Matching degree initial screening method, system and medium based on big data analysis

Abstract

The embodiment of the present invention provides a matching degree initial screening method based on big data analysis, which includes: person-job matching degree parameter calculation: using the measured resume data and standardized assessment indicators as samples, fitting and calculating the person-job matching degree weight coefficient; determining the threshold for passing the initial screening according to the actual number of job requirements and the number of candidates of the enterprise; candidate information analysis: pre-processing the resume to be analyzed, converting each resume information into a numerical indication form, and calculating the person-job matching degree score of the person in combination with the person-job matching degree weight coefficient; feedback of the analysis result; the person-job matching screening method reduces the ineffective actions of both parties, allowing the enterprise to find suitable talents, and avoiding the waste of time of candidates with low matching degree in irrelevant positions. The research and development of the resume screening system realizes the transformation of results from algorithm to platform, truly reducing the burden on recruiters, improving recruitment efficiency, and saving recruitment costs.

Description

Technical Field

[0001] The present invention relates to the field of big data analysis, and in particular to a matching degree preliminary screening method, system and medium based on big data analysis. Background Art

[0002] With the continuous development of Internet technology, the talent recruitment method of receiving resumes through emails has been increasingly favored by various recruiters. At present, recruiters generally publish recruitment emails for receiving resumes in the recruitment notices they publish, so that applicants can send their resumes directly to the recruitment email according to the recruitment email published by the recruiting unit. Then, recruiters manually screen the resumes received in the recruitment email and forward the selected excellent resumes to the actual recruiting department.

[0003] However, the process of recruiters manually screening resumes is very time-consuming and brain-consuming, so some programs for automatically screening resumes have appeared on the market. The existing patent CN201811469975.3 "A method and device for processing recruitment information" discloses a program for screening resumes by extracting keywords from resumes and calculating the similarity with recruitment job requirements. However, this program can only perform quantitative screening, that is, it can only determine whether the job seeker has the ability required for the position, but cannot determine the degree of mastery of the ability required for the position. Therefore, the accuracy of resume screening is limited.

[0004] Therefore, proposing a matching degree initial screening method, system and medium based on big data analysis has become a new technical requirement. Summary of the invention

[0005] To solve the above problems, the present application provides a matching degree initial screening method, system and medium based on big data analysis to quickly and accurately screen out resumes that match the position.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] In the first aspect, the present application provides a matching degree initial screening method based on big data analysis, the method comprising: step 1, establishing a resume sample database: obtaining resumes received in the past few years in the specified enterprise, including resumes that have not passed the initial screening and resumes that have passed the initial screening, and numbering the resumes in sequence as 1, 2, ...i, ...n in a set order to establish a resume sample database; step 2, resume sample data preprocessing: using statistical measurement concepts to characterize the characteristics and attributes of resume information, and converting resume information into a numerical representation; step 3, employee benefit database establishment: obtaining benefit data of employees who have joined the specified enterprise in the past few years, and constructing an employee benefit database; step 4, employee benefit data preprocessing: analyzing the benefit data, determining assessment indicators based on the actual situation of the enterprise, sorting the assessment indicators according to the degree of concern of the enterprise, dividing qualitative assessment indicators, etc. Step 5: Calculation of person-job matching parameters: Using the measured resume data and standardized assessment indicators as samples, fit and calculate the person-job matching weight coefficient; determine the threshold for passing the initial screening based on the actual number of job requirements and candidates of the enterprise; Step 6: Candidate information analysis: pre-process the resumes to be analyzed, convert each resume information into a numerical form, and calculate the person-job matching score of the person in combination with the person-job matching weight coefficient; Step 7: Feedback of analysis results: Determine the candidate's resume based on the initial screening threshold and enter the next stage after passing the initial screening. Visually display the scores of all resumes and feedback them to the human resources department. At the same time, add the processed resumes to the resume sample database to help update the person-job matching weight coefficient in real time.

