodule receives the adjusted knowledge point weights and intelligent recommendation information transmitted by the job knowledge graph module, as well as the assessment results generated by the job analysis module, and integrates them for display to the user.

[0013] The learning situation data collection module passes the collected learning situation data to the skill analysis module, the skill analysis module passes the analyzed skill assessment results to the job knowledge graph module, the job knowledge graph module passes the adjusted knowledge point weights and intelligent recommendation information to the job analysis module, and the job analysis module passes the generated assessment results to the job skill assessment visualization module.

[0014] The above scheme further comprises:

[0015] Furthermore, the data collection module includes a data collection unit, a data integration unit, a data update unit, a data synchronization unit and a data verification unit. The data collection unit is responsible for collecting data from various subsystems such as the course system, task management system, report scoring system, examination system, employee self-evaluation system, teacher evaluation system and enterprise feedback system. The data integration unit cleans, organizes and classifies the data collected from each subsystem to form a unified data format. The data update unit updates the integrated data to the skill analysis module in real time to ensure the timeliness and accuracy of the data. The data synchronization unit is responsible for exchanging data with external data sources to realize data import and export functions. The data verification unit verifies the integrity, consistency, validity and other aspects of the collected data to ensure To ensure the accuracy of the data, the data acquisition unit passes the collected raw data to the data integration unit, the data integration unit cleans and organizes the raw data, the data integration unit passes the organized data to the data update unit, the data update unit updates the data to the skill analysis module, the data integration unit passes the data that needs to be synchronized to the data synchronization unit, the data synchronization unit exchanges data with an external data source, the data integration unit passes the organized data to the data verification unit, the data verification unit verifies the data and feeds back the verification result to the data integration unit for data correction or re-collection, the data update unit pushes the updated data to the skill analysis module in real time, and the skill analysis module performs skill analysis based on the updated data.

[0016] Furthermore, the skill analysis module includes a data preprocessing unit, a skill analysis model unit, a learning coverage breadth assessment unit, an engineering application capability assessment unit, and a feedback and suggestion generation unit. The data preprocessing unit cleans, denoises, and standardizes the data collected by the data collection module to ensure the accuracy and consistency of the data. The skill analysis model unit is used to analyze the skill level of employees and personalize the model according to the individual differences and learning styles of employees. The learning coverage breadth assessment unit calculates the learning coverage breadth of employees on each knowledge point based on the results of the skill analysis model. The engineering application capability assessment unit evaluates the engineering application capability of employees in combination with the project practice data, homework completion status, etc. of the employees. The feedback and suggestion generation unit generates personalized learning suggestions and resource recommendations based on the analysis results. The data preprocessing unit passes the processed standard data to the skill analysis model unit. The skill analysis model unit passes the analysis results to the learning coverage breadth assessment unit and the engineering application capability assessment unit. The learning coverage breadth assessment unit and the engineering application capability assessment unit pass the assessment results to the feedback and suggestion generation unit.

[0017] Furthermore, the job knowledge graph module includes a data collection and processing unit, a knowledge graph construction unit, a weight distribution and optimization unit, a demand analysis and cooperation unit, a graph update and maintenance unit, a relationship description and display unit, and a data interaction and interface unit. The data collection and processing unit is responsible for collecting job demand information from various channels such as recruitment platforms and industry reports, and cleaning, sorting and analyzing the collected data. The knowledge graph construction unit constructs an initial job knowledge graph based on the information provided by the data collection and processing unit, and defines the knowledge points and their relationships in the graph. The weight distribution and optimization unit uses data analysis methods such as factor analysis and cluster analysis to determine the weight distribution of knowledge points, and provides a user interface to allow users to adjust the weights according to actual needs. The demand analysis and cooperation unit cooperates with industry experts to conduct in-depth analysis of job requirements, regularly update job demand information, and ensure the accuracy and timeliness of the graph. The graph update and maintenance unit is responsible for the regular update and maintenance of the knowledge graph, monitors industry trends, and ensures the timeliness and accuracy of the graph. The relationship description and display The unit establishes a relationship description mechanism between knowledge points. The data interaction and interface unit provides an interface with other modules to realize data interaction and sharing, and supports integration with other systems, such as recruitment platforms, learning management systems, etc. The data collection and processing unit passes the processed job requirement information to the knowledge graph construction unit. The knowledge graph construction unit constructs an initial job knowledge graph based on this information. The knowledge graph construction unit passes the constructed knowledge graph to the weight distribution and optimization unit. The weight distribution and optimization unit passes the weight distribution result to the demand analysis and cooperation unit. The demand analysis and cooperation unit passes the updated job requirement information and expert opinions to the graph update and maintenance unit. The graph update and maintenance unit updates and maintains the knowledge graph based on this information. The graph update and maintenance unit passes the updated knowledge graph to the relationship description and display unit. The relationship description and display unit establishes a relationship description mechanism and displays the graph based on the knowledge points and their relationships in the graph. The relationship description and display unit passes the knowledge graph and its relationship description to the data interaction and interface unit.

