A career assessment method and system based on recruitment big data

Abstract

The application discloses a kind of career assessment method and system based on recruitment big data, comprising: collecting relevant recruitment information;The recruitment information is carried out data preprocessing, obtains the data set corresponding to the recruitment information;Text mining is carried out to the data set, obtains the keyword that has influence on average salary and the average salary corresponding to the keyword;The information of the job applicant is substituted into the Lasso regression model for salary prediction constructed in advance, and the salary expected value of the job applicant is obtained;According to the keyword and the regression coefficient of the Lasso regression model, the career assessment scheme of the job applicant is determined;The career assessment scheme is used to carry out career assessment to the job applicant. Solve the problem that the job seeker cannot reasonably position oneself and cannot understand market demand in time.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of evaluation technology, and particularly relates to a professional evaluation method and system based on recruitment big data. BACKGROUND

[0002] At present, most of the enterprises in China need to go through resume initial screening and multiple interviews when carrying out talent recruitment, and finally form a conclusion. The process of recruitment is actually a comprehensive professional evaluation. The professional evaluation provides an effective reference for the enterprise to evaluate talents, and to some extent, reflects the level of the quality of the subjects with data. When the enterprise develops to a certain stage, the evaluation plays an important role. At present, the demand for professional evaluation of foreign enterprises, large state-owned enterprises, state-owned enterprises and banks has become a rigid demand. However, the reality is that the evaluation standards in the enterprise are often not timely, and the talent demand and salary standards do not keep pace with the market. In addition, the traditional professional evaluation is more focused on personality and ability test, and there is a certain lag between the pace of development of the times. It often cannot quickly grasp the latest trends of market requirements for various talents. SUMMARY

[0003] In order to solve the above problems, the present application provides a professional evaluation method based on recruitment big data, comprising:

[0004] Collecting relevant recruitment information; data preprocessing is performed on the recruitment information to obtain a data set corresponding to the recruitment information;

[0005] Text mining is performed on the data set to obtain keywords that have an impact on average salary and average salary corresponding to the keywords;

[0006] The information of the job seeker is substituted into a pre-constructed Lasso regression model for salary prediction to obtain a salary expectation value of the job seeker;

[0007] According to the keywords and the regression coefficient of the Lasso regression model, a professional evaluation scheme of the job seeker is determined; and the professional evaluation scheme is used to perform professional evaluation on the job seeker.

[0008] Preferably, the data preprocessing of the recruitment information to obtain the data set corresponding to the recruitment information comprises:

[0009] Relevant recruitment information is regularly grabbed from a recruitment platform;

[0010] Data cleaning and data preprocessing are performed on the recruitment information to obtain a data set corresponding to the recruitment information.

[0011] Preferably, the text mining of the data set to obtain the keywords that have an impact on the average salary comprises:

[0012] For the text type variable in the data set, the method of text analysis is used to analyze the influence of the requirements of the position and the corresponding average salary, and obtain the keywords that have an influence on the average salary.

[0013] Preferably, the pre-constructed Lasso regression model for salary prediction is trained and tested by the data set corresponding to the recruitment information.

[0014] Preferably, the professional evaluation scheme of the job seeker is determined according to the keywords and the regression coefficients of the Lasso regression model, specifically including:

[0015] Select keywords from the keywords that represent the average salary of the job seeker, including but not limited to education level, work experience, professional quality, work experience and software mastery;

[0016] The regression coefficient is used as the weight of the education level and the work experience; the average salary corresponding to the keyword is used as the weight of the professional quality, the work experience and the software mastery; and the weight is standardized;

[0017] The professional evaluation scheme of the job seeker is determined according to the keywords, the regression coefficients and the weights.

[0018] Preferably, the professional evaluation of the job seeker is performed through the professional evaluation scheme, including:

[0019] According to the evaluation scheme, each aspect of the information contained in the job seeker is scored to obtain the professional evaluation score of the job seeker.

[0020] Preferably, it further includes:

[0021] The suggested salary of the job seeker is obtained through the Lasso regression model.