[0008] Optionally, the resume sample data is preprocessed, and the specific processing process includes the following steps: step 2-1, classifying the resume information, and classifying the resume information into attributes with clear values ​​in a range, attributes without clear values ​​but with a certain internal ranking order, and attributes without internal order classification. The above three attributes are respectively recorded as S, O, and N, where S includes age, expected income, and college English scores, O includes attributes of graduation schools, internship experience, and student activity experience, and N includes gender; step 2-2, classifying the three attributes Different measurement methods are set for gender. A fixed ratio measurement is used for S, and the value of S is directly used. An ordinal measurement is used for O, and different values ​​are assigned to different grades. A categorical measurement is used for N, and different values ​​are assigned to different categories. The graduate school attribute in O is divided into 985 colleges, 211 colleges, general key colleges, general undergraduate colleges, other junior colleges and private colleges, and the values ​​are assigned T1, T2, T3, T4, and T5 respectively, satisfying T1>T2>T3>T4>T5. The gender attribute in N is divided into male and female, and the values ​​are assigned to 1 and 2 respectively.

[0009] Optionally, the assessment indicators are determined according to the actual situation of the enterprise, and the assessment indicators are ranked according to the degree of concern of the enterprise.

[0010] Optionally, the preprocessing of employee benefit data includes the following specific steps: Step 4-1, classifying the indicators into extremely large indicators, extremely small indicators, medium indicators and range indicators according to the standard of the interval in which the company expects the indicator values ​​to fall, wherein the workload completed by the employee is an extremely large indicator, the salary cost of the employee is an extremely small indicator, the age of the employee is a medium indicator, and the number of times the employee aligns and pulls through should be a range indicator; Step 4-2, unifying the four indicators into extremely large indicators, directly taking the negative of the extremely small indicators, and processing the medium indicators according to the following formula:

[0011]

[0012] In the above formula, v i is the value of the i-th evaluation index, h and l are the maximum and minimum values ​​allowed for the evaluation index, α is the evaluation coefficient factor, and v' i To obtain consistent indicator values, the range-type indicators are processed according to the following formula:

[0013]

[0014] In the above formula, v i is the value of the i-th evaluation index, b il and b ir are the left and right boundaries of the optimal range of the evaluation index, α and β are the evaluation coefficient factors, and c is the boundary coefficient factor. The specific calculation process is as follows:

[0015] c=max(b il ―l,h―b ir )

[0016] In the above formula, h and l are the maximum and minimum values ​​allowed for the evaluation index, respectively;

[0017] Step 4-3: All indicators are dimensionally standardized to eliminate the impact of the difference in numerical units and magnitudes on the evaluation process. In order to reduce the impact of outliers, the following formula is used:

[0018]

[0019] In the above formula, v i is the value of the i-th evaluation index, v' i is the index value after unification, v m is the median of the evaluation index value, q d is the interquartile range, which is calculated as follows:

[0020] q d =q3―q1

[0021] In the above formula, q3 is the upper quartile and q1 is the lower quartile.

[0022] Optionally, represent all the preprocessed resume data as x1, x2...x i …x n , where n is the number of resume samples, x i =[v 1, v2...v j, ...v p ],v j Represents different resume attributes, and all preprocessed assessment index data are expressed as y = [y1, y2...y i …y m ], where m is the number of assessment indicators, and the weight parameter matrix W to be estimated is set.

[0023] Optionally, the person-job matching parameter calculation process is as follows:

[0024] Step 5-1: Determine the cost function Where c is the kernel function factor;

[0025] Step 5-2: Use the resume sample data and assessment index data, and use the gradient descent method to fit the weight parameter matrix W that minimizes the cost function.

[0026] Optionally, the calculation formula of the job matching initial screening threshold is as follows:

[0027]

[0028] In the above formula, n1 is the number of job requirements, n2 is the number of candidates, ρ is the conversion function, and κ is the compensation factor for the impact of policy factors, etc.

[0029] Optionally, the analysis result feedback, the specific feedback process is: step 7-1, based on the candidate's value coefficient γ and threshold η, if γ>η, the candidate passes the initial screening, the system pushes the candidate's resume to the human resources department, and displays the candidate's basic information and score on the visual terminal; step 7-2, the candidate's resume is added to the resume database. If the subsequent candidate passes various interview processes and joins the company, the candidate's post-work benefits and assessment results are added to the employee benefit database, and the weight coefficient is recalculated when the data accumulates to a certain level.

