Employment guidance method based on enterprise recruitment data and college entrance examination application filling linkage

Abstract

The invention discloses an employment guidance method based on enterprise recruitment data and college entrance examination application filling linkage. The method comprises the steps of S1, collecting social enterprise recruitment data, public unit plan recruitment data and public unit actual admission data in real time; s2, integrating the cleaned data into a unified analysis data set through a data tagging technology; s3, constructing an occupational development prediction model; s4, generating an examinee personalized portrait model of each college entrance examination volunteer filling examinee by using a multi-dimensional data analysis method; s5, dynamically generating a volunteer filling professional selection suggestion for each college entrance examination volunteer filling examinee based on the vocational development prediction model and the examinee personalized portrait model; s6, constructing a Pareto optimal solution set by using a multi-objective optimization algorithm based on the volunteer filling professional selection suggestion; and S7, carrying out statistical analysis on the Pareto optimal solution set and the selection result of the college entrance examination volunteer filling examinees. According to the method, the scientificity and practicability of volunteer filling guidance are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of employment guidance, and in particular to an employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting. Background Art

[0002] As filling out college entrance examination applications has gradually become a key part of students' career planning, more and more research and tools are trying to introduce data technology into the decision-making process of application filling in order to improve the scientificity and rationality of students' choices. However, the current application filling guidance methods still have significant shortcomings and cannot meet the growing personalized and precise needs of candidates and parents.

[0003] At present, most guidance on filling out college entrance examination applications mainly relies on manual consultation or static analysis based on historical data. Manual consultation is usually provided by experienced experts or educational institutions, who provide advice by analyzing candidates' scores, interests and college admissions information. Although it has a certain degree of pertinence, it is difficult to systematize and promote on a large scale because it relies on individual experience. At the same time, the decision-making process of manual consultation usually lacks comprehensive consideration of future employment market needs, and it is difficult to accurately reflect the actual correlation between professional choices and career development.

[0004] In recent years, some volunteer application systems based on artificial intelligence have begun to be used, using machine learning algorithms to model candidate data and university information and trying to add some employment market data for correlation analysis. However, the existing technology has exposed the following problems in actual application: First, the existing system is not able to respond to the dynamic changes in the employment market, and the analysis data is relatively limited, making it difficult to accurately reflect real-time job requirements and skill changes; second, the existing technology can only make personalized recommendations based on simple interest or score data, lacking multi-dimensional mining of candidate portraits, resulting in insufficient pertinence and scientificity of the recommendation results; in addition, existing methods usually fail to provide optimization options or trade-offs, and candidates lack effective support for their right to choose between satisfaction and career prospects.

[0005] To sum up, the existing technologies have significant defects in the dynamic adaptation, personalized recommendations and option weighing of volunteer applications, and it is difficult to meet the diverse needs of students in volunteer applications. There is an urgent need for an innovative method that can combine dynamic employment market demands, personalized candidate portraits and multi-objective optimization to solve the above problems. Summary of the invention

[0006] One purpose of the present invention is to propose an employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting. The present invention significantly improves the scientificity and practicality of volunteer reporting guidance, and provides efficient and reliable technical support for students' academic choices and career planning.

[0007] According to an embodiment of the present invention, an employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer application includes the following steps:

[0008] S1. Real-time collection of recruitment data of social enterprises, planned recruitment data of public institutions and actual admission data of public institutions;

[0009] S2. Clean the collected social enterprise recruitment data, public unit planned recruitment data and public unit actual admission data, and integrate the cleaned data into a unified analysis data set through data labeling technology;

[0010] S3. Construct a career development prediction model, which is used to analyze the changing trends of job requirements, job skill requirements, and salary levels in corporate recruitment information;

[0011] S4. Combine the personal information of the candidates who fill in the college entrance examination voluntary application form and use the multidimensional data analysis method to generate a personalized portrait model of each candidate who fills in the college entrance examination voluntary application form;

[0012] S5. Dynamically generate suggestions for each candidate to fill in their college entrance examination application form based on the career development prediction model and the candidate's personalized portrait model;

[0013] S6. Based on the recommendations for major selection in the college entrance examination, a multi-objective optimization algorithm is used to construct a Pareto optimal solution set. The Pareto optimal solution set is formed by optimizing the candidate satisfaction target, the job market matching target, and the professional competition intensity target. The recommended results of different trade-offs in the Pareto optimal solution set are fed back to the college entrance examination volunteer candidates for their selection;

[0014] S7. Conduct statistical analysis on the Pareto optimal solution set and the selection results of the candidates filling out their college entrance examination applications, and feed back the analysis results to colleges and enterprises through a data interface. The feedback is used to guide colleges and universities to adjust their major settings and enrollment plans and to guide enterprises to optimize recruitment needs and job definitions.