[0022] The application also provides a professional evaluation system based on recruitment big data, including:

[0023] A data acquisition module is used to acquire relevant recruitment information; the recruitment information is pre-processed to obtain a data set corresponding to the recruitment information;

[0024] A mining module is used to perform text mining on the data set to obtain keywords that have an influence on the average salary;

[0025] A salary expectation value acquisition module is used to substitute the information of the job seeker into a pre-constructed Lasso regression model for salary prediction to obtain the salary expectation value of the job seeker;

[0026] The professional evaluation module is configured to determine a professional evaluation scheme for the job candidate according to the keywords and the regression coefficients of the Lasso regression model, and perform professional evaluation on the job candidate through the professional evaluation scheme.

[0027] Preferably, the professional evaluation module comprises:

[0028] The keyword selection subunit is configured to select keywords representing the average salary of the job candidate from the keywords, wherein the keywords include but are not limited to education level, working years, professional accomplishment, work experience and software mastery.

[0029] The standardization processing subunit is configured to take the regression coefficients as the weights of the education level and the working years, take the average salaries corresponding to the keywords as the weights of the professional accomplishment, the work experience and the software mastery, and perform standardization processing on the weights.

[0030] The evaluation scheme determination subunit is configured to determine a professional evaluation scheme for the job candidate according to the keywords, the regression coefficients and the weights.

[0031] Preferably, the professional evaluation module comprises:

[0032] The evaluation score acquisition subunit is configured to score each aspect of the information of the job candidate according to the evaluation scheme, and acquire a professional evaluation score of the job candidate.

[0033] Preferably, the professional evaluation module further comprises:

[0034] The recommended salary acquisition module is configured to acquire a recommended salary of the job candidate through the Lasso regression model. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 is a flowchart of a professional evaluation method based on recruitment big data provided by the present application;

[0036] Figure 2 is a system flowchart of a job candidate tested by the professional evaluation method provided by the present application;

[0037] Figure 3 is a structural schematic diagram of a professional evaluation system based on recruitment big data provided by the present application. DETAILED DESCRIPTION

[0038] In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of the present application. However, the present application can be practiced in many different ways beyond the specific embodiments described herein, and it is understood that similar modifications and adaptations of the present application can be made by those skilled in the art in the light of the above teachings without departing from the scope of the present application. Therefore, the present application is not limited to the specific implementation disclosed below.

[0039] Figure 1 is a flowchart of a career assessment method based on recruitment big data provided by the present application, comprising the following steps:

[0040] Step S101, collecting relevant recruitment information; data preprocessing is performed on the recruitment information to obtain a data set corresponding to the recruitment information.

[0041] The relevant recruitment information is regularly captured from the recruitment platform; data cleaning and data preprocessing are performed on the recruitment information to obtain a data set corresponding to the recruitment information. Data collection supports cloud collection and local collection. At the same time, the captured data is stored. In order to ensure the standardization of the collected data, the data field mapping relationship is set, that is, the source data field, the source data type, the target data field and the target data type are used to complete the writing and storage of the data, and a certain field is set as a unique identifier.

[0042] For the text type variable in the data set, the method of text analysis is used to analyze the influence of the requirements of the position and the corresponding average salary, and to obtain the keywords for career assessment. After data collection and data storage, data cleaning and data preprocessing are automatically completed according to the collected structured data to obtain a recruitment information data set. Each observation in the data set represents the recruitment information of a position.

[0043] Step S102, text mining is performed on the data set to obtain keywords that have an impact on the average salary and the average salary corresponding to the keywords.

[0044] For the text type variable in the data set, the method of text analysis is used to analyze the influence of the requirements of the position and the corresponding average salary, and to obtain the keywords for career assessment. After data collection and data storage, data cleaning and data preprocessing are automatically completed according to the collected structured data to obtain a recruitment information data set. Each observation in the data set represents the recruitment information of a position.

[0045] Specifically, the keywords include position name, average salary level, industry, company type, company size, city, establishment date, registered capital, work duration, work experience requirement, education requirement, ability requirement, experience requirement, software requirement, professional requirement, etc.

[0046] Step S103, substituting the information of the job seeker into the pre-constructed Lasso regression model for salary prediction to obtain the expected salary value of the job seeker.

