Position matching method and device for business administration major

Abstract

The invention relates to the technical field of position planning, and particularly provides a position matching method and device for the business administration major, and the method comprises the following steps: carrying out the keyword extraction of a pre-filled student resume based on a recruitment keyword set; generating a position score based on the position of the resume keyword in the corresponding recruitment information, and generating a frequency score based on the occurrence frequency of the resume keyword in the corresponding recruitment information; generating a domain score corresponding to each resume keyword based on the similarity between the resume keyword and the professional keyword; calculating a matching score corresponding to each resume keyword according to the position score, the frequency score and the field score corresponding to the resume keyword; screening out target recruitment information from all the recruitment information based on the sum of all the matching scores corresponding to the same recruitment information; according to the method, the problem that high-quality matching of the recruitment information and the resume cannot be realized due to the fact that position matching is performed only depending on simple appearance of resume keywords can be effectively solved.

Description

Technical Field

[0001] This application relates to the field of job planning technology, and more specifically, to a job matching method and apparatus for business administration professionals. Background Technology

[0002] In the specific application scenario of job preparation for business administration students in universities, students need to use job matching systems to recommend suitable positions from job postings on various recruitment websites based on their resumes. These job matching systems rely on the number of resume keywords (keywords extracted from the resume) to recommend suitable positions. That is, the system defines positions corresponding to job postings with a preset number of resume keywords as suitable. However, because this technology relies solely on the simple occurrence of resume keywords for job matching, and different job postings show varying degrees of focus on the same keywords, it suffers from the problem of failing to achieve high-quality matching between job postings and resumes, resulting in low job matching accuracy.

[0003] Currently, there is no effective technical solution to the above-mentioned problems. It should be noted that the information disclosed in this section is only for understanding the background of the present invention and therefore may include information that does not constitute prior art. Summary of the Invention

[0004] The purpose of this application is to provide a job matching method and apparatus for business administration professionals, which can effectively solve the problem that high-quality matching of recruitment information and resumes cannot be achieved due to relying solely on the simple occurrence of keywords in resumes for job matching.

[0005] Firstly, this application provides a job matching method for business administration professionals, which includes the following steps: S1. Extract keywords from multiple job postings sequentially to obtain multiple sets of job posting keywords. Then, perform word segmentation, stop word removal, and synonym expansion on the sets of job posting keywords. Each set of job posting keywords corresponds to one job posting and includes multiple job posting keywords. S2. Extract keywords from pre-filled student resumes based on the recruitment keyword set, so that each recruitment information corresponds to a resume keyword set, and the resume keyword set includes multiple resume keywords; S3. Generate position scores for each resume keyword based on its position in the corresponding job posting, and generate frequency scores for each resume keyword based on its frequency of occurrence in the corresponding job posting. S4. Generate domain scores for each resume keyword based on the similarity between resume keywords and professional keywords in a pre-built business administration professional knowledge base. S5. Calculate the matching score for each resume keyword based on the position score, frequency score, and domain score corresponding to the resume keywords. S6. Based on the sum of all matching scores corresponding to the same job posting, filter out the target job posting from all job postings to complete the job matching.

[0006] This application provides a job matching method for business administration majors. First, it generates position scores, frequency scores, and domain scores corresponding to resume keywords. Then, based on these scores, it filters target job postings from all recruitment information. In other words, this application recommends suitable jobs from three perspectives: the position of resume keywords in the job postings, the frequency of their occurrence, and the similarity between resume keywords and professional keywords. Therefore, this application analyzes whether job postings match student resumes from multiple angles, effectively solving the problem of failing to achieve high-quality matching between job postings and resumes due to relying solely on the simple occurrence of resume keywords, thereby effectively improving job matching accuracy.

[0007] Optionally, step S3 includes: S31. Select a keyword from a resume; S32. Obtain job type information based on the recruitment information corresponding to the selected resume keywords; S33. Generate a position score based on the position of the selected resume keywords in the corresponding job posting and their corresponding job type information; S34. Generate a frequency score based on the frequency of the selected resume keywords in the corresponding job postings and their corresponding job type information. S35. Analyze whether there are any resume keywords that have not generated position scores and frequency scores. If so, select any resume keyword that has not generated position scores and frequency scores and return to step S32. If not, proceed to step S4.

[0008] Since different types of jobs pay varying degrees of attention to the position and frequency of the same resume keywords in job postings, this technical solution first obtains job type information, then generates a position score based on the position of the selected resume keywords in the corresponding job postings and their corresponding job type information, and generates a frequency score based on the frequency of the selected resume keywords in the corresponding job postings and their corresponding job type information. Therefore, this technical solution effectively considers the influence of job type on the degree of attention paid to the position and frequency of the resume keywords in job postings when generating position and frequency scores, thereby effectively improving the accuracy of position and frequency scores and thus effectively improving job matching accuracy.

[0009] Optionally, step S33 includes: S331. Obtain the position correction coefficient based on the position of the selected resume keywords in the corresponding job information, the corresponding job type information, and the first preset conversion relationship; S332. Calculate the position score based on the position correction coefficient and the preset preliminary position score.