[0015] Optionally, the S1 includes the following steps:

[0016] S11. Collect information on job demand, job skill requirements, salary levels and industry trends from social enterprises, and use the trend of changes in the number of recruitment positions to evaluate the development prospects of related majors, including fluctuations in professional demand, increases and decreases in the number of positions and trends in salary changes, to determine the future development potential of the major and employment market trends, and obtain recruitment data for social enterprises;

[0017] S12. Collect recruitment notice data issued by public institutions, civil service system, central enterprises and state-owned enterprises, including recruitment position information, planned number of people, educational requirements and professional requirements, and use data to analyze which majors have advantages when entering specific units, and obtain planned recruitment data of public institutions;

[0018] S13. Collect admission publicity data from public institutions, civil service system, central enterprises and state-owned enterprises, including the gender, education background, graduation school and graduation major of the admitted personnel, and combine it with the planned recruitment data to determine which majors and schools are more likely to enter the corresponding units. Through data analysis, clarify the correlation between "unit-major-school" to provide candidates with employment path guidance from major to unit, and reversely deduce adaptability suggestions from unit to school to obtain the actual admission data of public institutions.

[0019] Optionally, S2 includes the following steps:

[0020] S21. Detect and eliminate redundant data from the collected social enterprise recruitment data, public unit planned recruitment data, and public unit actual admission data to obtain a non-redundant analysis data set D′ i ;

[0021] S22. Non-redundant analysis data set D′ i Perform format standardization and define a standardized mapping function to transform each data d∈D′ i Converted into data d′ in a unified format. The attributes of data d′ include job demand information, job skill requirements information, salary level information, industry trend information, recruitment plan number information, admitted personnel major information and admitted personnel college information, forming a standardized analysis data set

[0022] S23. Standardized analysis of data sets Detect and remove erroneous data to obtain a cleaned analysis data set

[0023] S24. Analysis of the cleaned data set Label the data by defining a label generation function for each piece of data Generate the corresponding label set {t1,t2,…,t n}, after labeling is completed, a labeled analysis data set is formed

[0024] S25. Integrate labeled analysis datasets from different sources into a unified analysis dataset D norm .

[0025] Optionally, S3 includes the following steps:

[0026] S31. Based on a unified analysis data set D norm Construct a job demand change trend model, perform time series analysis on the extracted job demand information R, and obtain the job demand function P(t) at time t:

[0027]

[0028] Among them, a0 and a k is the regression coefficient parameter, f k (t) is a time-dependent basis function, which is used to characterize the changing characteristics of job requirements at different time scales. i is the i-th job demand data point extracted from the unified analysis data set, |R| is the number of job demand data points, t is the time dimension parameter, and n represents the number of data points of job demand information;

[0029] S32. Build a job skill requirement change trend model based on a unified analysis data set, normalize and aggregate the extracted skill requirement set S and corresponding frequencies, and obtain the skill requirement model M. S :

[0030]

[0031] Among them, Freq(s i ) is the skill i The number of occurrences in the unified analysis data set, ψ s With χ s is a control parameter in the skill demand model, which highlights or weakens the importance of specific skills in the overall skill demand structure through nonlinear transformation. Using the extracted job demand data x Weighting the skill demand distribution to reflect the impact of job demand changes at the skill level;

[0032] S33. Build a salary level change trend model based on a unified analysis data set, extract the salary level information set L from the data set, and build a salary level model M through distribution fitting L :

[0033]

[0034] Among them, μ l With σ l They represent the mean and standard deviation of the salary estimated from the unified analysis data set, which are used to characterize the center and dispersion of the salary level distribution, ν l To adjust the parameters of the salary weighting term, (1+r i ) νl The acquired job demand data iIntegrate it into the salary distribution, so that the salary level model M L Present salary trends based on job demand;

[0035] S34. Combining the job demand change trend model P(t), the job skill demand change trend model and the salary level change trend model, a career development prediction model Q(p,s,l) of job demand and professional choice is constructed:

[0036]

[0037] Among them, p is a parameter related to professional selection, s is a parameter related to skill demand, l is a parameter related to salary level, α, β, and γ are weight coefficients, which are used to balance the influence of job demand trend, skill demand distribution and salary change trend in the comprehensive matching degree, ρ1, ρ2, ρ3 and τ are parameters that control the contribution and nonlinear transformation of different dimensions in the matching degree model.

[0038] Optionally, S4 includes the following steps:

[0039] S41. Extract the multidimensional feature vector set V = {v1, v2, ..., v n}, where each eigenvector v i Defined as:

[0040] v i =[I i ,A i ,R i ,F i ];

[0041] Among them, I i represents the interest preference score of candidate i, A i represents the academic performance of candidate i, R i represents the region selection preference of candidate i, F i represents the family background characteristics of candidate i;

[0042] S42, constructing a portrait feature weight matrix W based on a multi-dimensional feature vector set extracted from the personal information of the candidates filling out the college entrance examination application form;

[0043] S43, using the multi-dimensional feature vector set and the portrait feature weight matrix to comprehensively calculate the candidate portrait features, and obtain the individual portrait feature value U of each candidate i :

[0044]

[0045] Among them, w k is the weight of the k-th image feature, v ikis the score of candidate i on the kth portrait feature;

[0046] S44, set the image feature value U of all candidates i Conduct cluster analysis to form a personalized portrait model M for candidates P .