[0018] Furthermore, the job analysis module includes a skill analysis receiving unit, a job knowledge graph matching unit, an evaluation calculation unit, a result generation and display unit, a comparative analysis and reporting unit, and an interactive unit personalized suggestion generation unit. The skill analysis receiving unit is responsible for receiving the skill evaluation results transmitted by the skill analysis module, preprocessing the received data, and ensuring the accuracy and completeness of the data. The job knowledge graph matching unit matches the knowledge points in the job knowledge graph according to the skill evaluation results, determines the weight of each knowledge point in the job, and provides a basis for the evaluation calculation. The evaluation calculation unit combines the skill evaluation results and the knowledge point weights to calculate the comprehensive score of the job skills and generate preliminary data of the evaluation results. The result generation and display unit converts the data provided by the evaluation calculation unit into multiple forms of evaluation results, generates visual charts such as job capability radar charts, and intuitively displays the evaluation results. The comparative analysis and reporting unit supports the comparison of evaluation results of different jobs or different time periods, generates comparative analysis reports, and helps users discover the changing trends and existing problems of job skills. The personalized suggestion generation unit generates personalized skill improvement suggestions based on the evaluation results and job requirements. Provide suggestion content and implementation plan to help users develop targeted improvement plans. After receiving the skill evaluation result transmitted by the skill analysis module, the skill analysis receiving unit transmits it to the job knowledge graph matching unit. The job knowledge graph matching unit matches the knowledge points in the job knowledge graph according to the received skill evaluation result and determines the weight. The job knowledge graph matching unit transmits the matched knowledge points and their weights to the evaluation calculation unit. The evaluation calculation unit calculates the comprehensive score of the job skills according to the knowledge point weights and the skill evaluation results. The evaluation calculation unit transmits the calculated evaluation result to the result generation and display unit. The result generation and display unit generates various forms of display content according to the evaluation results, such as evaluation scores and job capability radar charts. The result generation and display unit transmits the generated evaluation result to the comparison analysis and reporting unit. The comparison analysis and reporting unit performs comparison analysis according to the evaluation results and generates a comparison analysis report. The comparison analysis and reporting unit transmits the comparison analysis result and the evaluation result to the personalized suggestion generation unit. The personalized suggestion generation unit generates personalized skill improvement suggestions according to the comparison analysis result and job requirements.

[0019] Furthermore, the user interaction submodule receives the user's instructions and inputs, and passes them to the post-processing submodule, the post-processing submodule processes this information, and passes the evaluation results, etc. to the user interaction submodule for display, the post-processing submodule passes the user's learning situation data to the skill analysis module, the skill analysis module generates a skill evaluation result based on these data, and passes it to the post-processing submodule, the job knowledge graph module provides the adjusted knowledge point weights and intelligent recommendation information to the post-processing submodule, the job analysis module generates a job evaluation result, and passes it to the post-processing submodule, the user selects the evaluation position through the user interaction submodule and inputs the learning situation, the user interaction submodule passes the input learning situation to the post-processing submodule, the post-processing submodule passes the learning situation data to the skill analysis module, the skill analysis module generates a skill evaluation result, and passes it to the post-processing submodule, at the same time, the job knowledge graph module provides the knowledge point weights and intelligent recommendation information to the post-processing submodule, the job analysis module generates a job evaluation result, and passes it to the post-processing submodule, the post-processing submodule integrates the skill evaluation result, knowledge point weights, intelligent recommendation information and job evaluation result, and displays them to the user through the user interaction submodule.

[0020] Furthermore, the skill analysis model unit uses a recurrent neural network. The skill analysis model unit receives the learning behavior sequence of the employee (such as job completion, online test scores, learning time, etc.) and predicts the employee's future skill level or learning trend. Through training, the skill analysis model unit learns the individual differences and learning styles of the employee, thereby providing personalized skill assessment. The specific steps are:

[0021] Input Layer:

[0022] Input vector x t represents the learning behavior of employees at time step t, and the input layer converts x t Passed to the hidden layer;

[0023] Hidden Layer:

[0024] The state of the hidden layer h t From the current input x t and the hidden state h at the previous time step t-1 Jointly determined, the calculation formula is: h t =σ(W h h t-1 +W x x t +b h ), where σ is the activation function (such as Tanh or ReLU), W h and W x is the weight matrix, b his the bias term;

[0025] Output layer:

[0026] The output layer is based on the hidden state h t Calculate the predicted result y t , the calculation formula is: t =\softmax(W y h t +b y ), where \softmax is the activation function for multi-classification problems, W y is the weight matrix, b y is the bias term;

[0027] Loss function and optimization:

[0028] The loss function is used to measure the prediction result y t The difference between the actual result and the actual result. In skill analysis, the loss function is the cross entropy loss. Through the back propagation algorithm, the weight matrix W is continuously adjusted. h , W x and W y and b h and b y A bias term to minimize the loss function;

[0029] Loops and Iterations:

[0030] The skill analysis model unit processes the entire learning behavior sequence through a loop iteration. At each time step t, the skill analysis model unit updates the hidden state h t And calculate the predicted result y t , ultimately, the output of the RNN can be a sequence (such as the predicted employee skill level sequence) or a single value (such as the employee’s final skill level).