[0010] Optionally, step S4 includes: S41. Select a keyword from a resume; S42. Obtain the domain score based on the maximum similarity between the selected resume keywords and the professional keywords in the pre-built business administration professional domain knowledge base and the second preset conversion relationship; S43. Analyze whether there are any resume keywords that have not generated domain scores. If so, select any resume keyword that has not generated domain scores and return to step S42. If not, proceed to step S5.

[0011] Because this technical solution obtains the domain score based on the maximum similarity between the selected resume keywords and the professional keywords in the pre-built business administration professional knowledge base, and the second preset conversion relationship, that is, the domain score of each resume keyword can reflect its degree of association with the most relevant professional knowledge in the business administration professional knowledge base, this technical solution can effectively improve the accuracy of the domain score, thereby further improving the job matching accuracy.

[0012] Optionally, step S42 includes: S421. Obtain job type information based on the recruitment information corresponding to the selected resume keywords; S422. Obtain the first correction coefficient corresponding to different professional keywords based on job type information and the third preset conversion relationship; S423. Calculate the similarity between the selected resume keywords and all professional keywords in the pre-built business administration professional knowledge base to obtain the preliminary similarity corresponding to each professional keyword. S424. Calculate the similarity between each professional keyword and the selected resume keywords based on the preliminary similarity and the first correction coefficient corresponding to the professional keywords, and obtain the domain score based on the maximum similarity and the second preset conversion relationship.

[0013] Since different types of positions pay varying degrees of attention to the same professional keywords, this technical solution first obtains the first correction coefficient corresponding to different professional keywords based on the job type information, and then obtains the preliminary similarity corresponding to each professional keyword. Next, it calculates the similarity between each professional keyword and the selected resume keywords based on the preliminary similarity and the first correction coefficient. Finally, it obtains the domain score based on the maximum similarity. In other words, this technical solution considers the influence of job type on the degree of attention to professional keywords when calculating the domain score. Therefore, this technical solution can further improve the accuracy of the domain score, thereby further improving the job matching accuracy.

[0014] Optionally, the third preset transformation relationship is a pre-constructed mapping relationship regarding job type, professional keywords, and correction coefficients. The pre-construction process of the third preset transformation relationship includes the following steps: Based on the clustering algorithm, the historical recruitment data is divided into multiple job type clusters according to the job descriptions and skill requirements in the historical recruitment data. Each job type cluster corresponds to a job type. The set of correction coefficients for each job type cluster is calculated based on the frequency of occurrence of different professional keywords in all historical recruitment data corresponding to the same job type cluster. Each set of correction coefficients includes multiple sets of professional keywords and their corresponding correction coefficients, so that each job type corresponds to a set of correction coefficients. A third preset transformation relationship is constructed based on the job type and the corresponding set of correction coefficients.

[0015] Optionally, the student's resume includes internship experience, and step S6 includes: S61. Calculate the sum of all matching scores corresponding to each recruitment information to obtain a preliminary total score. Each preliminary total score corresponds to one recruitment information. S62. Obtain the second correction coefficient corresponding to each recruitment information based on the text similarity and the fourth preset conversion relationship. The text similarity is the similarity between the description text of the internship experience and the description text of the recruitment information. S63. Calculate the matching total score for each recruitment information based on the preliminary total score and the second correction coefficient. S64. Select the job posting with the highest total score from all job postings as the target job posting to complete the job matching.

[0016] This technical solution performs job matching based on two dimensions: the degree of matching between resume keywords and job postings, and the similarity between internship experience and job postings. Therefore, this technical solution can effectively improve the matching quality between job postings and student resumes, thereby further improving the accuracy of job matching.

[0017] Optionally, step S6 may also include steps performed prior to step S63: S65. Obtain the number of user clicks and posting duration of recruitment information, and calculate the job popularity corresponding to each recruitment information based on the number of user clicks and posting duration using a time decay function. S66. Obtain the third correction coefficient for each job posting based on job popularity and the fifth preset conversion relationship; Step S63 includes: S631. Calculate the matching total score for each recruitment information based on the preliminary total score, the second correction coefficient, and the third correction coefficient.

[0018] Since job popularity is positively correlated with job competition intensity and recruitment standards, changes in job competition intensity and recruitment standards can lead to changes in students' job search success rate. This technical solution takes into account the impact of job popularity when matching jobs, thus enabling the target recruitment information to better match students' actual situation and market demand, thereby effectively improving the accuracy and quality of job matching.

[0019] Optionally, step S6 may also include steps performed prior to step S63: S67. Calculate the number of job openings for different job types based on all recruitment information, and calculate the fourth correction coefficient based on the number of job openings corresponding to the job type of the recruitment information and the total number of job openings. Step S631 includes: S6311. Calculate the matching total score for each recruitment information based on the preliminary total score, the second correction coefficient, the third correction coefficient, and the fourth correction coefficient.

[0020] Because this technical solution introduces a fourth correction coefficient when calculating the total matching score, and this fourth correction coefficient can reflect the market demand ratio for different job types, the calculation of the total matching score of this technical solution takes into account the market demand for different job types. Therefore, this technical solution can make the job matching results more comprehensive, balanced and in line with the market supply and demand relationship, thereby effectively improving the rationality of job matching.