[0047] Optionally, S5 includes the following steps:

[0048] S51. Define the dynamic recommendation score function D for each candidate i for major j based on the career development prediction model and the candidate personalized portrait model. ij :

[0049] D ij =α1·Q(p j ,s j ,l j )+β1·U i +γ1·C ij ;

[0050] Among them, α1, β1, γ1 are the weight parameters of the score function, Q(p j ,s j ,l j ) represents the career development matching degree of major j, U i represents the profile feature value of candidate i, C ij represents the compatibility score between candidate i’s interest preference and major j. The compatibility score is calculated by the candidate’s profile interest dimension I i and professional feature keyword set K j Similarity calculation:

[0051]

[0052] Among them, K j,k represents the kth characteristic keyword of major j, Sim(I i ,K j,k ) is the similarity function between interest preference and keywords;

[0053] S52. Using dynamic recommendation score function D ij Generate a set of professional selection suggestions S for each candidate i based on his / her interests, abilities and career development potential i :

[0054] S i ={j|D ij ≥δ};

[0055] Among them, δ is the set recommendation score threshold, satisfying D ij Major j greater than or equal to the threshold is recommended to candidate i, and the major selection suggestion set S iSort and display the matching score, job market demand trend and interest preference compatibility score of each recommended major.

[0056] Optionally, the S6 includes the following steps:

[0057] S61. Suggested selection of majors based on voluntary application S i And the dynamic recommendation score function D ij Construct a multi-objective optimization problem, the objectives include:

[0058] Candidate satisfaction goal: maximize candidate i's recommendation score D for major j ij ;

[0059] Job market matching goal: maximize the job market demand trend E of major j j ;

[0060] Professional competition intensity objective: Minimize the competition intensity C of major j j :

[0061]

[0062] Among them, TotalDemand j Indicates the total number of seats required for major j, AvailableSeats j represents the planned enrollment number of major j;

[0063] The objective function set is constructed by integrating the candidate satisfaction target, the job market matching target and the professional competition intensity target:

[0064]

[0065] S62, using NSGA-II algorithm to calculate the objective function set F(S i ) is iteratively optimized to obtain the Pareto frontier solution set P:

[0066]

[0067] in, represents the kth optimization solution, Represents the Pareto optimal dominance relationship, each optimization solution To achieve the best balance among candidate satisfaction, job market matching and professional competition intensity;

[0068] S63. Optimize the solution for each Pareto frontier Calculate its comprehensive trade-off score W k :

[0069]

[0070] Among them, ω1, ω2, and ω3 are target weight coefficients, which are used to regulate the influence of different targets in the comprehensive score;

[0071] S64. Select the solution with the highest comprehensive trade-off score W from the Pareto frontier solution set P. k Optimization plan Generate the final professional selection recommendation for candidate i:

[0072]

[0073] Recommend the final major The major selection in the program is sorted according to the dynamic recommendation score, and the satisfaction score, job market matching and competition intensity index of each major are output.

[0074] The beneficial effects of the present invention are:

[0075] (1) The present invention constructs a job demand change trend model, a job skill demand change trend model and a salary level change trend model, and effectively combines time series analysis, clustering algorithm and distribution fitting technology to achieve high-precision modeling and real-time updating of dynamic changes in the employment market. By dynamically updating the career development prediction model, it can not only accurately capture the market demand fluctuations for different majors, but also provide real-time feedback on job skill requirements and salary changes, and provide candidates with targeted professional selection suggestions. Experimental results show that the accuracy of the employment market dynamic analysis of the present invention is improved by 12.5% ​​compared with the traditional method, and has achieved significant advantages in predicting the matching degree between majors and job requirements.

[0076] (2) The present invention is the first to use multi-dimensional data (including interest preferences, academic performance, regional selection preferences, and family background characteristics) to construct a personalized portrait model for candidates, and to determine the weights of portrait features through hierarchical analysis and weighted analysis, thereby effectively improving the personalization level and pertinence of recommendation results. Compared with the limitation of existing methods that only make recommendations based on a single data dimension, the present invention combines the compatibility analysis between interest preferences and career needs, and significantly improves the rationality of professional selection recommendations through a dynamic recommendation score function. Test results show that the satisfaction of candidates with the recommendation solutions generated by the present invention is 15% higher than that of traditional methods, and the score in the comprehensive matching evaluation is increased by 8%.