[0031] Furthermore, the weight allocation and optimization unit uses the DBSCAN algorithm to determine the weight allocation of knowledge points. Given a data set D = {x1, x2, ..., xn}, each data point xi represents a d-dimensional vector, which may represent various features related to the knowledge point (such as employee's academic performance, learning time, etc.), □ neighborhood: For data point xi, its □ neighborhood contains all data points whose distance to xi is less than or equal to □, that is, N(xi) = {xj|dist(xi,xj)≤□}, core object: If the □ neighborhood of data point xi contains at least MinPts data points (including xi itself), then xi is a core object, directly density reachable: For data points xi and xj, if xj is in the □ neighborhood of xi, and xi is a core object, then xj is said to be directly density reachable to xi, density reachable: For data points xi and xj, if there exists a data point sequence p1, p2, ..., pl, where p1 = xi, pl = xj, and pi + 1 is directly density reachable to pi, then xj is said to be density reachable to xi, density connected: For data points xi and xj, if there exists a data point xk such that both xi and xj are density reachable to xk, then xi and xj are said to be density connected:

[0032] Initialize and mark all data points as unclassified;

[0033] For each data point xi, determine whether it is a core object. If it is a core object, build a new cluster and add all data points in xi and its □ neighborhood (including xi itself and data points directly reachable by density) to the cluster;

[0034] For each data point of each core object (a data point that has been added to a cluster), recursively add its density-reachable data points (i.e., those that can be reached through a series of direct density-reachable relationships) to the same cluster;

[0035] Continue to do the above operation for the unclassified data points until all data points are classified into a cluster, or there are no more data points that can be added to any cluster;

[0036] The above process is repeated until all core objects have been processed and all data points have been classified into corresponding clusters.

[0037] Evaluation method and steps of job skill evaluation system:

[0038] S1: Comprehensive data collection: Obtain employees’ learning progress, task completion, online test scores, etc. from the course learning system; collect employees’ practice reports, project assignments and their scores to evaluate their practical ability; understand employees’ cognition of their own learning through self-evaluation; obtain teachers’ evaluation feedback, including classroom performance, homework completion, etc.; receive feedback from companies to understand employees’ application of skills during internships or work;

[0039] Real-time data update: The data collection module updates the data in the certification platform system regularly or in real time to ensure that the skill analysis results can accurately reflect the current skill level of employees. It allows users (such as teachers and administrators) to manually import data from external data sources, such as employees' learning records or certificate information on other platforms, to keep the data comprehensive and up-to-date;

[0040] S2: Skill level analysis: The skill analysis module uses a recurrent neural network to conduct in-depth analysis of the collected learning information and predict employees' future skill levels or learning trends. Through training, it learns about employees' individual differences and learning styles. Based on employees' individual differences and learning styles, it makes personalized adjustments to the model and analyzes the breadth and depth of employees' learning coverage in various knowledge points.

[0041] S3: Build and maintain job knowledge graph: Based on the job requirement information on the recruitment platform, the DBSCAN algorithm is used to determine the weight distribution of knowledge points, allowing users to adjust the weights according to actual needs to adapt to the skill requirements of different positions;

[0042] S4: Receive skill assessment results: The job analysis module receives the skill assessment results delivered by the skill analysis module, including the employee’s learning coverage breadth and depth of each knowledge point;

[0043] S5: Generate evaluation results based on knowledge point weights: Generate employee job skill evaluation results based on knowledge point weights in the job knowledge graph. The evaluation results can include evaluation scores, job capability radar charts, and other forms to intuitively display employee skill levels and job adaptability.

[0044] The present invention has the following beneficial effects:

[0045] 1. In the present invention, the skill analysis module can make personalized adjustments based on the collected data according to the individual differences and learning styles of employees, so as to calculate more accurate skill assessment results. By calculating the learning coverage breadth and engineering application capabilities of employees at various knowledge points, the system can more comprehensively evaluate the skill level of employees and provide a more accurate data basis for job skill assessment.

[0046] 2. In the present invention, an objective data analysis method is used to determine the weight distribution of knowledge points. The job knowledge graph module dynamically adjusts the weights of knowledge points according to the job requirement information on the recruitment platform, ensuring the timeliness of the knowledge graph. This dynamic adjustment mechanism enables the system to keep up with industry development trends. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a system block diagram of the job skill assessment system proposed by the present invention;

[0048] Figure 2 The present invention provides a flowchart of the evaluation method of the job skill evaluation system. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0050] See also Figure 1-Figure 2 As shown, the present invention is a job skill assessment system, comprising:

[0051] Data collection module: responsible for comprehensively collecting employees' learning data, including course learning status, task completion status, report scores, test scores, and obtaining data from multiple dimensions such as employee self-evaluation, teacher evaluation, and corporate feedback. At the same time, it updates the data in the certification platform system in real time to ensure that the skill analysis results reflect the current skill level of employees;

[0052] Skill analysis module: responsible for conducting skill analysis based on the collected learning data, making personalized adjustments to the skill analysis model based on the individual differences and learning styles of employees, and calculating the learning coverage and engineering application capabilities of employees at each knowledge point;

[0053] Job knowledge graph module: responsible for building and maintaining the job knowledge graph, and adjusting the weight of knowledge points according to job demand information on the recruitment platform;

[0054] Position analysis module: responsible for conducting position skill assessment and generating assessment results based on skill analysis results and position knowledge graphs. The assessment results can be in various forms such as assessment scores and position capability radar charts;

[0055] Job skill assessment module: includes a user interaction submodule and a post-processing submodule. The user interaction submodule is responsible for interacting with the user, receiving the user's instructions and input, and displaying the assessment results. The post-processing submodule is responsible for processing the information transmitted by the user interaction submodule, including transmitting the learning situation data to the skill analysis module, receiving the skill assessment results transmitted by the skill analysis module, and transmitting the assessment results to the user interaction submodule for display. At the same time, the post-processing submodule receives the adjusted knowledge point weights and intelligent recommendation information transmitted by the job knowledge graph module, as well as the assessment results generated by the job analysis module, and integrates them for display to the user.