[0021] Secondly, this application also provides a job matching device for business administration professionals, comprising: The recruitment keyword acquisition module is used to extract keywords from multiple recruitment information in sequence to obtain multiple recruitment keyword sets. The recruitment keyword sets are then segmented, stop words are removed, and synonyms are expanded. Each recruitment keyword set corresponds to one recruitment information and includes multiple recruitment keywords. The resume keyword acquisition module is used to extract keywords from pre-filled student resumes based on a set of recruitment keywords, so that each recruitment information corresponds to a set of resume keywords, and the set of resume keywords includes multiple resume keywords. The position score and frequency score acquisition module is used to generate position scores for each resume keyword based on the position of the resume keywords in the corresponding job postings, and to generate frequency scores for each resume keyword based on the frequency of their appearance in the corresponding job postings. The domain score acquisition module is used to generate domain scores for each resume keyword based on the similarity between resume keywords and professional keywords in a pre-built business administration professional domain knowledge base. The matching score acquisition module is used to calculate the matching score for each resume keyword based on the position score, frequency score, and domain score corresponding to the resume keywords. The job matching module is used to filter target job information from all job postings based on the sum of all matching scores corresponding to the same job posting, in order to complete job matching.

[0022] This application provides a job matching device for business administration majors. It first generates position scores, frequency scores, and domain scores corresponding to resume keywords. Then, based on these scores, it filters target job postings from all recruitment information. In other words, this application recommends suitable jobs from three perspectives: the position of resume keywords in the job postings, the frequency of their occurrence, and the similarity between resume keywords and professional keywords. Therefore, this application analyzes whether job postings match student resumes from multiple angles, effectively solving the problem of failing to achieve high-quality matching between job postings and resumes due to relying solely on the simple occurrence of resume keywords, thereby effectively improving job matching accuracy.

[0023] As can be seen from the above, the job matching method and apparatus for business administration majors provided in this application first generates position scores, frequency scores, and domain scores corresponding to resume keywords, and then filters target recruitment information from all recruitment information based on the position scores, frequency scores, and domain scores. That is, this application recommends suitable positions from three perspectives: the position of resume keywords in recruitment information, the frequency of resume keywords in recruitment information, and the similarity between resume keywords and professional keywords. Therefore, this application is equivalent to analyzing whether recruitment information matches student resumes from multiple perspectives, thereby effectively solving the problem that high-quality matching of recruitment information and resumes cannot be achieved due to relying solely on the simple occurrence of resume keywords for job matching, and thus effectively improving the accuracy of job matching. Attached Figure Description

[0024] Figure 1 A flowchart illustrating a job matching method for business administration professionals, provided as an embodiment of this application.

[0025] Figure 2 This is a schematic diagram of a job matching device for business administration professionals, provided as an embodiment of this application.

[0026] Attached labels: 1. Recruitment keyword acquisition module; 2. Resume keyword acquisition module; 3. Location score and frequency score acquisition module; 4. Domain score acquisition module; 5. Matching score acquisition module; 6. Job matching module. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0028] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0029] Firstly, such as Figure 1As shown, this application provides a job matching method for business administration professionals, which includes the following steps: S1. Extract keywords from multiple job postings sequentially to obtain multiple sets of job posting keywords. Then, perform word segmentation, stop word removal, and synonym expansion on the sets of job posting keywords. Each set of job posting keywords corresponds to one job posting and includes multiple job posting keywords. S2. Extract keywords from pre-filled student resumes based on the recruitment keyword set, so that each recruitment information corresponds to a resume keyword set, and the resume keyword set includes multiple resume keywords; S3. Generate position scores for each resume keyword based on its position in the corresponding job posting, and generate frequency scores for each resume keyword based on its frequency of occurrence in the corresponding job posting. S4. Generate domain scores for each resume keyword based on the similarity between resume keywords and professional keywords in a pre-built business administration professional knowledge base. S5. Calculate the matching score for each resume keyword based on the position score, frequency score, and domain score corresponding to the resume keywords. S6. Based on the sum of all matching scores corresponding to the same job posting, filter out the target job posting from all job postings to complete the job matching.

[0030] Step S1 can utilize the existing TF-IDF algorithm to extract keywords from recruitment information. Step S1 can also utilize the existing jieba word segmentation tool to segment the recruitment keyword set. Step S1 can remove stop words from the recruitment keyword set using a pre-collected general stop word list. Step S1 can also expand the recruitment keyword set with synonyms using existing thesaurus such as WordNet. Step S1 reduces noise interference and improves the quality of recruitment keywords by segmenting the recruitment information and removing stop words. Since Step S2 needs to extract keywords from student resumes based on the recruitment keyword set, Step S1 can avoid the situation where some keywords matching the recruitment keywords are missed due to differences in word choice by expanding the recruitment keyword set with synonyms.

[0031] Step S2 can utilize the existing TF-IDF algorithm to extract keywords from pre-filled student resumes based on the recruitment keyword set, thereby obtaining multiple resume keyword sets. Each recruitment information corresponds to a resume keyword set, and each resume keyword set includes multiple resume keywords. Since the resume keywords are obtained by extracting keywords from student resumes based on recruitment keywords, each resume keyword corresponds to a recruitment keyword.