[0077] (3) The present invention introduces a multi-objective optimization algorithm into dynamic major recommendation, and constructs a Pareto frontier solution set by combining the three core goals of candidate satisfaction, job market matching, and professional competition intensity. The recommendation scheme is optimized through a dynamic recommendation score function and trade-off analysis. Through the dynamic optimization of the Pareto frontier solution set, multiple trade-off options can be provided for candidates to choose from, and multi-dimensional decision-making can be supported through recommendation visualization results. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0079] Figure 1 This is a flow chart of an employment guidance method proposed by the present invention based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting. DETAILED DESCRIPTION

[0080] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0081] refer to Figure 1 , an employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer filling, comprising the following steps:

[0082] S1. Real-time collection of social enterprise recruitment data, public unit planned recruitment data and public unit actual admission data;

[0083] S2. Clean the collected social enterprise recruitment data, public unit planned recruitment data and public unit actual admission data, and integrate the cleaned data into a unified analysis data set through data labeling technology;

[0084] S3. Construct a career development prediction model, which is used to analyze the changing trends of job requirements, job skill requirements, and salary levels in corporate recruitment information;

[0085] S4. Combine the personal information of the candidates who fill in the college entrance examination application form and use the multidimensional data analysis method to generate a personalized portrait model of each candidate who fills in the college entrance examination application form;

[0086] S5. Dynamically generate suggestions for each candidate to fill out their college entrance examination application form based on the career development prediction model and the candidate's personalized portrait model;

[0087] S6. Based on the recommendations for major selection in the college entrance examination, a multi-objective optimization algorithm is used to construct a Pareto optimal solution set. The Pareto optimal solution set is formed by optimizing the candidate satisfaction target, the job market matching target, and the professional competition intensity target. The recommended results of different trade-offs in the Pareto optimal solution set are fed back to the college entrance examination volunteer candidates for their selection;

[0088] S7. Conduct statistical analysis on the Pareto optimal solution set and the selection results of candidates filling out college entrance examination applications, and feed back the analysis results to colleges and enterprises through data interfaces. The feedback is used to guide colleges and universities to adjust their major settings and enrollment plans, and to guide enterprises to optimize recruitment needs and job definitions.

[0089] In this implementation, S1 includes the following steps:

[0090] S11. Collect information on job demand, job skill requirements, salary levels and industry trends from social enterprises, and use the trend of changes in the number of recruitment positions to evaluate the development prospects of related majors, including fluctuations in professional demand, increases and decreases in the number of positions and trends in salary changes, to determine the future development potential of the major and employment market trends, and obtain recruitment data for social enterprises;

[0091] S12. Collect recruitment notice data issued by public institutions, civil service system, central enterprises and state-owned enterprises, including recruitment position information, planned number of people, educational requirements and professional requirements, and use data to analyze which majors have advantages when entering specific units, and obtain planned recruitment data of public institutions;

[0092] S13. Collect admission publicity data from public institutions, civil service system, central enterprises and state-owned enterprises, including the gender, education background, graduation school and graduation major of the admitted personnel, and combine it with the planned recruitment data to determine which majors and schools are more likely to enter the corresponding units. Through data analysis, clarify the correlation between "unit-major-school" to provide candidates with employment path guidance from major to unit, and reversely deduce adaptability suggestions from unit to school to obtain the actual admission data of public institutions.

[0093] In this implementation, S2 includes the following steps:

[0094] S21. Detect and eliminate redundant data from the collected social enterprise recruitment data, public unit planned recruitment data, and public unit actual admission data to obtain a non-redundant analysis data set D′ i ;

[0095] S22. Non-redundant analysis data set D′ i Perform format standardization and define a standardized mapping function to transform each data d∈D′ i Converted into data d′ in a unified format. The attributes of data d′ include job demand information, job skill requirements information, salary level information, industry trend information, recruitment plan number information, admitted personnel major information and admitted personnel college information, forming a standardized analysis data set

[0096] S23. Standardized analysis of data sets Detect and remove erroneous data to obtain a cleaned analysis data set

[0097] S24. Analysis of the cleaned data set Label the data by defining a label generation function for each piece of data Generate the corresponding label set {t1,t2,…,t n}, after labeling is completed, a labeled analysis data set is formed

[0098] S25. Integrate labeled analysis datasets from different sources into a unified analysis dataset D norm .

[0099] In this implementation, S3 includes the following steps:

[0100] S31. Based on a unified analysis data set D norm Construct a job demand change trend model, perform time series analysis on the extracted job demand information R, and obtain the job demand function P(t) at time t:

[0101]

[0102] Among them, a0 and a k is the regression coefficient parameter, f k (t) is a time-dependent basis function, which is used to characterize the changing characteristics of job requirements at different time scales. i is the i-th job demand data point extracted from the unified analysis data set, |R| is the number of job demand data points, t is the time dimension parameter, and n represents the number of data points of job demand information;

[0103] S32. Build a job skill requirement change trend model based on a unified analysis data set, normalize and aggregate the extracted skill requirement set S and corresponding frequencies, and obtain the skill requirement model M. S :

[0104]

[0105] Among them, Freq(s i ) is the skill i The number of occurrences in the unified analysis data set, ψ s With χ s is a control parameter in the skill demand model, which highlights or weakens the importance of specific skills in the overall skill demand structure through nonlinear transformation. Using the extracted job demand data x Weighting the skill demand distribution to reflect the impact of job demand changes at the skill level;

[0106] S33. Build a salary level change trend model based on a unified analysis data set, extract the salary level information set L from the data set, and build a salary level model M through distribution fitting L :

[0107]