[0030] On the second aspect, the present application provides a matching degree initial screening system based on big data analysis, which includes: a resume sample database establishment module, which is used to obtain resumes received in the past few years in the specified enterprise, including resumes that have not passed the initial screening and resumes that have passed the initial screening, and complete the establishment of the sample database after numbering; a resume sample data preprocessing module, which is used to characterize the characteristics and attributes of resume information and convert resume information into a numerical representation; an employee benefit database module, which is used to obtain the benefit data of employees who have joined the specified enterprise in the past few years and complete the establishment of the employee benefit database; an employee benefit data preprocessing module, which is used to determine the assessment indicators that the company is concerned about, quantify the qualitative indicators, and Different reference data are set for the qualitative and quantitative indicators for standardization processing; the person-job matching parameter calculation module is used to fit and calculate the person-job matching weight coefficient and calculate the initial screening threshold; the candidate information analysis module is used to pre-process the resumes to be analyzed, convert various resume information into numerical representation, and then calculate the person-job matching score of the person in combination with the person-job matching weight coefficient; the analysis result feedback terminal is used to complete the preliminary screening of resumes, visualize the scores of all resumes and feedback them to the human resources department, and add the resumes processed this time to the database; the information database is used to store resume sample data, employee benefit sample data and person-job matching weight coefficient.

[0031] In a third aspect, the present application provides a matching degree preliminary screening medium based on big data analysis, wherein the job matching degree preliminary screening storage medium is burned with a computer program, and the computer program implements the above method when running in the memory of the server.

[0032] The present application provides a matching degree initial screening method based on big data analysis, which accurately screens candidate resumes for positions by quantifying the assessment indicators in the resumes, and then screens job seekers who meet the job requirements; by performing weighted calculations on the resume data in the resume database, the accuracy of resume screening is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the implementation of the present invention or the scheme in the prior art, the following briefly introduces the drawings required for the implementation or the prior art description. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0034] The structures, proportions, sizes, etc. illustrated in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantial technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0035] Attached photos:

[0036] Figure 1 A flowchart of a matching degree preliminary screening method based on big data analysis provided in Example 1 of the present invention;

[0037] Figure 2 A schematic diagram of the structure of a matching degree preliminary screening system based on big data analysis provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0038] 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 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 should fall within the scope of protection of the present invention.

[0039] For the convenience of explanation, the technical solution of the embodiment of the present invention is described in detail with reference to the accompanying drawings. It should be understood that this is only an example for illustrating the technical solution of the embodiment of the present invention and does not constitute any limitation to the embodiment of the present invention.

[0040] Figure 1 A matching degree preliminary screening method based on big data analysis is provided in an embodiment of the present invention, such as Figure 1 The flowchart of the matching degree initial screening method based on big data analysis is shown, and the method includes the following steps:

[0041] S101, establishing a resume sample database: obtaining resumes received by a designated enterprise in the past few years, including resumes that have not passed the initial screening and resumes that have passed the initial screening, and numbering the resumes in sequence as 1, 2, ...i, ...n in a set order to establish a resume sample database.

[0042] S102, resume sample data preprocessing: using statistical measurement concepts to characterize the characteristics and attributes of resume information, and converting resume information into a numerical representation.

[0043] Optionally, the preprocessing of the resume sample data includes the following steps:

[0044] Step 2-1, classify the resume information, and classify the resume information into attributes with clear values ​​in a range, attributes without clear values ​​but with a certain internal ranking order, and attributes without internal order classification. The above three attributes are denoted as S, O, and N respectively, where S includes age, expected income, and college English scores, O includes graduate school attributes, internship experience, and student activity experience, and N includes gender.

[0045] Step 2-2, set different measurement methods for the three attributes. Use ratio measurement for S and directly use the value of S. Use ordinal measurement for O and assign different values ​​to different grades. Use categorical measurement for N and assign different values ​​to different categories. Divide the graduate school attribute in O into 985 colleges, 211 colleges, general key colleges, general undergraduate colleges, other junior colleges and private colleges, and assign them T1, T2, T3, T4, T5 respectively, satisfying T1>T2>T3>T4>T5. Divide the gender attribute in N into male and female, and assign them 1 and 2 respectively.

