KR20220080338A - System for matching human resources by selectively weighting keyword matching - Google Patents

Abstract

The present invention includes: a resume review unit for receiving a resume uploaded through an expert terminal, and extracting a specification keyword by reviewing the resume; a recruit review unit that receives a recruit document uploaded through a corporate terminal of a recruiting company, reviews the request document, and extracts keywords for needs; a code matching unit for rearranging a predetermined order by matching the specification keyword and the needs keyword with each other, arranging them in a predetermined order, and assigning a matching rate weight according to the past actual matching rate; And discloses a technical idea, characterized in that it comprises a record unit that provides mutual recommendations between the expert and the hiring company according to the rearranged predetermined order.

Description

{System for matching human resources by selectively weighting keyword matching}

The present invention relates to a system for matching human resources, and more specifically, through mutual keyword matching on the Rejume of an expert, such as a freelancer, and a Recruit Document of a hiring company, and AI deep learning of these matching, the expert freelancer is optimized It is a technical field related to the human resource matching system that allows matching with companies of

The freelance economic market is showing rapid growth as the global market economy is revitalized and the “gig economy” based on online payment platforms and social media (SNS) is expanding.

The freelance economy, which has seen explosive growth over the past decade, is booming with a generational shift happening around the world. Millennials-Generation Z, who seek independence and new business opportunities, are attracting attention in the freelance market as an attractive option to expand their careers as well as a new form of work with freedom and flexibility.

According to the '2020 Global Freelance Market' report conducted by Payoneer for 7,000 freelancers in 150 countries around the world who use its services, about 70% of the freelancers who participated in the survey were under the age of 35, and those under the age of 25 were 21 %, the younger generation is leading the market.

On the other hand, the income of those 55 and older, known as baby boomers, was more than double that of the younger generation (18-24). In Asia Pacific, the percentage of freelancers under the age of 35 was 82%, the highest among regions worldwide, and North America had the lowest rate at 47%.

As the proportion of the younger generation increased, the most popular fields for freelancers were web and graphic design (30%), programming (19%), and design/technology fields such as IT (10%). Finance, quality control (QA), sales, project management, and customer service sectors showed a low proportion of 1-2%.

The proportion of women in the freelance workforce was 24%, lower than the proportion of women in the global workforce (39%), but it showed a slight increase. The ratio of female freelancers was high in fields such as translation (50%), administrative/business management (46%), and content writers (43%). The percentage of female freelancers by country is highest in the Philippines at 62%, followed by the United States at 46%.

The average hourly wage for freelancers worldwide is $21 an hour, up about 10% from two years ago. The average hourly wage for women freelancers ($18.15) was only 84% of that of men ($21.67), so the wage gap still existed.

In finance and project management, women's average hourly wages were 53% and 56% lower than men's. In the fields of marketing (+17%), web and graphic design (+4%), and multimedia production (+0.3%), women received higher hourly wages than men.

The average hourly wage of female freelancers in Asia Pacific was the smallest at 87% of men, while the gap was the largest in North America at 69%.

It was found that freelancers actively use social media to promote and communicate with customers. Freelancers using social media increased from 65% in 2018 to 74% in 2019. Facebook (34%) is the most used, followed by LinkedIn (28%), Instagram (21%), Twitter (13%) and YouTube (9%). Instagram has grown significantly, from 12% in 2018 to 21% in 2020.

About 80% of respondents said they were satisfied with their freelance life. Freelancers in the business/administrative management field had the lowest average hourly wage, but the highest level of satisfaction. Conversely, workers in the multimedia production field had the lowest job satisfaction-to-income ratio despite the highest wages. Although the use of shared offices is increasing, most of the freelancers said that they work from home (83%).

"Talent recommendation and evaluation system and method (Registration No. 10-1292336, Patent Document 1)" exists as a published prior patent document.

In the case of the invention according to Patent Document 1, it relates to a talent recommendation and evaluation system and method, which promotes smooth employment of high-quality personnel by establishing an online employment marketplace to provide optimal job and job opportunities between companies and job seekers it is to do

In the case of the invention according to Patent Document 1, a talent recommendation module (step) for requesting a job seeker's talent recommendation availability and work rating to the job seeker's existing company; a talent information request module (step) for requesting job seeker information for job-seeker job competency evaluation; Talent evaluation module for statistically quantifying and storing the job performance rating of talent based on the job-seeker's work rating for each field recommended through the talent recommendation module (step) and job seeker information collected through the talent information request module (step) (step) and; a talent search module (step) for enabling the search for job seekers by field stored in the talent evaluation module (step), information of the searched job seeker, and job performance rating; It is characterized in that it includes a talent salary display module (step) for receiving and disclosing the salary offer amount information in real time in an auction method from a company that wants to hire a job seeker searched through the talent search module (step).