[0108] Among them, μ l With σ l They represent the mean and standard deviation of the salary estimated from the unified analysis data set, which are used to characterize the center and dispersion of the salary level distribution, ν l To adjust the parameters of the salary weighting term, (1+r i ) νl The acquired job demand data i Integrate it into the salary distribution, so that the salary level model M L Present salary trends based on job demand;

[0109] S34. Combining the job demand change trend model P(t), the job skill demand change trend model and the salary level change trend model, a career development prediction model Q(p,s,l) of job demand and professional choice is constructed:

[0110]

[0111] Among them, p is a parameter related to professional selection, s is a parameter related to skill demand, l is a parameter related to salary level, α, β, and γ are weight coefficients, which are used to balance the influence of job demand trend, skill demand distribution and salary change trend in the comprehensive matching degree, ρ1, ρ2, ρ3 and τ are parameters that control the contribution and nonlinear transformation of different dimensions in the matching degree model.

[0112] In this implementation, S4 includes the following steps:

[0113] S41. Extract the multidimensional feature vector set V = {v1, v2, ..., v n}, where each eigenvector v i Defined as:

[0114] v i =[I i ,A i ,R i ,F i ];

[0115] Among them, I i represents the interest preference score of candidate i, A i represents the academic performance of candidate i, Ri represents the region selection preference of candidate i, F i represents the family background characteristics of candidate i;

[0116] S42, constructing a portrait feature weight matrix W based on a multi-dimensional feature vector set extracted from the personal information of the candidates filling out the college entrance examination application form;

[0117] S43, using the multi-dimensional feature vector set and the portrait feature weight matrix to comprehensively calculate the candidate portrait features, and obtain the individual portrait feature value U of each candidate i :

[0118]

[0119] Among them, w k is the weight of the k-th image feature, v ik is the score of candidate i on the kth portrait feature;

[0120] S44, set the image feature value U of all candidates i Conduct cluster analysis to form a personalized portrait model M for candidates P .

[0121] In this implementation, S5 includes the following steps:

[0122] S51. Define the dynamic recommendation score function D for each candidate i for major j based on the career development prediction model and the candidate personalized portrait model. ij :

[0123] D ij =α1·Q(p j ,s j ,l j )+β1·U i +γ1·C ij ;

[0124] Among them, α1, β1, γ1 are the weight parameters of the score function, Q(p j ,s j ,l j ) represents the career development matching degree of major j, U i represents the profile feature value of candidate i, C ij represents the compatibility score between candidate i’s interest preference and major j. The compatibility score is calculated by the candidate’s profile interest dimension I i and professional feature keyword set K j Similarity calculation:

[0125]

[0126] Among them, K j,krepresents the kth characteristic keyword of major j, Sim(I i ,K j,k ) is the similarity function between interest preference and keywords;

[0127] S52. Using dynamic recommendation score function D ij Generate a set of professional selection suggestions S for each candidate i based on his / her interests, abilities and career development potential i :

[0128] S i ={j|D ij ≥δ};

[0129] Among them, δ is the set recommendation score threshold, satisfying D ij Major j greater than or equal to the threshold is recommended to candidate i, and the major selection suggestion set S i Sort and display the matching score, job market demand trend and interest preference compatibility score of each recommended major.

[0130] In this implementation, S6 includes the following steps:

[0131] S61. Suggested selection of majors based on voluntary application S i And the dynamic recommendation score function D ij Construct a multi-objective optimization problem, the objectives include:

[0132] Candidate satisfaction goal: maximize candidate i's recommendation score D for major j ij ;

[0133] Job market matching goal: maximize the job market demand trend E of major j j ;

[0134] Professional competition intensity objective: Minimize the competition intensity C of major j j :

[0135]

[0136] Among them, TotalDemand j Indicates the total number of seats required for major j, AvailableSeats j represents the planned enrollment number of major j;

[0137] The objective function set is constructed by integrating the candidate satisfaction target, the job market matching target and the professional competition intensity target:

[0138]

[0139] S62, using NSGA-II algorithm to calculate the objective function set F(Si ) is iteratively optimized to obtain the Pareto frontier solution set P:

[0140]

[0141] in, represents the kth optimization solution, Represents the Pareto optimal dominance relationship, each optimization solution To achieve the best balance among candidate satisfaction, job market matching and professional competition intensity;

[0142] S63. Optimize the solution for each Pareto frontier Calculate its comprehensive trade-off score W k :

[0143]

[0144] Among them, ω1, ω2, and ω3 are target weight coefficients, which are used to regulate the influence of different targets in the comprehensive score;

[0145] S64. Select the solution with the highest comprehensive trade-off score W from the Pareto frontier solution set P. k Optimization plan Generate the final professional selection recommendation for candidate i:

[0146]

[0147] Recommend the final major The major selection in the program is sorted according to the dynamic recommendation score, and the satisfaction score, job market matching and competition intensity index of each major are output.

[0148] Embodiment 1:

[0149] The embodiment is applied in a high school in District A. Xiao Ming, a senior student in a high school in District A, is preparing to fill in his application. Xiao Ming is interested in computer science and likes programming, but he is not sure whether to choose "computer science and technology" or "information security". At the same time, his parents hope that the choice of major can guarantee future employment competitiveness, so they put forward high requirements for the guidance of filling in college entrance examination applications. The school organized a simulated guidance activity for filling in applications, and used traditional methods and the method of the present invention to guide Xiao Ming's choice of application.