[0047] The average salary in the keyword is taken as the dependent variable, the best value of the adjustment parameter λ is selected by cross-validation, and a plurality of Lasso regression models for different types of positions are constructed; the regression coefficients of the Lasso regression model are taken as the level weight involved in the relevant dimension of the career assessment scheme of the job seeker. The Lasso regression model is trained and tested by the data set corresponding to the recruitment information.

[0048] Step S104, determining the career assessment scheme of the job seeker according to the keyword and the regression coefficient of the Lasso regression model; and performing career assessment on the job seeker through the career assessment scheme.

[0049] The keyword representing the average salary of the job seeker is selected from the keyword, the keyword includes but is not limited to education level, working years, professional accomplishment, work experience and software mastering situation; the regression coefficient is taken as the weight of the education level and the working years; the average salary corresponding to the keyword is taken as the weight of the professional accomplishment, the work experience and the software mastering situation; and the weight is standardized; and the career assessment scheme of the job seeker is determined according to the keyword, the regression coefficient and the weight.

[0050] After obtaining the Lasso regression model, a professional evaluation scheme is formulated according to the results of text mining and the regression coefficients. For example, the candidate can be scored from the aspects of "education level", "work experience", "professional quality", "work experience" and "software mastery". The weights of each level involved in the two dimensions of "education level" and "work experience" mainly come from the regression coefficients; the weights of each level involved in the other three dimensions come from the average salary under each level in the text analysis. Further, the weights of each level in each dimension are standardized, the maximum value is set to 100 points, the minimum value is set to 0 points, and the intermediate value is subtracted from the minimum value and then divided by the range (maximum value-minimum value) to obtain the specific score. Specifically, assume that the education level has three levels: undergraduate, master and doctor. Then, the regression coefficient of undergraduate is 100, the regression coefficient of master is 200, and the regression coefficient of doctor is 300. Through the maximum and minimum processing, the score of undergraduate is 100*(100-100) / (300-100), the score of master is 100*(200-100) / (300-100), and the score of doctor is 100*(300-100) / (300-100). Work experience, through text mining analysis, there are three levels: management experience, project experience and operation experience, and the corresponding average salaries are 1, 2 and 3 respectively. Therefore, the score of management experience is 100*(1-1) / (3-1), the score of project experience is 100*(2-1) / (3-1), and the score of operation experience is 100*(3-1) / (3-1). If a candidate has both project experience and operation experience, its score should be the average of the operation score and the project experience score.

[0051] According to the evaluation scheme, each aspect contained in the information of the candidate is scored to obtain the professional evaluation score of the candidate. Based on the standardized candidate information, relevant information such as education background, work experience, whether having management experience, whether having communication ability, whether mastering a certain technology, etc. is extracted as the input of the professional evaluation function unit by using natural language processing technology.

[0052] The candidate's corresponding characteristics are identified from the candidate information reading function unit, and the candidate is scored and profiled according to the 5-dimensional scoring mechanism given by the modeling analysis function unit, and a net analysis diagram is output to help the candidate self-positioning. In addition, the Lasso regression model is used to obtain the recommended salary of the candidate. The system substitutes the candidate information and the recruitment post situation into the regression model to obtain the predicted value of the salary, which can provide a reference for personal salary expectation and company salary customization.

[0053] Figure 2 is a system flowchart for testing a candidate using the professional evaluation method provided by the present application, which specifically comprises:

[0054] 1. Data collection, mainly including data collection and data storage function units.

[0055] Data collection unit. This unit can realize the regular collection of relevant recruitment information from major Internet recruitment platforms according to keywords, and supports cloud collection and local collection.

[0056] Data storage unit. This unit can realize the storage of the data collected by the system. In order to ensure the standardization of the collected data, the system supports the setting of data field mapping relationship, that is, through the source data field, source data type, target data field and target data type to complete the writing and storage of data, and supports setting a certain field as a unique identifier.

[0057] 2. Data preprocessing, which is the basis for establishing a career assessment and user portrait system, including a data preprocessing unit.