In addition, "a wallet system-based talent matching system using machine learning and a method therefor (registration number 10-1875923, Patent Document 2)" also exist.

In the case of the invention according to Patent Document 2, the talent matching server according to one aspect classifies the job seeker information into quantitative information that can be quantified and qualitative information that cannot be quantified, and uses the quantitative information in the first machine learning method determines the job suitable for the job seeker by the method, extracts feature words by data mining using qualitative information, calculates the job suitability of the job seeker by the second machine learning method using the extracted words, and the quantitative information a job search module that calculates suitability for each final job by using the job determined using , and the job fitness calculated using qualitative information; and a job matching module that determines a recommended job seeker and a recommended job by using the job search information for each job of the company and the job seeker's suitability for each final job.

And "user-customized job search mediation method (Publication No. 10-2020-0062797, Patent Document 3)" also exists.

In the case of the invention according to Patent Document 3, the step of storing the job offer information received from the job offer company terminal and the job search information received from the job seeker terminal in a database, respectively, a recommendation based on a matching score calculated through matching between the job information and the job search information Creating a list of job seekers, based on the job seeker location information obtained from the terminal of the job seeker who meets a predetermined criterion matching score among the job seekers listed in the list, and the location information of the job seeker obtained from the terminal of the job seeker to generate an interview / employment list Step and the step of providing interview/recruitment information to the job seeker terminal of at least one job seeker included in the interview/recruitment list, and the job offer information includes working period information, career information, salary information, gender information, age information, and appearance Information is included, and the matching score can be calculated by giving and processing different ranges of weights for working period information, career information, salary information, gender information, age information, and appearance information.

The above prior arts stop at mechanical matching of various keywords, and there is a problem in that the matching direction and matching rate fluctuate greatly depending on the keywords used in documents or words written from the point of view of a hiring company or job seeker.

Although matching corrections are made according to the application of weights, etc., this is only a correction of primary keyword matching due to the introduction of new secondary keywords, so the reliability of the matching recommendation is greatly increased from the position of professional freelancers or recruiting companies. There was a problem with falling.

Registration No. 10-1292336 Registration No. 10-1875923 Publication No. 10-2020-0062797

The human resource matching system through selective weighting of keyword matching according to the present invention has been devised to solve the conventional problems as described above, and presents the following problems to be solved.

First, it allows freelancers and companies to match each other's needs.

Second, the freelancers required by the company and the company's projects that can utilize the professionalism of the freelancers should be optimally connected.

The problems to be solved of the present invention are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

The human resource matching system through selective weighting of keyword matching according to the present invention has the following problem solving means for the above-mentioned problems.

In the case of a human resource matching system through selective weighting of keyword matching according to the present invention, a resume review unit that receives a resume uploaded through an expert terminal, reviews the resume, and extracts a specification keyword; a recruit review unit that receives a recruit document uploaded through a corporate terminal of a recruiting company, reviews the recruit document, and extracts a keyword of a need; a code matching unit for rearranging the predetermined order by matching the specification keyword and the needs keyword with each other, arranging them in a predetermined order, and assigning a matching rate weight according to a past actual matching rate; And it may be characterized in that it comprises a recommendation unit (recommend unit) that provides a mutual recommendation between the expert and the recruitment company according to the rearranged in the predetermined order.

In the case of the human resource matching system through selective weighting of keyword matching according to the present invention, the resume review unit may include: a resume receiving unit connected to the expert terminal and networking to receive the resume uploaded by the expert terminal; a keyword extraction unit that analyzes the received resume, and recognizes and extracts the specification keyword written by the expert in the resume; and a keyword ordering unit arranging the specification keywords according to a preset order.

In the case of the human resource matching system through selective weighting of keyword matching according to the present invention, the recruit review unit is networked with the corporate terminal of the hiring company, and receives the recruit document uploaded by the corporate terminal. ; a keyword extracting unit that analyzes the received request document and extracts the necessary keywords written by the recruiting company in the recruit document; and a keyword arranging unit for arranging the necessary keywords according to a predetermined order.