[0150] The traditional method is based on the analysis of college admission scores and enrollment plans in previous years. Xiao Ming's total college entrance examination score is 640 points. According to the college admission scores in 2023, the simulation system recommends that he apply for the "Computer Science and Technology" major of a 985 university, with an admission probability of 85%. However, the recommendation of the traditional method ignores the changes in the future job market and the interests and preferences of students, and fails to provide a detailed comparative analysis of the "Information Security" major.

[0151] This system collects real-time recruitment data of enterprises across the country from 2021 to 2024, including job requirements of Internet companies, financial technology companies and national units. Specifically, it is found that in the first quarter of 2024, the demand for information security positions across the country has shown a quarterly increase, and the number of recruitments has increased from 3,200 positions in the previous quarter to 3,900, an increase of 21.9%. Through further cleaning and processing of recruitment data, duplicate positions and non-professional technical positions are excluded, and effective job demand data is obtained:

[0152] Computer science and technology related positions: Skill requirements cover a wide range, and the median salary is 15,000 yuan / month;

[0153] Information security-related positions: Demand has increased significantly, with a median salary of RMB 18,000 per month, but there are higher requirements for advanced skills such as “network attack and defense” and “vulnerability detection”.

[0154] The system generates a personalized portrait from Xiaoming’s multi-dimensional data on his interests, academic performance, and family background:

[0155] Interests: Programming (score 89 points), Cybersecurity (score 83 points);

[0156] Academic performance: Mathematics 130 points, Physics 132 points, Chemistry 120 points, overall excellent performance;

[0157] Regional preference: Region A is preferred;

[0158] Family background: Both parents are corporate employees and tend to choose majors with high employment stability.

[0159] The comprehensive portrait shows that Xiao Ming has a high potential score in computer and information security related fields, and is particularly well matched with technical R&D positions driven by his logical ability and interest.

[0160] The job demand change trend model constructed by the system predicts that in the next five years, the average annual growth rate of job demand in the information security field will be 12.5%, much higher than the 7.3% of computer science and technology. In addition, among the skill demand weights in the information security field, the score weight of "network attack and defense" skill is 0.36, and the score of "encryption algorithm" skill is 0.24, indicating that this professional direction has both depth and breadth of professional skill requirements.

[0161] Based on Xiao Ming's personal portrait and career development prediction model, the system generates a dynamic recommendation score:

[0162] "Computer Science and Technology" major recommendation score: 85.4;

[0163] "Information Security" professional recommendation score: 89.7.

[0164] In the Pareto optimal solution set, the "Information Security" major is listed as the first choice by the system due to its significant advantages in salary expectations and demand growth, but the "Computer Science and Technology" major is listed as an alternative option for Xiao Ming's family to refer to due to its wide demand and higher admission score.

[0165] The system displays all analysis results to Xiao Ming and his parents in an intuitive and visual way:

[0166] The employment trend comparison chart shows that the job growth rate of "Information Security" from 2024 to 2029 is significantly higher than that of "Computer Science and Technology";

[0167] The matching radar chart highlights Xiaoming’s personal interests and skills in the field of “information security”;

[0168] The student satisfaction analysis shows that Xiao Ming’s satisfaction with the recommendation for the “Information Security” direction is 92%.

[0169] In the end, with the support of his parents, Xiao Ming chose the "Information Security" major and expressed confidence in his future career development.

[0170] This embodiment demonstrates the practical application of the method of the present invention in the guidance of filling out college entrance examination applications through real data and a complete process, which significantly improves the scientificity, rationality and personalization level of the recommendation scheme, and solves the technical problems that the traditional method cannot dynamically adapt to the employment market demand and lacks personalized recommendations.

[0171] The present invention constructs a job demand change trend model, a job skill demand change trend model and a salary level change trend model, and effectively combines time series analysis, clustering algorithm and distribution fitting technology to achieve high-precision modeling and real-time updating of dynamic changes in the employment market. By dynamically updating the career development prediction model, it can not only accurately capture the market demand fluctuations for different majors, but also provide real-time feedback on job skill requirements and salary changes, and provide candidates with targeted professional selection suggestions. Experimental results show that the employment market dynamic analysis accuracy of the present invention is improved by 12.5% ​​compared with traditional methods, and has achieved significant advantages in predicting the matching degree between majors and job requirements.

[0172] The present invention is the first to use multi-dimensional data (including interest preferences, academic performance, regional selection preferences, and family background characteristics) to construct a personalized portrait model for candidates, and determines the weights of portrait features through hierarchical analysis and weighted analysis, which effectively improves the personalization level and pertinence of recommendation results. Compared with the limitations of existing methods that only make recommendations based on a single data dimension, the present invention combines the compatibility analysis between interest preferences and career needs, and significantly improves the rationality of professional selection suggestions through a dynamic recommendation score function. Test results show that the satisfaction of candidates with the results of the recommendation scheme generated by the present invention is 15% higher than that of traditional methods, and the score in the comprehensive matching evaluation is increased by 8%.