[0058] Data preprocessing unit. After the system completes data collection and data storage, the system will automatically complete data cleaning and data preprocessing according to the collected structured data to obtain a recruitment information dataset. Each observation in the dataset represents the recruitment information of a position. Among them, the average salary level is the main research target, so it is regarded as the dependent variable, and there are 13 independent variables, including recruitment company information (company type, company size, etc.) and job requirements (work experience requirement, education requirement, etc.). The specific variables in the data are: position name, average salary level, industry, company type, company size, city, establishment date, registered capital, work duration, work experience requirement, education requirement, ability requirement, experience requirement, software requirement, and professional requirement.

[0059] 3. Mining modeling, including text mining and modeling analysis function units.

[0060] Text mining function unit. For the text type variables in the data set, the method of text analysis is used to analyze the influence of job requirements and corresponding average salary. All job requirements are segmented, and the top 50 words with the highest frequency are extracted to draw a word cloud diagram, and the average salary comparison analysis result of whether this post requires this ability is given.

[0061] The modeling analysis function unit takes the "log average salary" as the dependent variable, and establishes multiple Lasso regression models for different types of positions. When constructing the model, cross-validation is used to select the best value of the adjustment parameter λ. After obtaining the final model, the system will develop a career assessment program based on the results of text mining and regression coefficients. For example, the candidate can be scored from five aspects: "education level", "work experience", "professional competence", "work experience", and "software proficiency". The weights of each level in the "education level" and "work experience" dimensions are mainly derived from the regression coefficients; the weights of each level in the other three dimensions are derived from the average salary under each level in the text analysis. Further, the weights of each level in each dimension are standardized, with the maximum value set to 100 points and the minimum value set to 0 points. The intermediate value is subtracted from the minimum value and then divided by the range (maximum value - minimum value) to obtain the specific score.

[0062] 4. Career assessment, including candidate information reading and career assessment function units.

[0063] The candidate information reading function unit uses natural language processing technology to extract relevant information such as education background, work experience, management experience, communication skills, and technical skills based on standardized candidate information as input for the career assessment function unit.

[0064] The career assessment function unit identifies the corresponding characteristics of the candidate from the candidate information reading function unit and scores and profiles the candidate according to the 5-dimensional scoring mechanism provided by the modeling analysis function unit, outputs a network analysis diagram, and helps the candidate to self-position. In addition, the system can obtain the predicted value of the salary by substituting the candidate information and the recruitment position information into the regression model, which can provide a reference for personal salary expectations and company salary customization.

[0065] Based on the same inventive concept, the present application provides a career assessment system 300 based on recruitment big data, as shown in Figure 3 , comprising:

[0066] The data acquisition module 310 is used to acquire relevant recruitment information; the recruitment information is preprocessed to obtain a data set corresponding to the recruitment information;

[0067] The mining module 320 is used to perform text mining on the data set to obtain keywords that affect the average salary;

[0068] The salary expectation value acquisition module 330 is used to substitute the information of the candidate into the pre-constructed Lasso regression model for salary prediction to obtain the salary expectation value of the candidate;

[0069] The professional evaluation module 340 is configured to determine a professional evaluation scheme of the candidate according to the keywords and the regression coefficients of the Lasso regression model, and perform professional evaluation on the candidate through the professional evaluation scheme.

[0070] Preferably, the professional evaluation module comprises:

[0071] The keyword selection subunit is configured to select keywords representing average salary of the candidate from the keywords, wherein the keywords include but are not limited to education level, working years, professional accomplishment, work experience and software mastering situation.

[0072] The standardization processing subunit is configured to take the regression coefficients as weights of the education level and the working years, take the average salary corresponding to the keywords as weights of the professional accomplishment, the work experience and the software mastering situation, and perform standardization processing on the weights.

[0073] The evaluation scheme determination subunit is configured to determine a professional evaluation scheme of the candidate according to the keywords, the regression coefficients and the weights.

[0074] Preferably, the professional evaluation module comprises:

[0075] The evaluation score acquisition subunit is configured to score each aspect of the information of the candidate according to the evaluation scheme, and acquire a professional evaluation score of the candidate.