[0046] S103, establishment of employee benefit database: obtaining benefit data of employees who have joined the designated enterprise in the past few years and constructing an employee benefit database.

[0047] Specifically, benefit data may include: performance data, task volume data, task quality data, etc.

[0048] S104, preprocessing of employee benefit data: analyze the benefit data, determine the assessment indicators according to the actual situation of the enterprise, sort the assessment indicators according to the degree of concern of the enterprise, divide the qualitative assessment indicators into grades, quantify them according to the grades, and set different reference data for the quantified qualitative and quantitative indicators for standardization.

[0049] Determine the assessment indicators based on the actual situation of the enterprise and rank the assessment indicators according to the degree of concern of the enterprise.

[0050] The specific steps of preprocessing employee benefit data are as follows:

[0051] Step 4-1. Classify the indicators into extremely large indicators, extremely small indicators, medium indicators and range indicators according to the range in which the company expects the indicator values ​​to fall. Among them, the workload completed by employees is an extremely large indicator, the salary cost of employees is an extremely small indicator, the age of employees is a medium indicator, and the number of times employees align should be a range indicator.

[0052] Step 4-2: Unify the four indicators into very large indicators, directly negate the very small indicators, and process the medium indicators according to the following formula:

[0053]

[0054] In the above formula, v i is the value of the i-th evaluation index, h and l are the maximum and minimum values ​​allowed for the evaluation index, α is the evaluation coefficient factor, and v' i To obtain consistent indicator values, the range-type indicators are processed according to the following formula:

[0055]

[0056] In the above formula, v i is the value of the i-th evaluation index, b il and b ir are the left and right boundaries of the optimal range of the evaluation index, α and β are the evaluation coefficient factors, and c is the boundary coefficient factor. The specific calculation process is as follows:

[0057] c=max(b il ―l,h―b ir )

[0058] In the above formula, h and l are the maximum and minimum values ​​allowed for the evaluation index, respectively.

[0059] Step 4-3: All indicators are dimensionally standardized to eliminate the impact of the difference in numerical units and magnitudes on the evaluation process. In order to reduce the impact of outliers, the following formula is used:

[0060]

[0061] In the above formula, v i is the value of the i-th evaluation index, v' i is the index value after unification, v m is the median of the evaluation index value, q d is the interquartile range, which is calculated as follows:

[0062] q d =q3―q1

[0063] In the above formula, q3 is the upper quartile and q1 is the lower quartile.

[0064] S105, Calculation of person-job matching parameters: Using the measured resume data and standardized assessment indicators as samples, fit and calculate the person-job matching weight coefficient; determine the threshold for passing the initial screening based on the actual number of job requirements and candidates in the enterprise.

[0065] Represent all the preprocessed resume data as x1, x2...x i …x n , where n is the number of resume samples, x i=[v 1, v2...v j, ...v p ],v j Represents different resume attributes, and all preprocessed assessment index data are expressed as y = [y1, y2...y i …y m ], where m is the number of assessment indicators, and the weight parameter matrix W to be estimated is set.

[0066] Optionally, the person-job matching parameter calculation process is as follows:

[0067] Step 5-1: Determine the cost function Where c is the kernel function factor.

[0068] Step 5-2: Use the resume sample data and assessment index data, and use the gradient descent method to fit the weight parameter matrix W that minimizes the cost function.

[0069] Optionally, the calculation formula of the job matching initial screening threshold is as follows:

[0070]

[0071] In the above formula, n1 is the number of job requirements, n2 is the number of candidates, ρ is the conversion function, and κ is the compensation factor for the impact of policy factors, etc.

[0072] S106, candidate information analysis: pre-process the resume to be analyzed, convert each resume information into a numerical value, and then calculate the person-job matching score of the candidate in combination with the person-job matching weight coefficient.

[0073] S107, analysis result feedback: determine the candidate's resume according to the initial screening threshold and enter the next stage after passing the initial screening, visualize the scores of all resumes and feedback to the human resources department, and add the processed resumes to the resume sample database to help update the person-job matching weight coefficient in real time.