[0173] The present invention introduces a multi-objective optimization algorithm in dynamic major recommendation, and constructs a Pareto frontier solution set based on the three core goals of candidate satisfaction, job market matching and professional competition intensity. The recommendation scheme is optimized through dynamic recommendation scoring function and trade-off analysis. Through dynamic optimization of the Pareto frontier solution set, multiple trade-off options can be provided for candidates to choose from, and multi-dimensional decision-making can be supported through recommendation visualization results.

[0174] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting, characterized in that: The steps include: S1. Real-time collection of recruitment data of social enterprises, planned recruitment data of public institutions and actual admission data of public institutions; S2. Clean the collected social enterprise recruitment data, public unit planned recruitment data and public unit actual admission data, and integrate the cleaned data into a unified analysis data set through data labeling technology; S3. Construct a career development prediction model, which is used to analyze the changing trends of job requirements, job skill requirements, and salary levels in corporate recruitment information; S4. Combine the personal information of the candidates who fill in the college entrance examination voluntary application form and use the multidimensional data analysis method to generate a personalized portrait model of each candidate who fills in the college entrance examination voluntary application form; S5. Dynamically generate suggestions for each candidate to fill in their college entrance examination application form based on the career development prediction model and the candidate's personalized portrait model; S6. Based on the recommendations for major selection in the college entrance examination, a multi-objective optimization algorithm is used to construct a Pareto optimal solution set. The Pareto optimal solution set is formed by optimizing the candidate satisfaction target, the job market matching target, and the professional competition intensity target. The recommended results of different trade-offs in the Pareto optimal solution set are fed back to the college entrance examination volunteer candidates for their selection; S7. Conduct statistical analysis on the Pareto optimal solution set and the selection results of the candidates filling out their college entrance examination applications, and feed back the analysis results to colleges and enterprises through a data interface. The feedback is used to guide colleges and universities to adjust their major settings and enrollment plans and to guide enterprises to optimize recruitment needs and job definitions.

2. According to claim 1, a method for employment guidance based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting is characterized in that: The S1 comprises the following steps: S11. Collect information on job demand, job skill requirements, salary levels and industry trends from social enterprises, and use the trend of changes in the number of recruitment positions to evaluate the development prospects of related majors, including fluctuations in professional demand, increases and decreases in the number of positions and trends in salary changes, to determine the future development potential of the major and employment market trends, and obtain recruitment data for social enterprises; S12. Collect recruitment notice data issued by public institutions, civil service system, central enterprises and state-owned enterprises, including recruitment position information, planned number of people, educational requirements and professional requirements, and use data to analyze which majors have advantages when entering specific units, and obtain planned recruitment data of public institutions; S13. Collect admission publicity data from public institutions, civil service system, central enterprises and state-owned enterprises, including the gender, education background, graduation school and graduation major of the admitted personnel, and combine it with the planned recruitment data to determine which majors and schools are more likely to enter the corresponding units. Through data analysis, clarify the correlation between "unit-major-school" to provide candidates with employment path guidance from major to unit, and reversely deduce adaptability suggestions from unit to school to obtain the actual admission data of public institutions.

3. The employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting according to claim 1 is characterized in that: The S2 comprises the following steps: S21. Detect and eliminate redundant data from the collected social enterprise recruitment data, public unit planned recruitment data, and public unit actual admission data to obtain a non-redundant analysis data set D′ i ; S22. Non-redundant analysis data set D′ i Perform format standardization and define a standardized mapping function to transform each data d∈D′ i Converted into data d′ in a unified format. The attributes of data d′ include job demand information, job skill requirements information, salary level information, industry trend information, recruitment plan number information, admitted personnel major information and admitted personnel college information, forming a standardized analysis data set S23. Standardized analysis of data sets Detect and remove erroneous data to obtain a cleaned analysis data set S24. Analysis of the cleaned data set Label the data by defining a label generation function for each piece of data Generate the corresponding label set {t1,t2,…,t n }, after labeling is completed, a labeled analysis data set is formed S25. Integrate labeled analysis datasets from different sources into a unified analysis dataset D norm .