[0056] The learning situation data collection module passes the collected learning situation data to the skill analysis module, the skill analysis module passes the analyzed skill assessment results to the job knowledge graph module, the job knowledge graph module passes the adjusted knowledge point weights and intelligent recommendation information to the job analysis module, and the job analysis module passes the generated assessment results to the job skill assessment visualization module.

[0057] In one embodiment, for the above-mentioned data collection module, the data collection module includes a data collection unit, a data integration unit, a data update unit, a data synchronization unit and a data verification unit. The data collection unit is responsible for collecting data from various subsystems such as the course system, task management system, report scoring system, examination system, employee self-evaluation system, teacher evaluation system and enterprise feedback system. The data integration unit cleans, organizes and classifies the data collected from each subsystem to form a unified data format. The data update unit updates the integrated data to the skill analysis module in real time to ensure the timeliness and accuracy of the data. The data synchronization unit is responsible for exchanging data with external data sources to realize the data import and export functions. The data verification unit verifies the integrity and consistency of the collected data. The data collection unit will pass the collected raw data to the data integration unit, which will clean and organize the raw data. The data integration unit will pass the organized data to the data update unit, which will update the data to the skill analysis module. The data integration unit will pass the data that needs to be synchronized to the data synchronization unit, which will exchange data with the external data source. The data integration unit will pass the organized data to the data verification unit, which will verify the data and feed back the verification result to the data integration unit for data correction or re-collection. The data update unit will push the updated data to the skill analysis module in real time, and the skill analysis module will perform skill analysis based on the updated data.

[0058] In one embodiment, for the above-mentioned skill analysis module, the skill analysis module includes a data preprocessing unit, a skill analysis model unit, a learning coverage breadth assessment unit, an engineering application capability assessment unit, and a feedback and suggestion generation unit. The data preprocessing unit cleans, denoises, and standardizes the data collected by the data collection module to ensure the accuracy and consistency of the data. The skill analysis model unit is used to analyze the skill level of the employees and personalize the model according to the individual differences and learning styles of the employees. The learning coverage breadth assessment unit calculates the learning coverage breadth of the employees on each knowledge point based on the results of the skill analysis model. The engineering application capability assessment unit evaluates the engineering application capability of the employees in combination with the project practice data, homework completion status, etc. of the employees. The feedback and suggestion generation unit generates personalized learning suggestions and resource recommendations based on the analysis results. The data preprocessing unit passes the processed standard data to the skill analysis model unit. The skill analysis model unit passes the analysis results to the learning coverage breadth assessment unit and the engineering application capability assessment unit. The learning coverage breadth assessment unit and the engineering application capability assessment unit pass the assessment results to the feedback and suggestion generation unit.

[0059] In one embodiment, for the above-mentioned job knowledge graph module, the job knowledge graph module includes a data collection and processing unit, a knowledge graph construction unit, a weight distribution and optimization unit, a demand analysis and cooperation unit, a graph update and maintenance unit, a relationship description and display unit, and a data interaction and interface unit. The data collection and processing unit is responsible for collecting job demand information from various channels such as recruitment platforms and industry reports, and cleaning, sorting and analyzing the collected data. The knowledge graph construction unit constructs an initial job knowledge graph based on the information provided by the data collection and processing unit, defines the knowledge points and their relationships in the graph, and the weight distribution and optimization unit uses data analysis methods such as factor analysis and cluster analysis to determine the weight distribution of knowledge points, and provides a user interface to allow users to adjust the weights according to actual needs. The demand analysis and cooperation unit cooperates with industry experts to conduct in-depth analysis of job requirements, regularly updates job demand information, and ensures the accuracy and timeliness of the graph. The graph update and maintenance unit is responsible for the regular update and maintenance of the knowledge graph, monitors industry trends, and ensures the timeliness of the graph and accuracy, the relationship description and display unit establishes a relationship description mechanism between knowledge points, the data interaction and interface unit provides interfaces with other modules, realizes data interaction and sharing, and supports integration with other systems, such as recruitment platforms, learning management systems, etc. The data collection and processing unit passes the processed job requirement information to the knowledge graph construction unit, and the knowledge graph construction unit constructs an initial job knowledge graph based on this information. The knowledge graph construction unit passes the constructed knowledge graph to the weight allocation and optimization unit, and the weight allocation and optimization unit passes the weight allocation result to the demand analysis and cooperation unit. The demand analysis and cooperation unit passes the updated job requirement information and expert opinions to the graph update and maintenance unit. The graph update and maintenance unit updates and maintains the knowledge graph based on this information. The graph update and maintenance unit passes the updated knowledge graph to the relationship description and display unit. The relationship description and display unit establishes a relationship description mechanism and displays the graph based on the knowledge points and their relationships in the graph. The relationship description and display unit passes the knowledge graph and its relationship description to the data interaction and interface unit.