[0074] The analysis result feedback, the specific feedback process is:

[0075] Step 7-1: Based on the candidate's value coefficient γ and threshold η, if γ>η, the candidate passes the initial screening, and the system pushes the candidate's resume to the human resources department, and displays the candidate's basic information and score on the visual terminal.

[0076] Step 7-2: Add the candidate's resume to the resume database. If the subsequent candidate passes the interview process and joins the company, the candidate's post-work performance and assessment results will be added to the employee performance database. When the data accumulates to a certain level, the weight coefficient will be recalculated.

[0077] Corresponding to the above-mentioned embodiment 1, embodiment 2 of the present invention further provides a matching degree preliminary screening system based on big data analysis, which includes: a resume sample database establishment module 201, which is used to obtain resumes received in the past few years in the specified enterprise, including resumes that have not passed the preliminary screening and resumes that have passed the preliminary screening, and complete the establishment of the sample database after numbering; a resume sample data preprocessing module 202, which is used to characterize the characteristics and attributes of resume information and convert resume information into a numerical representation form; an employee benefit database module 203, which is used to obtain the benefit data of employees who have joined the specified enterprise in the past few years and complete the establishment of the employee benefit database; an employee benefit data preprocessing module 204, which is used to determine the assessment indicators that the company is concerned about, and to perform quantitative analysis on qualitative indicators. quantification, setting different reference data for the quantified qualitative indicators and quantitative indicators for standardization; person-job matching parameter calculation module 205, used to fit and calculate the person-job matching weight coefficient and calculate the initial screening threshold; candidate information analysis module 206, used to pre-process the resume to be analyzed, convert each resume information into a numerical representation, and calculate the person-job matching score of the person in combination with the person-job matching weight coefficient; analysis result feedback terminal 207, used to complete the preliminary screening of resumes, visualize the scores of all resumes and feedback to the human resources department, and add the resumes processed this time to the database; information library 208, used to store resume sample data, employee benefit sample data and person-job matching weight coefficient.

[0078] The functions performed by the various components in the automatic question-generating device based on knowledge point expressions provided by the embodiment of the present invention have been introduced in detail in the above-mentioned embodiment 1, so they will not be elaborated here.

[0079] Corresponding to the above-mentioned embodiment 1, embodiment 3 of the present invention also provides a matching degree preliminary screening medium based on big data analysis, and the position matching degree preliminary screening storage medium is burned with a computer program. When the computer program is running in the memory of the server, it implements the above-mentioned matching degree preliminary screening method based on big data analysis.

[0080] The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor for execution, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The processor reads the information in the storage medium and completes the steps of the above method in combination with its hardware.

[0081] The storage medium may be a memory, which may be, for example, a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memory.

[0082] The storage media described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0083] The present application provides a matching degree initial screening method based on big data analysis, which accurately screens candidate resumes for positions by quantifying the assessment indicators in the resumes, and then screens job seekers who meet the job requirements; by performing weighted calculations on the resume data in the resume database, the accuracy of resume screening is higher.

[0084] Those skilled in the art will appreciate that in one or more of the above examples, the functions described in the present invention can be implemented using a combination of hardware and software. When software is used, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any media that facilitates the transmission of computer programs from one place to another. Storage media can be any available media that can be accessed by general or special-purpose computers.