4. The employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting according to claim 1 is characterized in that: The S3 comprises the following steps: S31. Based on a unified analysis data set D norm Construct a job demand change trend model, perform time series analysis on the extracted job demand information R, and obtain the job demand function P(t) at time t: Among them, a0 and a k is the regression coefficient parameter, f k (t) is a time-dependent basis function, which is used to characterize the changing characteristics of job requirements at different time scales. i is the i-th job demand data point extracted from the unified analysis data set, |R| is the number of job demand data points, t is the time dimension parameter, and n represents the number of data points of job demand information; S32. Build a job skill requirement change trend model based on a unified analysis data set, normalize and aggregate the extracted skill requirement set S and corresponding frequencies, and obtain the skill requirement model M. S : Among them, Freq(s i ) is the skill i The number of occurrences in the unified analysis data set, ψ s With χ s is a control parameter in the skill demand model, which highlights or weakens the importance of specific skills in the overall skill demand structure through nonlinear transformation. Using the extracted job demand data x Weighting the skill demand distribution to reflect the impact of job demand changes at the skill level; S33. Build a salary level change trend model based on a unified analysis data set, extract the salary level information set L from the data set, and build a salary level model M through distribution fitting L : Among them, μ l With σ l They represent the mean and standard deviation of the salary estimated from the unified analysis data set, which are used to characterize the center and dispersion of the salary level distribution, ν l To adjust the parameters of the salary weighting term, (1+r i ) νl The acquired job demand data i Integrate it into the salary distribution, so that the salary level model M L Present salary trends based on job demand; S34. Combining the job demand change trend model P(t), the job skill demand change trend model and the salary level change trend model, a career development prediction model Q(p,s,l) of job demand and professional choice is constructed: Among them, p is a parameter related to professional selection, s is a parameter related to skill demand, l is a parameter related to salary level, α, β, and γ are weight coefficients, which are used to balance the influence of job demand trend, skill demand distribution and salary change trend in the comprehensive matching degree, ρ1, ρ2, ρ3 and τ are parameters that control the contribution and nonlinear transformation of different dimensions in the matching degree model.

5. The employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting according to claim 1 is characterized in that: The S4 comprises the following steps: S41. Extract the multidimensional feature vector set V = {v1, v2, ..., v n }, where each eigenvector v i Defined as: v i =[I i ,A i ,R i ,F i ]; Among them, I i represents the interest preference score of candidate i, A i represents the academic performance of candidate i, R i represents the region selection preference of candidate i, F i represents the family background characteristics of candidate i; S42, constructing a portrait feature weight matrix W based on a multi-dimensional feature vector set extracted from the personal information of the candidates filling out the college entrance examination application form; S43, using the multi-dimensional feature vector set and the portrait feature weight matrix to comprehensively calculate the candidate portrait features, and obtain the individual portrait feature value U of each candidate i : Among them, w k is the weight of the k-th image feature, v ik is the score of candidate i on the kth portrait feature; S44, set the image feature value U of all candidates i Conduct cluster analysis to form a personalized portrait model M for candidates P .

6. The employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting according to claim 1 is characterized in that: The S5 comprises the following steps: S51. Define the dynamic recommendation score function D for each candidate i for major j based on the career development prediction model and the candidate personalized portrait model. ij : D ij =α1·Q(p j ,s j ,l j )+β1·U i +γ1·C ij ; Among them, α1, β1, γ1 are the weight parameters of the score function, Q(p j ,s j ,l j ) represents the career development matching degree of major j, U i represents the profile feature value of candidate i, C ij represents the compatibility score between candidate i’s interest preference and major j. The compatibility score is calculated by the candidate’s profile interest dimension I i and professional feature keyword set K j Similarity calculation: Among them, K j,k represents the kth characteristic keyword of major j, Sim(I i ,K j,k ) is the similarity function between interest preference and keywords; S52. Using dynamic recommendation score function D ij Generate a set of professional selection suggestions S for each candidate i based on his / her interests, abilities and career development potential i : S i ={j|D ij ≥δ}; Among them, δ is the set recommendation score threshold, satisfying D ij Major j greater than or equal to the threshold is recommended to candidate i, and the major selection suggestion set S i Sort and display the matching score, job market demand trend and interest preference compatibility score of each recommended major.

7. The employment guidance method based on the linkage between enterprise recruitment data and college entrance examination volunteer reporting according to claim 1 is characterized in that: The S6 comprises the following steps: S61. Suggested selection of majors based on voluntary application S i And the dynamic recommendation score function D ij Construct a multi-objective optimization problem, the objectives include: Candidate satisfaction goal: maximize candidate i's recommendation score D for major j ij ; Job market matching goal: maximize the job market demand trend E of major j j ; Professional competition intensity objective: Minimize the competition intensity C of major j j : Among them, TotalDemand j Indicates the total number of seats required for major j, AvailableSeats j represents the planned enrollment number of major j; The objective function set is constructed by integrating the candidate satisfaction target, the job market matching target and the professional competition intensity target: S62, using NSGA-II algorithm to calculate the objective function set F(S i ) is iteratively optimized to obtain the Pareto frontier solution set P: in, represents the kth optimization solution, Represents the Pareto optimal dominance relationship, each optimization solution To achieve the best balance among candidate satisfaction, job market matching and professional competition intensity; S63. Optimize the solution for each Pareto frontier Calculate its comprehensive trade-off score W k : Among them, ω1, ω2, and ω3 are target weight coefficients, which are used to regulate the influence of different targets in the comprehensive score; S64. Select the solution with the highest comprehensive trade-off score W from the Pareto frontier solution set P. k Optimization plan Generate the final professional selection recommendation for candidate i: Recommend the final major The major selection in the program is sorted according to the dynamic recommendation score, and the satisfaction score, job market matching and competition intensity index of each major are output.

Images