[0032] This embodiment allows for configuring different weight coefficients for different positions within the recruitment information. Step S3, generating a position score for the resume keywords based on their position in the corresponding recruitment information, can be achieved by: determining the corresponding weight coefficient based on the position of the resume keywords in the corresponding recruitment information; and multiplying the preset position score by the weight coefficient corresponding to the resume keywords to obtain the position score. This embodiment can also set multiple frequency ranges and assign a corresponding score to each frequency range. The frequency of resume keywords appearing in the corresponding recruitment information reflects the number of times the resume keywords appear in the corresponding recruitment information. Step S3, generating a frequency score for the resume keywords based on their frequency of appearance in the corresponding recruitment information, can be achieved by: using the score corresponding to the frequency range in which the resume keywords appear in the corresponding recruitment information as the frequency score.

[0033] Step S4 stores multiple professional keywords in the business administration professional field database. These keywords can be professional terms, skills, or knowledge points in the business administration patent field. Step S4 can use existing word vector models such as word2vec or fasttext to calculate the similarity between the resume keywords and the professional keywords in the pre-built business administration professional field knowledge base. This embodiment can set multiple similarity ranges and assign a corresponding score to each similarity range. The process of generating the field score for each resume keyword based on the similarity between the resume keywords and the professional keywords in the pre-built business administration professional field knowledge base in Step S4 can be as follows: the score corresponding to the similarity range between the resume keywords and the professional keywords in the pre-built business administration professional field knowledge base is used as the frequency score. Step S5 can calculate the matching score corresponding to the resume keywords by summing the position score, frequency score, and field score corresponding to the resume keywords. Step S6 can first calculate the sum of all matching scores corresponding to each job posting, and then filter out the job posting with the largest sum of matching scores from all job postings as the target job posting. Since job postings are associated with positions, step S6 can complete job matching by filtering out the target job posting from all job postings based on the sum of all matching scores corresponding to the same job posting.

[0034] The working principle of this embodiment is as follows: the position of keywords in job postings reflects the degree of attention a company pays to those keywords. For example, a company pays more attention to keywords appearing in the title or job requirements than to keywords appearing in the company profile. Therefore, the position score generated based on the position of resume keywords in the corresponding job postings reflects the degree of matching between the resume keywords and the job postings. Furthermore, the frequency of keyword appearance in job postings also reflects the company's degree of attention to those keywords. For example, if "financial analysis" appears repeatedly in job postings, it indicates a high degree of attention from the company to this keyword. Therefore, the frequency score generated based on the frequency of resume keywords in the corresponding job postings also reflects the degree of matching between the resume keywords and the job postings. The similarity score reflects the degree of matching between resume keywords and job posting information. Since professional keywords are professional terms, skills, or knowledge points in the field of business administration, the similarity between resume keywords and professional keywords in a pre-built business administration professional knowledge base can reflect the likelihood that a student possesses the professional knowledge corresponding to that keyword. The higher the similarity, the greater the likelihood that the student possesses the corresponding professional knowledge. In other words, the similarity score can reflect the student's professional ability level, which is also a key focus for enterprises. Therefore, the student's professional ability level is positively correlated with the degree of matching between the student and the job posting information. Thus, the domain score generated based on the similarity between resume keywords and professional keywords in a pre-built business administration professional knowledge base can also reflect the degree of matching between resume keywords and job posting information.

[0035] This application provides a job matching method for business administration majors. First, it generates position scores, frequency scores, and domain scores corresponding to resume keywords. Then, based on these scores, it filters target job postings from all recruitment information. In other words, this application recommends suitable jobs from three perspectives: the position of resume keywords in the job postings, the frequency of their occurrence, and the similarity between resume keywords and professional keywords. Therefore, this application analyzes whether job postings match student resumes from multiple angles, effectively solving the problem of failing to achieve high-quality matching between job postings and resumes due to relying solely on the simple occurrence of resume keywords, thereby effectively improving job matching accuracy.

[0036] In some preferred embodiments, step S3 includes: S31. Select a keyword from a resume; S32. Obtain job type information based on the recruitment information corresponding to the selected resume keywords; S33. Generate a position score based on the position of the selected resume keywords in the corresponding job posting and their corresponding job type information; S34. Generate a frequency score based on the frequency of the selected resume keywords in the corresponding job postings and their corresponding job type information. S35. Analyze whether there are any resume keywords that have not generated position scores and frequency scores. If so, select any resume keyword that has not generated position scores and frequency scores and return to step S32. If not, proceed to step S4.

[0037] This embodiment can select resume keywords by traversal, that is, selecting resume keywords sequentially according to their order in the resume keyword set. The job type information in step S32 is extracted from the recruitment information corresponding to the selected resume keywords. For example, the job type information is determined by analyzing keywords in the title or job description of the recruitment information. This job type information can be marketing, financial management, or human resources, etc. The specific process for generating the position score in step S33 can be as follows: first, a preliminary position score is generated based on the position of the selected resume keywords in the corresponding recruitment information; then, the preliminary position score is adjusted based on the job type information (for example, financial management positions focus more on the appearance of resume keywords in professional positions such as skill requirements or project experience, while marketing positions may focus more on the appearance of resume keywords in core positions such as job description or job requirements. Therefore, for financial management positions, this embodiment will increase the preliminary position score of resume keywords appearing in skill requirements or project experience in the recruitment information; for marketing positions, this embodiment will increase the preliminary position score of resume keywords appearing in job description or job requirements in the recruitment information) to obtain the final position score. The specific process for generating the position score in step S34 can be as follows: First, a preliminary score is generated based on the frequency of the selected resume keywords in the corresponding job postings. Then, the preliminary frequency score is adjusted based on the job type information (for example, for financial management positions, since financial management positions pay more attention to resume keywords related to financial management than to resume keywords related to communication skills, this embodiment will adjust the preliminary frequency score of resume keywords related to financial management to be higher than that of resume keywords related to communication skills when the frequencies are the same), to obtain the frequency score. Since different types of positions pay different amounts of attention to the position and frequency of the same resume keyword in job postings, this embodiment first obtains the job type information, then generates a position score based on the position of the selected resume keywords in the corresponding job postings and its corresponding job type information, and generates a frequency score based on the frequency of the selected resume keywords in the corresponding job postings and its corresponding job type information. Therefore, this embodiment effectively considers the influence of job type on the attention paid to the position and frequency of the resume keywords in job postings when generating the position score and frequency score, thereby effectively improving the accuracy of the position score and frequency score, and thus effectively improving the job matching accuracy.