[0060] In one embodiment, for the above-mentioned job analysis module, the job analysis module includes a skill analysis receiving unit, a job knowledge graph matching unit, an evaluation calculation unit, a result generation and display unit, a comparative analysis and reporting unit, and an interactive unit personalized suggestion generation unit. The skill analysis receiving unit is responsible for receiving the skill evaluation results transmitted by the skill analysis module, preprocessing the received data, and ensuring the accuracy and completeness of the data. The job knowledge graph matching unit matches the knowledge points in the job knowledge graph according to the skill evaluation results, determines the weight of each knowledge point in the job, and provides a basis for the evaluation calculation. The evaluation calculation unit combines the skill evaluation results and the knowledge point weights to calculate the comprehensive score of the job skills and generate preliminary data of the evaluation results. The result generation and display unit converts the data provided by the evaluation calculation unit into multiple forms of evaluation results, generates visual charts such as job capability radar charts, and intuitively displays the evaluation results. The comparative analysis and reporting unit supports the comparison of evaluation results of different jobs or different time periods, generates comparative analysis reports, and helps users discover the changing trends and existing problems of job skills. The personalized suggestion generation unit generates personalized suggestions based on the evaluation results and job requirements. The skill analysis receiving unit receives the skill evaluation result transmitted by the skill analysis module and transmits it to the job knowledge graph matching unit. The job knowledge graph matching unit matches the knowledge points in the job knowledge graph according to the received skill evaluation result and determines the weight. The job knowledge graph matching unit transmits the matched knowledge points and their weights to the evaluation calculation unit. The evaluation calculation unit calculates the comprehensive score of the job skills according to the knowledge point weights and the skill evaluation results. The evaluation calculation unit transmits the calculated evaluation result to the result generation and display unit. The result generation and display unit generates various forms of display content according to the evaluation results, such as evaluation scores and job capability radar charts. The result generation and display unit transmits the generated evaluation result to the comparative analysis and reporting unit. The comparative analysis and reporting unit performs comparative analysis according to the evaluation results and generates a comparative analysis report. The comparative analysis and reporting unit transmits the comparative analysis results and the evaluation results to the personalized suggestion generation unit. The personalized suggestion generation unit generates personalized skill improvement suggestions according to the comparative analysis results and job requirements.

[0061] In one embodiment, for the above-mentioned user interaction submodule, the user interaction submodule receives the user's instructions and inputs, and passes them to the post-processing submodule. The post-processing submodule processes this information and passes the evaluation results to the user interaction submodule for display. The post-processing submodule passes the user's learning situation data to the skill analysis module. The skill analysis module generates a skill evaluation result based on this data and passes it to the post-processing submodule. The job knowledge graph module provides the adjusted knowledge point weights and intelligent recommendation information to the post-processing submodule. The job analysis module generates a job evaluation result and passes it to the post-processing submodule. The user selects the evaluation position through the user interaction submodule and enters the learning situation. The user interaction submodule passes the input learning situation to the post-processing submodule. The post-processing submodule passes the learning situation data to the skill analysis module. The skill analysis module generates a skill evaluation result and passes it to the post-processing submodule. At the same time, the job knowledge graph module provides the knowledge point weights and intelligent recommendation information to the post-processing submodule. The job analysis module generates a job evaluation result and passes it to the post-processing submodule. The post-processing submodule integrates the skill evaluation results, knowledge point weights, intelligent recommendation information and job evaluation results, and displays them to the user through the user interaction submodule.

[0062] In one embodiment, for the skill analysis model unit, the skill analysis model unit uses a recurrent neural network. The skill analysis model unit receives the learning behavior sequence of the employee (such as job completion, online test scores, learning time, etc.), and predicts the employee's future skill level or learning trend. Through training, the skill analysis model unit learns the individual differences and learning styles of the employee, thereby providing personalized skill assessment. The specific steps are:

[0063] Input Layer:

[0064] Input vector x t represents the learning behavior of employees at time step t, and the input layer converts x t Passed to the hidden layer;

[0065] Hidden Layer:

[0066] The state of the hidden layer h t From the current input x t and the hidden state h at the previous time step t-1 Jointly determined, the calculation formula is: h t =σ(W h h t-1 +W x x t +b h ), where σ is the activation function (such as Tanh or ReLU), W h and W x is the weight matrix, bh is the bias term;

[0067] Output layer:

[0068] The output layer is based on the hidden state h t Calculate the predicted result y t , the calculation formula is: t =\softmax(W y h t +b y ), where \softmax is the activation function for multi-classification problems, W y is the weight matrix, b y is the bias term;

[0069] Loss function and optimization:

[0070] The loss function is used to measure the prediction result y t The difference between the actual result and the actual result. In skill analysis, the loss function is the cross entropy loss. Through the back propagation algorithm, the weight matrix W is continuously adjusted. h , W x and W y and b h and b y A bias term to minimize the loss function;

[0071] Loops and Iterations:

[0072] The skill analysis model unit processes the entire learning behavior sequence through loop iteration. At each time step t, the skill analysis model unit updates the hidden state h t And calculate the predicted result y t , ultimately, the output of the RNN can be a sequence (such as the predicted employee skill level sequence) or a single value (such as the employee’s final skill level).