[0085] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A resume initial screening method for matching positions based on big data analysis, characterized in that: The method includes: Step 1, Establishing a resume sample database: Obtain the resumes received by the specified company in the past few years, including resumes that have not passed the initial screening and resumes that have passed the initial screening, and number the resumes in sequence as 1, 2, ...i, ...n in the set order to establish a resume sample database; Step 2: Resume sample data preprocessing: Use statistical measurement concepts to characterize the characteristics and attributes of resume information and convert resume information into numerical representations; Step 3: Establish employee benefit database: Obtain benefit data of employees who have joined the designated enterprise in the past few years and build an employee benefit database; Step 4: Preprocessing of employee benefit data: Analyze benefit data, determine assessment indicators based on the actual situation of the enterprise, sort assessment indicators according to the degree of concern of the enterprise, divide qualitative assessment indicators into grades, quantify them according to the grades, and set different reference data for the quantified qualitative and quantitative indicators for standardization. The specific steps of preprocessing employee benefit data are as follows: Step 4-1. Categorize the indicators into extremely large indicators, extremely small indicators, medium indicators and range indicators according to the range in which the company expects the indicator values ​​to fall. The workload completed by employees is an extremely large indicator, the salary cost of employees is an extremely small indicator, the age of employees is a medium indicator, and the number of times employees have completed alignment should be a range indicator. Step 4-2: Unify the four indicators into very large indicators, directly negate the very small indicators, and process the medium indicators according to the following formula: In the above formula, v i is the value of the i-th evaluation index, h and l are the maximum and minimum values ​​allowed for the evaluation index, α is the evaluation coefficient factor, and v′ i To obtain consistent indicator values, the range-type indicators are processed according to the following formula: In the above formula, v i is the value of the i-th evaluation index, bi l and bi r are the left and right boundaries of the optimal range of the evaluation index, α and β are the evaluation coefficient factors, and c is the boundary coefficient factor. The specific calculation process is as follows: c=max(b il -l,h-b ir ) In the above formula, h and l are the maximum and minimum values ​​allowed for the evaluation index, respectively; Step 4-3: All indicators are dimensionally standardized to eliminate the impact of the difference in numerical units and magnitudes on the evaluation process. In order to reduce the impact of outliers, the following formula is used: In the above formula, v i is the value of the i-th evaluation index, v′ i is the index value after unification, v m is the median of the evaluation index value, q d is the interquartile range, which is calculated as follows: q d =q3-q1 In the above formula, q3 is the upper quartile and q1 is the lower quartile; Step 5: Calculation of person-job matching parameters: Using the measured resume data and standardized assessment indicators as samples, fit and calculate the person-job matching weight coefficient; determine the threshold for passing the initial screening based on the actual number of job requirements and candidates of the enterprise; Step 6, candidate information analysis: pre-process the resume to be analyzed, convert each resume information into a numerical value, and calculate the candidate's job matching score based on the job matching weight coefficient; Step 7, analysis result feedback: Determine the candidate's resume based on the initial screening threshold and enter the next stage after passing the initial screening. Visualize the scores of all resumes and feedback to the human resources department. At the same time, add the processed resumes to the resume sample database to help update the person-job matching weight coefficient in real time.

2. The resume initial screening method for matching positions based on big data analysis as claimed in claim 1, characterized in that: The specific processing process of the resume sample data preprocessing includes the following steps: Step 2-1, classify the resume information, and classify the resume information into attributes with clear values ​​in a range, attributes without clear values ​​but with a certain internal ranking order, and attributes without internal order classification. The above three attributes are recorded as S, O, and N respectively, where S includes age, expected income, and college English scores, O includes attributes of graduation schools, internship experience, and student activity experience, and N includes gender; Step 2-2, set different measurement methods for the three attributes. Use ratio measurement for S and directly use the value of S. Use ordinal measurement for O and assign different values ​​to different grades. Use categorical measurement for N and assign different values ​​to different categories. Divide the graduate school attribute in O into 985 colleges, 211 colleges, general key colleges, general undergraduate colleges, other junior colleges and private colleges, and assign them T1, T2, T3, T4, T5 respectively, satisfying T1>T2>T3>T4>T5. Divide the gender attribute in N into male and female, and assign them 1 and 2 respectively.

3. The resume initial screening method for matching positions based on big data analysis as claimed in claim 1, characterized in that: Determine the assessment indicators based on the actual situation of the enterprise and rank the assessment indicators according to the degree of concern of the enterprise.

4. The resume initial screening method for matching positions based on big data analysis as claimed in claim 1, characterized in that: Represent all the preprocessed resume data as x1, x2...x i …x n , where n is the number of resume samples, x i =[v 1, v2...v j, ...v p ],v j Represents different resume attributes, and all preprocessed assessment index data are expressed as y = [y1, y2...y i …y m ], where m is the number of assessment indicators, and the weight parameter matrix W to be estimated is set.