[0038] In some preferred embodiments, step S33 includes: S331. Obtain the position correction coefficient based on the position of the selected resume keywords in the corresponding job information, the corresponding job type information, and the first preset conversion relationship; S332. Calculate the position score based on the position correction coefficient and the preset preliminary position score.

[0039] The first preset conversion relationship in this embodiment is a pre-constructed mapping relationship between job type, keyword position in corresponding recruitment information, and correction coefficient. This embodiment can extract the corresponding position correction coefficient from the first preset conversion relationship based on the position of the selected resume keywords in the corresponding recruitment information and the job type information. Step S332 calculates the position score by multiplying the position correction coefficient by a preset preliminary position score. It should be understood that the specific process of calculating the frequency score in step S34 is similar to the specific process of calculating the position score in this embodiment; therefore, this application will not discuss step S34 in detail.

[0040] In some preferred embodiments, step S4 includes: S41. Select a keyword from a resume; S42. Obtain the domain score based on the maximum similarity between the selected resume keywords and the professional keywords in the pre-built business administration professional domain knowledge base and the second preset conversion relationship; S43. Analyze whether there are any resume keywords that have not generated domain scores. If so, select any resume keyword that has not generated domain scores and return to step S42. If not, proceed to step S5.

[0041] This embodiment can use a cosine similarity algorithm or other text similarity calculation methods to calculate the similarity between selected resume keywords and professional keywords. After obtaining a series of similarity values, this embodiment selects the one with the largest value. This maximum value represents the degree of association between the resume keywords and the most relevant professional keywords in the business administration professional knowledge base. The second preset transformation relationship in this embodiment can be a pre-constructed functional relationship or a mapping relationship between similarity and domain score. This embodiment can obtain the domain score by inputting the maximum similarity value into the second preset transformation relationship. Since this embodiment obtains the domain score based on the maximum similarity between the selected resume keywords and the professional keywords in the pre-constructed business administration professional knowledge base and the second preset transformation relationship, that is, the domain score of each resume keyword can reflect its degree of association with the most relevant professional knowledge in the business administration professional knowledge base, this embodiment can effectively improve the accuracy of the domain score, thereby further improving the job matching accuracy.

[0042] In some preferred embodiments, step S42 includes: S421. Obtain job type information based on the recruitment information corresponding to the selected resume keywords; S422. Obtain the first correction coefficient corresponding to different professional keywords based on job type information and the third preset conversion relationship; S423. Calculate the similarity between the selected resume keywords and all professional keywords in the pre-built business administration professional knowledge base to obtain the preliminary similarity corresponding to each professional keyword. S424. Calculate the similarity between each professional keyword and the selected resume keywords based on the preliminary similarity and the first correction coefficient corresponding to the professional keywords, and obtain the domain score based on the maximum similarity and the second preset conversion relationship.

[0043] The third preset conversion relationship in this embodiment can be a pre-constructed mapping relationship regarding job type, professional keywords, and correction coefficients. Step S424 can calculate the similarity between professional keywords and selected resume keywords by multiplying the preliminary similarity corresponding to the professional keywords by the first correction coefficient. Since different types of jobs have different levels of attention to the same professional keyword, this embodiment first obtains the first correction coefficient corresponding to different professional keywords based on job type information, and obtains the preliminary similarity corresponding to each professional keyword. Then, it calculates the similarity between each professional keyword and the selected resume keywords based on the preliminary similarity and the first correction coefficient. Finally, it obtains the domain score based on the maximum similarity. That is, this embodiment is equivalent to considering the influence of job type on the degree of attention to professional keywords when calculating the domain score. Therefore, this embodiment can further improve the accuracy of the domain score, thereby further improving the job matching accuracy.

[0044] In some preferred embodiments, the third preset conversion relationship is a pre-constructed mapping relationship regarding job type, professional keywords, and correction coefficients. The pre-construction process of the third preset conversion relationship includes the following steps: Based on the clustering algorithm, the historical recruitment data is divided into multiple job type clusters according to the job descriptions and skill requirements in the historical recruitment data. Each job type cluster corresponds to a job type. The set of correction coefficients for each job type cluster is calculated based on the frequency of occurrence of different professional keywords in all historical recruitment data corresponding to the same job type cluster. Each set of correction coefficients includes multiple sets of professional keywords and their corresponding correction coefficients, so that each job type corresponds to a set of correction coefficients. A third preset transformation relationship is constructed based on the job type and the corresponding set of correction coefficients.