[0073] In one embodiment, for the above-mentioned weight allocation and optimization unit, the weight allocation and optimization unit uses the DBSCAN algorithm to determine the weight allocation of knowledge points. Given a data set D = {x1, x2, ..., xn}, each data point xi represents a d-dimensional vector, which may represent various features related to the knowledge point (such as the employee's academic performance, learning time, etc.), □ neighborhood: For a data point xi, its □ neighborhood contains all data points whose distance to xi is less than or equal to □, that is, N(xi) = {xj|dist(xi,xj)≤□}, core object: If the □ neighborhood of a data point xi contains at least MinP ts data points (including xi itself), then xi is a core object, directly density reachable: For data points xi and xj, if xj is in the □ neighborhood of xi, and xi is a core object, then xj is said to be directly density reachable to xi, density reachable: For data points xi and xj, if there exists a data point sequence p1, p2, ..., pl, where p1 = xi, pl = xj, and pi + 1 is directly density reachable to pi, then xj is said to be density reachable to xi, density connected: For data points xi and xj, if there exists a data point xk such that both xi and xj are density reachable to xk, then xi and xj are said to be density connected:

[0074] Initialize and mark all data points as unclassified;

[0075] For each data point xi, determine whether it is a core object. If it is a core object, build a new cluster and add all data points in xi and its □ neighborhood (including xi itself and data points directly reachable by density) to the cluster;

[0076] For each data point of each core object (a data point that has been added to a cluster), recursively add its density-reachable data points (i.e., those that can be reached through a series of direct density-reachable relationships) to the same cluster;

[0077] Continue to do the above operation for the unclassified data points until all data points are classified into a cluster, or there are no more data points that can be added to any cluster;

[0078] The above process is repeated until all core objects have been processed and all data points have been classified into corresponding clusters.

[0079] Evaluation method and steps of job skill evaluation system:

[0080] S1: Comprehensive data collection: Obtain employees’ learning progress, task completion, online test scores, etc. from the course learning system; collect employees’ practice reports, project assignments and their scores to evaluate their practical ability; understand employees’ cognition of their own learning through self-evaluation; obtain teachers’ evaluation feedback, including classroom performance, homework completion, etc.; receive feedback from companies to understand employees’ application of skills during internships or work;

[0081] Real-time data update: The data collection module updates the data in the certification platform system regularly or in real time to ensure that the skill analysis results can accurately reflect the current skill level of employees. It allows users (such as teachers and administrators) to manually import data from external data sources, such as employees' learning records or certificate information on other platforms, to keep the data comprehensive and up-to-date;

[0082] S2: Skill level analysis: The skill analysis module uses a recurrent neural network to conduct in-depth analysis of the collected learning information and predict employees' future skill levels or learning trends. Through training, it learns about employees' individual differences and learning styles. Based on employees' individual differences and learning styles, it makes personalized adjustments to the model and analyzes the breadth and depth of employees' learning coverage in various knowledge points.

[0083] S3: Build and maintain job knowledge graph: Based on the job requirement information on the recruitment platform, the DBSCAN algorithm is used to determine the weight distribution of knowledge points, allowing users to adjust the weights according to actual needs to adapt to the skill requirements of different positions;

[0084] S4: Receive skill assessment results: The job analysis module receives the skill assessment results delivered by the skill analysis module, including the employee’s learning coverage breadth and depth of each knowledge point;

[0085] S5: Generate evaluation results based on knowledge point weights: Generate employee job skill evaluation results based on knowledge point weights in the job knowledge graph. The evaluation results can include evaluation scores, job capability radar charts, and other forms to intuitively display employee skill levels and job adaptability.

[0086] Although 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 the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Job skills assessment system, characterized by: include: Data collection module: responsible for comprehensively collecting employees' learning data and obtaining data in multiple dimensions, while updating the data in the verification platform system in real time; Skill analysis module: responsible for conducting skill analysis based on the collected learning data, making personalized adjustments to the skill analysis model based on the individual differences and learning styles of employees, and calculating the learning coverage and engineering application capabilities of employees at each knowledge point; Job knowledge graph module: responsible for building and maintaining the job knowledge graph, and adjusting the weight of knowledge points according to job demand information on the recruitment platform; Position analysis module: responsible for conducting position skill assessment and generating assessment results based on skill analysis results and position knowledge graph; Job skill assessment module: includes a user interaction submodule and a post-processing submodule. The user interaction submodule is responsible for interacting with the user, receiving the user's instructions and inputs, and displaying the assessment results. The post-processing submodule is responsible for processing the information transmitted by the user interaction submodule. At the same time, the post-processing submodule receives the adjusted knowledge point weights and intelligent recommendation information transmitted by the job knowledge graph module, as well as the assessment results generated by the job analysis module, and integrates them for display to the user. The data collection module passes the collected learning data to the skill analysis module, the skill analysis module passes the analyzed skill assessment results to the job knowledge graph module, the job knowledge graph module passes the adjusted knowledge point weights and intelligent recommendation information to the job analysis module, and the job analysis module passes the generated assessment results to the job skill assessment visualization module.