In the case of the human resource matching system through selective weighting of keyword matching according to the present invention, the code matching unit aligns the specification keywords according to the preset order, and aligns the need keywords according to the predetermined order. , a keyword matching unit for mutually matching the specification keyword and the need keyword according to the mutually sorted order; and a matching ordering unit for rearranging the specification keyword and the need keyword matched according to the keyword matching unit in the predetermined order.

In the case of a human resource matching system through selective weighting of keyword matching according to the present invention, the code matching unit is configured to perform mutual recommendation by the record unit according to the predetermined order rearranged by the matching ordering unit. It may further include a feedback reflecting unit that calculates a real matching rate between the expert and the recruiting company, and re-arranges the predetermined order according to the real matching rate.

In the case of the human resource matching system through selective weighting of keyword matching according to the present invention, the code matching unit includes a hierarchical decision list, and according to a predetermined decision list provided on the hierarchical decision list, the predetermined It may be characterized in that it further comprises a decision tree for rearranging the order of.

In the case of a human resource matching system through selective weighting of keyword matching according to the present invention, the decision tree unit includes a separate check list collected from the expert and a separate check list collected from the recruiting company in the hierarchical decision list It can be characterized by reflecting individually.

In the case of the human resource matching system through selective weighting of keyword matching according to the present invention, the code matching unit is a project setting unit that obtains project information obtained from the job offer company and has a predetermined period and a predetermined mission ; and matching the keywords of the needs of the recruiting company that transmitted the project information and the specification keywords of the experts, and providing them to the decision tree part through the hierarchical decision list of the decision tree part By rearranging the order, it may further include a project recommendation unit for selectively recommending the expert to the project information of the recruiting company.

In the case of the human resource matching system through selective weighting of keyword matching according to the present invention, the project recommendation unit may include: a stream listing unit for arranging the project information as a bit stream of bit units and transmitting it to the expert terminal; a bit omission unit configured to set a predetermined section in the bit stream and selectively omit the predetermined section to transmit the bit stream; and a parallel transmission unit that independently processes missing bit information of the predetermined section selectively omitted from the bit stream.

In the case of the human resource matching system through selective weighting of keyword matching according to the present invention, the record unit selectively transmits the missing bit information separately transmitted by the parallel transmitter when the ID of the expert terminal is confirmed, After the missing bit information is transmitted, the project information is recovered by merging the missing bit information in the predetermined section of the bit stream.

The human resource matching system through selective weighting of keyword matching according to the present invention configured as described above provides the following effects.

First, primary filtering is performed by matching key keywords between Rejume and Recruit documents.

Second, the mutual matching combinations filtered through main keyword matching are given weights according to the actual matching rate, and are filtered secondary accordingly.

Third, it finds the optimal combination for the second matched combination filtered according to the decision tree, and combines the professionalism of freelancers with the needs of the hiring company to match the desired talent to the desired place.

Effects of the present invention are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a conceptual diagram of a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.
FIG. 2 is a conceptual diagram illustrating mutual keyword matching between an expert's rejume and a recruiting company's recruit document in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.
3 is a block diagram illustrating the main components of a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.
4 is a block diagram illustrating the configuration of a rejume review unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.
5 is a block diagram illustrating a configuration of a recruit review unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.
6 is a block diagram illustrating each configuration of a code matching unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.
7 is a conceptual diagram illustrating a decision tree unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.
8 is a block diagram illustrating a project recommendation unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.
9 is a conceptual diagram illustrating the function of each configuration of the project recommendation unit in the human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.

The human resource matching system through selective weighting of keyword matching according to the present invention can apply various changes and can have various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include all modifications, equivalents and substitutes included in the spirit and scope of the present invention.

1 is a conceptual diagram of a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention. FIG. 2 is a conceptual diagram illustrating mutual keyword matching between an expert's rejume and a recruiting company's recruit document in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention. 3 is a block diagram illustrating the main components of a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention. 4 is a block diagram illustrating the configuration of a rejume review unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention. 5 is a block diagram illustrating a configuration of a recruit review unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention. 6 is a block diagram illustrating each configuration of a code matching unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention. 7 is a conceptual diagram illustrating a decision tree unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention. 8 is a block diagram illustrating a project recommendation unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention. 9 is a conceptual diagram illustrating the function of each configuration of a project recommendation unit in a human resource matching system through selective weighting of keyword matching according to an embodiment of the present invention.

In the case of a human resource matching system through selective weighting of keyword matching according to the present invention, as shown in FIG. 1, experts, for example, specialized knowledge and accumulated know-how in a specific field, and practical experience in the field are supported. Disclosed is a technical idea that allows matching between freelancers and recruiting companies (2) that require the expertise of such freelancers.