[0045] The historical recruitment data in this embodiment can be recruitment information released by enterprises in previous years. This historical recruitment data includes job descriptions and skill requirements. The clustering algorithm in this embodiment is preferably the K-means algorithm. Since the higher the frequency of professional keywords in all historical recruitment data corresponding to the same job type cluster, the more important the professional keyword is in the job type corresponding to that job type cluster, this embodiment can calculate the correction coefficients corresponding to different professional keywords in the same job type cluster based on the frequency of different professional keywords in all historical recruitment data corresponding to the same job type cluster. Specifically, this embodiment can obtain the correction coefficients corresponding to different professional keywords in the same job type cluster by normalizing the frequency of different professional keywords in the same job type cluster. After completing the calculation of the set of correction coefficients, each correction coefficient corresponds to a professional keyword and a job type. Therefore, this embodiment can construct a third preset mapping relationship based on the existing mapping relationship construction method according to the job type and the corresponding set of correction coefficients.

[0046] In some preferred embodiments, the student resume includes internship experience, and step S6 includes: S61. Calculate the sum of all matching scores corresponding to each recruitment information to obtain a preliminary total score. Each preliminary total score corresponds to one recruitment information. S62. Obtain the second correction coefficient corresponding to each recruitment information based on the text similarity and the fourth preset conversion relationship. The text similarity is the similarity between the description text of the internship experience and the description text of the recruitment information. S63. Calculate the matching total score for each recruitment information based on the preliminary total score and the second correction coefficient. S64. Select the job posting with the highest total score from all job postings as the target job posting to complete the job matching.

[0047] The fourth preset conversion relationship in this embodiment can be a pre-constructed mapping relationship between text similarity and correction coefficient. Step S63 can calculate the total matching score corresponding to the recruitment information by multiplying the preliminary total score corresponding to the recruitment information by the second correction coefficient. In this embodiment, the text similarity is the similarity between the description text of the internship experience and the description text of the recruitment information. The higher the text similarity, the stronger the correlation between the internship experience and the recruitment information, and the higher the matching degree between the student and the recruitment information. This embodiment is equivalent to performing job matching from two dimensions: the matching degree between resume keywords and recruitment information, and the similarity between internship experience and recruitment information. Therefore, this embodiment can effectively improve the matching quality between recruitment information and student resumes, thereby further improving the job matching accuracy. Specifically, taking an internship experience description of a business administration student as "Interning in the marketing department of a well-known internet company, responsible for market research and data analysis, and participating in multiple marketing promotion activities" as an example, the initial total score calculated from the student's resume keywords for the position of "Marketing Manager" is 80 points. The initial total score calculated from the student's resume keywords for the position of "Financial Analyst" is also 80 points. However, because the similarity between the description of the internship experience and the description of the job posting for Marketing Manager is greater than the similarity between the description of the internship experience and the description of the job posting for Financial Analyst, the second correction coefficient for Marketing Manager will be greater than that for Financial Analyst. Therefore, the total matching score for Marketing Manager will be higher than that for Financial Analyst, and the final target job posting will be the job posting for Marketing Manager.

[0048] In some preferred embodiments, step S6 further includes steps performed before step S63: S65. Obtain the number of user clicks and posting duration of recruitment information, and calculate the job popularity corresponding to each recruitment information based on the number of user clicks and posting duration using a time decay function. S66. Obtain the third correction coefficient for each job posting based on job popularity and the fifth preset conversion relationship; Step S63 includes: S631. Calculate the matching total score for each recruitment information based on the preliminary total score, the second correction coefficient, and the third correction coefficient.

[0049] In this embodiment, user clicks reflect the number of times users view recruitment information and its popularity. The posting duration is the time difference between the time the recruitment information was posted and the current time. Both user clicks and posting duration can be obtained from the recruitment website's backend data interface. This embodiment calculates the job popularity corresponding to each recruitment posting based on a time decay function, considering user clicks and posting duration. With the same posting duration, higher user clicks indicate higher job popularity; conversely, with the same user clicks, longer posting duration indicates lower job popularity. The fifth preset conversion relationship in this embodiment can be a pre-constructed mapping relationship between job popularity and a correction coefficient. Since job popularity is positively correlated with job competition intensity and recruitment standards, changes in job competition intensity and recruitment standards can alter students' job search success rates. This embodiment considers the impact of job popularity when matching jobs, thus enabling target recruitment information to better match students' actual situations and market demands, effectively improving the accuracy and quality of job matching.

[0050] In some preferred embodiments, step S6 further includes steps performed before step S63: S67. Calculate the number of job openings for different job types based on all recruitment information, and calculate the fourth correction coefficient based on the number of job openings corresponding to the job type of the recruitment information and the total number of job openings. Step S631 includes: S6311. Calculate the matching total score for each recruitment information based on the preliminary total score, the second correction coefficient, the third correction coefficient, and the fourth correction coefficient.

[0051] The fourth correction coefficient in this embodiment can be the ratio of the number of job openings corresponding to the job type in the recruitment information to the total number of job openings. This fourth correction coefficient reflects the market demand ratio for different job types. Because this embodiment introduces the fourth correction coefficient when calculating the total matching score, and this fourth correction coefficient reflects the market demand ratio for different job types, the calculation of the total matching score in this embodiment takes into account the market demand for different job types. Therefore, this embodiment can make the job matching results more comprehensive, balanced, and in line with market supply and demand, thereby effectively improving the rationality of job matching.