5. The resume initial screening method for matching positions based on big data analysis as claimed in claim 1, characterized in that: The calculation process of the person-job matching parameter is as follows: Step 5-1: Determine the cost function Where c is the kernel function factor; Step 5-2: Use the resume sample data and assessment index data, and use the gradient descent method to fit the weight parameter matrix W that minimizes the cost function.

6. The resume initial screening method for matching positions based on big data analysis as claimed in claim 1, characterized in that: The calculation formula for the initial screening threshold of job matching degree is as follows: In the above formula, n1 is the number of job requirements, n2 is the number of candidates, ρ is the conversion function, and κ is the compensation factor for the impact of policy factors, etc.

7. The resume initial screening method for matching positions based on big data analysis as claimed in claim 1, characterized in that: The analysis result feedback, the specific feedback process is: Step 7-1: Based on the candidate's value coefficient γ and threshold η, if γ>η, the candidate passes the initial screening, and the system pushes the candidate's resume to the human resources department, and displays the candidate's basic information and score on the visual terminal; Step 7-2: Add the candidate's resume to the resume database. If the subsequent candidate passes the interview process and joins the company, the candidate's post-work performance and assessment results will be added to the employee performance database. When the data accumulates to a certain level, the weight coefficient will be recalculated.

8. A resume screening system for matching positions based on big data analysis, characterized in that: The system includes: The resume sample database establishment module is used to obtain the resumes received by the specified enterprise in the past few years, including resumes that have not passed the initial screening and resumes that have passed the initial screening, and complete the establishment of the sample database after numbering; The resume sample data preprocessing module is used to characterize the characteristics and attributes of resume information and convert resume information into a numerical representation form; The employee benefit database module is used to obtain the benefit data of employees who have joined the company in the past few years and complete the establishment of the employee benefit database; The employee benefit data preprocessing module is used to determine the assessment indicators that the company is concerned about, quantify the qualitative indicators, set different reference data for the quantified qualitative indicators and quantitative indicators for standardization, and preprocess the employee benefit data. The specific steps are as follows: Step 4-1. Categorize the indicators into extremely large indicators, extremely small indicators, medium indicators and range indicators according to the range in which the company expects the indicator values ​​to fall. The workload completed by employees is an extremely large indicator, the salary cost of employees is an extremely small indicator, the age of employees is a medium indicator, and the number of times employees have completed alignment should be a range indicator. Step 4-2: Unify the four indicators into very large indicators, directly negate the very small indicators, and process the medium indicators according to the following formula: In the above formula, v i is the value of the i-th evaluation index, h and l are the maximum and minimum values ​​allowed for the evaluation index, α is the evaluation coefficient factor, and v′ i To obtain consistent indicator values, the range-type indicators are processed according to the following formula: In the above formula, v i is the value of the i-th evaluation index, b il and b ir are the left and right boundaries of the optimal range of the evaluation index, α and β are the evaluation coefficient factors, and c is the boundary coefficient factor. The specific calculation process is as follows: c=max(b il -l,h-b ir ) In the above formula, h and l are the maximum and minimum values ​​allowed for the evaluation index, respectively; Step 4-3: All indicators are dimensionally standardized to eliminate the impact of the difference in numerical units and magnitudes on the evaluation process. In order to reduce the impact of outliers, the following formula is used: In the above formula, v i is the value of the i-th evaluation index, v′ i is the index value after unification, v m is the median of the evaluation index value, q d is the interquartile range, which is calculated as follows: q d =q3-q1 In the above formula, q3 is the upper quartile and q1 is the lower quartile; Person-job matching parameter calculation module, used to fit and calculate the person-job matching weight coefficient and calculate the initial screening threshold; The candidate information analysis module is used to pre-process the resume to be analyzed, convert the resume information into numerical values, and then calculate the candidate's job matching score based on the job matching weight coefficient; The analysis result feedback terminal is used to complete the preliminary screening of resumes, visualize the scores of all resumes and feedback them to the human resources department, and add the resumes processed this time to the database; The information database is used to store resume sample data, employee performance sample data, and person-job matching weight coefficients.

9. A resume screening medium for matching positions based on big data analysis, characterized in that: The job matching initial screening storage medium is burned with a computer program, and the computer program implements the method described in any one of claims 1 to 7 when running in the memory of the server.

Images