[0076] Preferably, the application further comprises:

[0077] The recommended salary acquisition module is configured to acquire a recommended salary of the candidate through the Lasso regression model.

[0078] The professional evaluation method and system based on recruitment big data provided by the application automatically realize collection of recruitment information of various recruitment positions by using internet cloud collection technology, and establish a corresponding professional evaluation model according to the real-time collected data, and compared with the traditional professional evaluation, the application has the advantages of being more professional and having vertical depth.

[0079] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application but not to limit it, and although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered in the scope of the claims of the present application.

Claims

1. A career assessment method based on recruitment big data, characterized in that, include: Collect relevant recruitment information; The recruitment information is preprocessed to obtain the dataset corresponding to the recruitment information; Text mining is performed on the dataset to obtain keywords that affect the average salary and the average salary corresponding to the keywords; The applicant's information is fed into a pre-built Lasso regression model for salary prediction to obtain the applicant's expected salary value. Based on the keywords and the regression coefficients of the Lasso regression model, determine the job applicant's career assessment plan; The job applicants were assessed using the aforementioned career assessment scheme. Based on the keywords and the regression coefficients of the Lasso regression model, determine the job applicant's career assessment plan, specifically including: Select keywords representing the average salary of applicants from the keywords, including education level, years of work experience, professional qualities, work experience and software proficiency; The regression coefficients are used as weights for education level and years of work experience; the average salary corresponding to the keywords is used as a weight for professional competence, work experience, and software proficiency; and the weights are standardized. Based on the keywords, regression coefficients, and weights, a career assessment plan for job applicants is determined.

2. The method according to claim 1, characterized in that, The recruitment information is preprocessed to obtain the dataset corresponding to the recruitment information, including: Regularly retrieve relevant job postings from recruitment platforms; The recruitment information is cleaned and preprocessed to obtain the dataset corresponding to the recruitment information.

3. The method according to claim 1, characterized in that, Text mining was performed on the dataset to extract keywords that influence average salaries, including: For the text-type variables in the dataset, text analysis methods are used to analyze the impact of job requirements and corresponding average salaries, and to identify keywords that affect the average salary.

4. The method according to claim 1, characterized in that, The pre-built Lasso regression model for salary prediction is trained and tested using the dataset corresponding to the recruitment information.

5. The method according to claim 1, characterized in that, The job applicants are assessed using the aforementioned career assessment scheme, including: According to the assessment scheme, scores are given to various aspects of the applicant's information to obtain the applicant's career assessment score.

6. The method according to claim 1 or 5, characterized in that, Also includes: The suggested salary for job applicants is obtained using the Lasso regression model.

7. A career assessment system based on recruitment big data, characterized in that, include: The data acquisition module is used to collect relevant recruitment information; The recruitment information is preprocessed to obtain the dataset corresponding to the recruitment information; The text mining module is used to perform text mining on the dataset to obtain keywords that have an impact on average salary; The salary expectation value acquisition module is used to input the applicant's information into a pre-built Lasso regression model for salary prediction to obtain the applicant's salary expectation value. The career assessment module is used to determine the career assessment plan for job applicants based on the keywords and the regression coefficients of the Lasso regression model. The job applicants were assessed using the aforementioned career assessment scheme. The career assessment module includes: The keyword selection subunit is used to select keywords representing the average salary of applicants from the keywords. The keywords include education level, years of work experience, professional qualities, work experience, and software proficiency. The standardization processing subunit is used to use the regression coefficients as weights for education level and years of work experience; to use the average salary corresponding to the keywords as weights for professional qualities, work experience, and software proficiency; and to standardize the weights. The assessment scheme determination sub-unit is used to determine the applicant's career assessment scheme based on the keywords, regression coefficients, and weights.

8. The system according to claim 7, characterized in that, The career assessment module includes: The assessment score acquisition subunit is used to score various aspects of the applicant's information according to the assessment scheme and obtain the applicant's career assessment score.

9. The system according to claim 7 or 8, characterized in that, Also includes: The suggested salary acquisition module is used to obtain the suggested salary for job applicants through the Lasso regression model.

Images