[0052] As can be seen from the above, the job matching method for business administration majors provided in this application first generates position scores, frequency scores, and domain scores corresponding to resume keywords, and then filters target recruitment information from all recruitment information based on the position scores, frequency scores, and domain scores. That is, this application recommends suitable positions from three perspectives: the position of resume keywords in recruitment information, the frequency of resume keywords in recruitment information, and the similarity between resume keywords and professional keywords. Therefore, this application is equivalent to analyzing whether recruitment information matches student resumes from multiple perspectives, thereby effectively solving the problem that high-quality matching of recruitment information and resumes cannot be achieved due to relying solely on the simple occurrence of resume keywords for job matching, and thus effectively improving the accuracy of job matching.

[0053] Secondly, such as Figure 2 As shown, this application also provides a job matching device for business administration professionals, which includes: The recruitment keyword acquisition module 1 is used to extract keywords from multiple recruitment information in sequence to obtain multiple recruitment keyword sets. The recruitment keyword sets are then segmented, stop words are removed, and synonyms are expanded. Each recruitment keyword set corresponds to one recruitment information and includes multiple recruitment keywords. The resume keyword acquisition module 2 is used to extract keywords from pre-filled student resumes based on a set of recruitment keywords, so that each recruitment information corresponds to a set of resume keywords, and the set of resume keywords includes multiple resume keywords. The position score and frequency score acquisition module 3 is used to generate position scores for each resume keyword based on the position of the resume keywords in the corresponding job information, and to generate frequency scores for each resume keyword based on the frequency of occurrence of the resume keywords in the corresponding job information. Domain score acquisition module 4 is used to generate domain scores for each resume keyword based on the similarity between resume keywords and professional keywords in a pre-built business administration professional domain knowledge base. The matching score acquisition module 5 is used to calculate the matching score for each resume keyword based on the position score, frequency score, and domain score corresponding to the resume keywords. Job matching module 6 is used to filter target job information from all job information based on the sum of all matching scores corresponding to the same job information, so as to complete job matching.

[0054] This application provides a job matching device for business administration professionals, comprising a recruitment keyword acquisition module 1, a resume keyword acquisition module 2, a location score and frequency score acquisition module 3, a domain score acquisition module 4, a matching score acquisition module 5, and a job matching module 6. This job matching device for business administration professionals is used to execute the steps of the job matching method for business administration professionals provided in the first aspect above. The principle of this job matching device for business administration professionals is the same as that of the job matching method for business administration professionals provided in the first aspect above, and will not be discussed in detail here.

[0055] As can be seen from the above, the job matching method and apparatus for business administration majors provided in this application first generates position scores, frequency scores, and domain scores corresponding to resume keywords, and then filters target recruitment information from all recruitment information based on the position scores, frequency scores, and domain scores. That is, this application recommends suitable positions from three perspectives: the position of resume keywords in recruitment information, the frequency of resume keywords in recruitment information, and the similarity between resume keywords and professional keywords. Therefore, this application is equivalent to analyzing whether recruitment information matches student resumes from multiple perspectives, thereby effectively solving the problem that high-quality matching of recruitment information and resumes cannot be achieved due to relying solely on the simple occurrence of resume keywords for job matching, and thus effectively improving the accuracy of job matching.

[0056] In the embodiments provided in this application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of the above units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple units or components may be combined or integrated into another robot, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0057] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0058] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0059] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A job matching method for business administration majors, characterized in that, The job matching method for business administration professionals includes the following steps: S1. Extract keywords from multiple job postings sequentially to obtain multiple sets of job posting keywords. Then, perform word segmentation, stop word removal, and synonym expansion on the sets of job posting keywords. Each set of job posting keywords corresponds to one job posting and includes multiple job posting keywords. S2. Extract keywords from the pre-filled student resumes based on the recruitment keyword set, so that each recruitment information corresponds to a resume keyword set, and the resume keyword set includes multiple resume keywords; S3. Generate a position score for each resume keyword based on its position in the corresponding job posting, and generate a frequency score for each resume keyword based on its frequency of occurrence in the corresponding job posting. S4. Generate a domain score for each resume keyword based on the similarity between the resume keywords and the professional keywords in the pre-built business administration professional knowledge base. S5. Calculate the matching score for each resume keyword based on the position score, frequency score, and domain score corresponding to the resume keywords; S6. Based on the sum of all the matching scores corresponding to the same recruitment information, the target recruitment information is filtered out from all the recruitment information to complete the job matching.

2. The job matching method for business administration professionals according to claim 1, characterized in that, Step S3 includes: S31. Select a keyword from a resume; S32. Obtain job type information based on the recruitment information corresponding to the selected resume keywords; S33. Generate a position score based on the position of the selected resume keywords in the corresponding job posting and their corresponding job type information; S34. Generate a frequency score based on the frequency of the selected resume keywords in the corresponding job postings and their corresponding job type information. S35. Analyze whether there are any resume keywords that have not generated position scores and frequency scores. If so, select any resume keyword that has not generated position scores and frequency scores and return to step S32. If not, proceed to step S4.