2. The job skill evaluation system according to claim 1, characterized in that: The data collection module includes a data collection unit, a data integration unit, a data update unit, a data synchronization unit and a data verification unit. The data collection unit is responsible for collecting data from each subsystem. The data integration unit cleans, organizes and classifies the data collected from each subsystem to form a unified data format. The data update unit updates the integrated data to the skill analysis module in real time. The data synchronization unit is responsible for exchanging data with an external data source to realize the data import and export functions. The data verification unit verifies the collected data. The data collection unit passes the collected raw data to the data integration unit. The data integration unit cleans and organizes the raw data. The data integration unit passes the organized data to the data update unit. The data update unit updates the data to the skill analysis module. The data integration unit passes the data that needs to be synchronized to the data synchronization unit. The data synchronization unit exchanges data with an external data source. The data integration unit passes the organized data to the data verification unit. The data verification unit verifies the data and feeds back the verification results to the data integration unit. The data update unit pushes the updated data to the skill analysis module in real time. The skill analysis module performs skill analysis based on the updated data.

3. The job skill evaluation system according to claim 1, characterized in that: The skill analysis module includes a data preprocessing unit, a skill analysis model unit, a learning coverage breadth assessment unit, an engineering application capability assessment unit, and a feedback and suggestion generation unit. The data preprocessing unit cleans, denoises, and standardizes the data collected by the data collection module. The skill analysis model unit is used to analyze the skill level of employees and personalize the model according to the individual differences and learning styles of employees. The learning coverage breadth assessment unit calculates the learning coverage breadth of employees on each knowledge point based on the results of the skill analysis model. The engineering application capability assessment unit assesses the engineering application capability of employees. The feedback and suggestion generation unit generates personalized learning suggestions and resource recommendations based on the analysis results. The data preprocessing unit passes the processed standard data to the skill analysis model unit. The skill analysis model unit passes the analysis results to the learning coverage breadth assessment unit and the engineering application capability assessment unit. The learning coverage breadth assessment unit and the engineering application capability assessment unit pass the assessment results to the feedback and suggestion generation unit.

4. The job skill evaluation system according to claim 1, characterized in that: The job knowledge graph module includes a data collection and processing unit, a knowledge graph construction unit, a weight distribution and optimization unit, a demand analysis and cooperation unit, a graph update and maintenance unit, a relationship description and display unit, and a data interaction and interface unit. The data collection and processing unit is responsible for collecting job demand information from various channels, and cleaning, sorting and analyzing the collected data. The knowledge graph construction unit constructs an initial job knowledge graph based on the information provided by the data collection and processing unit, and defines the knowledge points and their relationships in the graph. The weight distribution and optimization unit uses a data analysis method to determine the weight distribution of knowledge points, and provides a user interface to allow users to adjust the weights according to actual needs. The demand analysis and cooperation unit cooperates with industry experts to conduct in-depth analysis of job requirements and regularly update job requirement information. The graph update and maintenance unit is responsible for regular updates of the knowledge graph. and maintenance work, monitoring industry trends, the relationship description and display unit establishes a relationship description mechanism between knowledge points, the data interaction and interface unit provides an interface with other modules to realize data interaction and sharing, the data collection and processing unit passes the processed job requirement information to the knowledge graph construction unit, the knowledge graph construction unit constructs an initial job knowledge graph based on this information, the knowledge graph construction unit passes the constructed knowledge graph to the weight distribution and optimization unit, the weight distribution and optimization unit passes the weight distribution result to the demand analysis and cooperation unit, the demand analysis and cooperation unit passes the updated job requirement information and expert opinions to the graph update and maintenance unit, the graph update and maintenance unit passes the updated knowledge graph to the relationship description and display unit, the relationship description and display unit passes the knowledge graph and its relationship description to the data interaction and interface unit.

5. The job skill evaluation system according to claim 1, characterized in that: The job analysis module includes a skill analysis receiving unit, a job knowledge graph matching unit, an evaluation calculation unit, a result generation and display unit, a comparative analysis and reporting unit, and an interactive unit personalized suggestion generation unit. The skill analysis receiving unit is responsible for receiving the skill evaluation results transmitted by the skill analysis module. The job knowledge graph matching unit matches the knowledge points in the job knowledge graph according to the skill evaluation results and determines the weight of each knowledge point in the job. The evaluation calculation unit combines the skill evaluation results and the knowledge point weights to calculate the comprehensive score of the job skills and generate preliminary data of the evaluation results. The result generation and display unit converts the data provided by the evaluation calculation unit into evaluation results in various forms. The comparative analysis and reporting unit The reporting unit supports comparison of evaluation results of different positions or different time periods, and generates a comparative analysis report. The personalized suggestion generating unit generates personalized skill improvement suggestions according to the evaluation results and position requirements. The skill analysis receiving unit receives the skill evaluation results transmitted by the skill analysis module and transmits them to the position knowledge graph matching unit. The position knowledge graph matching unit transmits the matched knowledge points and their weights to the evaluation calculation unit. The evaluation calculation unit transmits the calculated evaluation results to the result generation and display unit. The result generation and display unit transmits the generated evaluation results to the comparative analysis and reporting unit. The comparative analysis and reporting unit transmits the comparative analysis results and the evaluation results to the personalized suggestion generating unit.