3. The job matching method for business administration professionals according to claim 2, characterized in that, Step S33 includes: S331. Obtain the position correction coefficient based on the position of the selected resume keywords in the corresponding job information, the corresponding job type information, and the first preset conversion relationship; S332. Calculate the position score based on the position correction coefficient and the preset preliminary position score.

4. The job matching method for business administration professionals according to claim 1, characterized in that, Step S4 includes: S41. Select a keyword from a resume; S42. Obtain the domain score based on the maximum similarity between the selected resume keywords and the professional keywords in the pre-built business administration professional domain knowledge base and the second preset conversion relationship; S43. Analyze whether there are any resume keywords that have not generated domain scores. If so, select any resume keyword that has not generated domain scores and return to step S42. If not, proceed to step S5.

5. The job matching method for business administration professionals according to claim 4, characterized in that, Step S42 includes: S421. Obtain job type information based on the recruitment information corresponding to the selected resume keywords; S422. Obtain the first correction coefficient corresponding to different professional keywords based on the job type information and the third preset conversion relationship; S423. Calculate the similarity between the selected resume keywords and all professional keywords in the pre-built business administration professional knowledge base to obtain the preliminary similarity corresponding to each professional keyword. S424. Calculate the similarity between each professional keyword and the selected resume keywords based on the preliminary similarity and the first correction coefficient corresponding to the professional keywords, and obtain the domain score according to the maximum value of the similarity and the second preset conversion relationship.

6. The job matching method for business administration professionals according to claim 5, characterized in that, The third preset conversion relationship is a pre-constructed mapping relationship regarding job type, professional keywords, and correction coefficients. The pre-construction process of the third preset conversion relationship includes the following steps: Based on the clustering algorithm, the historical recruitment data is divided into multiple job type clusters according to the job descriptions and skill requirements in the historical recruitment data, and each job type cluster corresponds to a job type. Based on the frequency of occurrence of different professional keywords in all historical recruitment data corresponding to the same job type cluster, the set of correction coefficients corresponding to each job type cluster is calculated. Each set of correction coefficients includes multiple sets of professional keywords and their corresponding correction coefficients, so that each job type corresponds to a set of correction coefficients. A third preset conversion relationship is constructed based on the job type and the corresponding set of correction coefficients.

7. The job matching method for business administration professionals according to claim 1, characterized in that, The student resume includes internship experience. Step S6 includes: S61. Calculate the sum of all the matching scores corresponding to each of the recruitment information to obtain a preliminary total score, where each preliminary total score corresponds to one of the recruitment information. S62. Obtain the second correction coefficient corresponding to each of the recruitment information based on the text similarity and the fourth preset conversion relationship, wherein the text similarity is the similarity between the description text of the internship experience and the description text of the recruitment information; S63. Calculate the matching total score corresponding to each of the recruitment information based on the preliminary total score and the second correction coefficient. S64. Select the recruitment information with the highest total matching score from all the recruitment information as the target recruitment information to complete the job matching.

8. The job matching method for business administration professionals according to claim 7, characterized in that, Step S6 also includes steps performed before step S63: S65. Obtain the number of user clicks and the posting duration of the recruitment information, and calculate the job popularity corresponding to each recruitment information based on the number of user clicks and the posting duration using a time decay function; S66. Obtain the third correction coefficient corresponding to each of the recruitment information based on the job popularity and the fifth preset conversion relationship; Step S63 includes: S631. Calculate the matching total score corresponding to each recruitment information based on the preliminary total score corresponding to the recruitment information, the second correction coefficient, and the third correction coefficient.

9. The job matching method for business administration professionals according to claim 8, characterized in that, Step S6 also includes steps performed before step S63: S67. Calculate the number of job requirements for different job types based on all the recruitment information, and calculate the fourth correction coefficient based on the number of job requirements corresponding to the job type of the recruitment information and the total number of job requirements. Step S631 includes: S6311. Calculate the matching total score corresponding to each recruitment information based on the preliminary total score corresponding to the recruitment information, the second correction coefficient, the third correction coefficient, and the fourth correction coefficient.

10. A job matching device for business administration majors, characterized in that, The job matching device for business administration professionals includes: The recruitment keyword acquisition module is used to extract keywords from multiple recruitment information in sequence to obtain multiple recruitment keyword sets, and to perform word segmentation, stop word removal and synonym expansion on the recruitment keyword sets. Each recruitment keyword set corresponds to one recruitment information and includes multiple recruitment keywords. The resume keyword acquisition module is used to extract keywords from pre-filled student resumes based on the recruitment keyword set, so that each recruitment information corresponds to a resume keyword set, and the resume keyword set includes multiple resume keywords. The position score and frequency score acquisition module is used to generate a position score for each resume keyword based on the position of the resume keyword in the corresponding job information, and to generate a frequency score for each resume keyword based on the frequency of occurrence of the resume keyword in the corresponding job information. The domain score acquisition module is used to generate a domain score for each resume keyword based on the similarity between the resume keywords and professional keywords in a pre-built business administration professional domain knowledge base. The matching score acquisition module is used to calculate the matching score corresponding to each of the resume keywords based on the position score, frequency score and domain score corresponding to the resume keywords. The job matching module is used to filter target job information from all the job information based on the sum of all the matching scores corresponding to the same job information, so as to complete job matching.

Images