6. The job skill evaluation system according to claim 1, characterized in that: The user interaction submodule receives the user's instructions and inputs, and passes them to the post-processing submodule, the post-processing submodule processes this information, and passes the evaluation results to the user interaction submodule for display, the post-processing submodule passes the user's learning situation data to the skill analysis module, the job knowledge graph module provides the adjusted knowledge point weights and intelligent recommendation information to the post-processing submodule, the job analysis module generates a job evaluation result, and passes it to the post-processing submodule, the user selects the evaluation position through the user interaction submodule and inputs the learning situation, the user interaction submodule passes the input learning situation to the post-processing submodule, the post-processing submodule passes the learning situation data to the skill analysis module, the skill analysis module generates a skill evaluation result, and passes it to the post-processing submodule, at the same time, the job knowledge graph module provides the knowledge point weights and intelligent recommendation information to the post-processing submodule, the job analysis module generates a job evaluation result, and passes it to the post-processing submodule, the post-processing submodule integrates the skill evaluation results, knowledge point weights, intelligent recommendation information and job evaluation results, and displays them to the user through the user interaction submodule.

7. The job skill evaluation system according to claim 3, characterized in that: The skill analysis model unit uses a recurrent neural network. The skill analysis model unit receives the learning behavior sequence of the employee and predicts the employee's future skill level or learning trend. Through training, the skill analysis model unit learns the individual differences and learning styles of the employee. The specific steps are: Input Layer: Input vector x t represents the learning behavior of employees at time step t, and the input layer converts x t Passed to the hidden layer; Hidden Layer: The state of the hidden layer h t From the current input x t and the hidden state h of the previous time step t-1 Jointly determined, the calculation formula is: h t =σ(W h h t-1 +W x x t +b h ), where σ is the activation function, W h and W x is the weight matrix, b h is the bias term; Output layer: The output layer is based on the hidden state h t Calculate the predicted result y t , the calculation formula is: t =\softmax(W y h t +b y ), where \softmax is the activation function for multi-classification problems, W y is the weight matrix, b y is the bias term; Loss function and optimization: The loss function is used to measure the prediction result y t The difference between the actual result and the actual result. In skill analysis, the loss function is the cross entropy loss. Through the back propagation algorithm, the weight matrix W is continuously adjusted. h , W x and W y and b h and b y A bias term to minimize the loss function; Loops and Iterations: The skill analysis model unit processes the entire learning behavior sequence through a loop iteration. At each time step t, the skill analysis model unit updates the hidden state h t And calculate the predicted result y t .

8. The job skill evaluation system according to claim 4, characterized in that: The weight assignment and optimization unit uses the DBSCAN algorithm to determine the weight assignment of knowledge points. Given a data set D = {x1, x2, ..., xn}, each data point xi represents a d-dimensional vector, which may represent various features related to the knowledge point. □ neighborhood: For a data point xi, its □ neighborhood contains all data points whose distance to xi is less than or equal to □, that is, N(xi) = {xj|dist(xi,xj)≤□}. Core object: If the □ neighborhood of a data point xi contains at least MinPts data points, then xi is a core object. Object, direct density reachable: For data points xi and xj, if xj is in the □ neighborhood of xi, and xi is a core object, then xj is said to be directly density reachable to xi, density reachable: For data points xi and xj, if there exists a data point sequence p1, p2, ..., pl, where p1 = xi, pl = xj, and pi + 1 is directly density reachable to pi, then xj is said to be density reachable to xi, density connected: For data points xi and xj, if there exists a data point xk such that both xi and xj are density reachable to xk, then xi and xj are said to be density connected: Initialize and mark all data points as unclassified; For each data point xi, determine whether it is a core object. If it is a core object, build a new cluster and add xi and all the data points in its □ neighborhood to the cluster; For each data point of each core object, recursively add its density-reachable data points to the same cluster; Continue to do the above operation for unclassified data points until all data points are classified into a cluster, or there are no more data points that can be added to any cluster; The above process is repeated until all core objects have been processed and all data points have been classified into corresponding clusters.

9. The evaluation method of the job skill evaluation system according to claim 1, characterized in that: Evaluation steps: S1: Comprehensive data collection: Obtain employees’ learning progress, task completion, and online test scores from the course learning system; collect employees’ practice reports, project assignments, and their scores to assess their practical abilities; understand employees’ perceptions of their own learning through self-evaluation; obtain teachers’ evaluation feedback; receive feedback from companies to understand employees’ skill application during internships or work; S2: Skill level analysis: The skill analysis module uses a recurrent neural network to conduct in-depth analysis of the collected learning information and predict employees' future skill levels or learning trends. Through training, it learns about employees' individual differences and learning styles. Based on employees' individual differences and learning styles, it makes personalized adjustments to the model and analyzes the breadth and depth of employees' learning coverage in various knowledge points. S3: Build and maintain job knowledge graph: Based on the job requirement information on the recruitment platform, use the DBSCAN algorithm to determine the weight distribution of knowledge points; S4: receiving skill evaluation results: the job analysis module receives the skill evaluation results transmitted by the skill analysis module; S5: Generate evaluation results based on knowledge point weights: Generate employee job skill evaluation results based on knowledge point weights in the job knowledge graph.

Images