As shown in FIG. 2, this mechanism is basically made through mutual keyword matching between the resume 11 written by the expert 1 and the recruit document 21 written by the job seeker 2, but this matching is conventional Let's focus on the part that is very different from the technique of

3, in the case of a human resource matching system through selective weighting of keyword matching according to the present invention, a resume review unit 110, a recruit review unit 120, a code It may include a matching unit (code matching unit, 130) and a recommendation unit (recommend unit, 140).

First, as described above, the expert (1) mainly means a freelancer supported by professional knowledge, accumulated know-how, and field experience in a specific specialized field, which is a specific major or a specific field of expertise. It may be a craftsman or a person who has accumulated knowledge in a specific field, or a professor in a major field, a researcher in the major field, or other natural person with business or practical knowledge.

As described above, the recruiting company (2) means a company that needs the above expert (1), and the hiring company (2) performs a specific mission with the above expert (1) as a regular or contract worker or outsourced person. It could be a company that wants you to do it.

The professional terminal 10 or the corporate terminal (not shown) is a terminal such as a general desktop PC, a laptop PC, a tablet PC, or a smart phone, and the type thereof is not limited. It is sufficient if the expert terminal 10 or the corporate terminal can transmit and receive target data through the system and data networking mentioned in the present invention.

In the case of the resume review unit 110 , it is configured to receive the uploaded recipe 11 through the expert terminal 10 .

Resume 11 does not mean a tangible, that is, a document in the form of paper or film, but refers to a digital document resume.

Resume 11 may be implemented in PDF, MS WORD, or Hangul (HWP), and the electronic form is not limited. However, it goes without saying that the resume 11 should contain the ID (identity) and specification information of the expert 1 .

The resume review unit 110 extracts the specification keyword of the specification information from the resume 11 written and written by the expert 1 .

The specification information may include the expert 1's ID (signature, age, gender, physical information, etc.) as well as the expert's academic background, career, and self-introduction.

In the case of the recruit review unit 120, it receives a recruit document (21) uploaded from the corporate terminal (not shown) of the recruiting company (2), and the company's job search needs keyword ( needs keyword) is extracted.

Recruitment document (21) does not limit the format, and corresponds to an electronic digital document that includes requirements, job announcements, job types, age groups, and other requirement information that companies offer for job openings.

In the case of the code matching unit 130 , the predetermined order may be rearranged by matching the specification keyword and the needs keyword with each other, arranging them in a predetermined order, and assigning a matching rate weight according to the past actual matching rate.

As described above, the code matching unit 130 mutually matches the specification keyword extracted from the recipe 11 of the expert 1 and the required keyword from the roasting company 2, but in a predetermined order, matching of these matching keywords It can be arbitrarily selected according to the number and accuracy of matching. Most of the conventional patents or practical technologies are based on adding weights by introducing other keywords when they stop here or go one step further.

Also in the present invention, a matching rate weight is given according to the mutual matching rate or the number of matches between these keywords, and after giving these weights, a predetermined order of the matching rate is rearranged.

In the case of the record unit 140, the recommendation is made between the expert (1) and the recruiting company (2) according to the rearranged predetermined order. The recommendation means that the expert (1) recommends a job offer company suitable for the expert terminal 10 to the expert (1), and recommends the expert (2) suitable for the company to the job offer company (2).

In the case of the resume review unit 110 , as shown in FIG. 4 , it may include a recipe receiving unit 111 , a keyword extracting unit 112 , and a keyword ordering unit 113 .

First, in the case of the recipe receiving unit 111 , it is connected to the expert terminal 10 by networking to receive the recipe 11 uploaded by the expert terminal 10 .

In the case of the keyword extraction unit 112, as described above, the proposal receiving unit 111 analyzes the received proposal 11, and recognizes the specification keyword written by the expert 1 in the proposal 11, and It is an extracting composition.

The keyword extraction unit 112 extracts the specification information of the expert 1, ie, the specification keyword, by using the OCR method or a commercialized electronic keyword extraction method.

In the case of the keyword ordering unit 113, the specification keywords are arranged in a preset order.

The preset order is based on the industry, business type, and other criteria among the keywords indicating the highest frequency among the specification keywords, words indicating major, words indicating a specific position or number of years in career, and words indicating a specific company during career. Of course, it may be set differently.

The request review unit 120 may include a request receiving unit 121 , a keyword extracting unit 122 , and a keyword arranging unit 123 .

In the case of the request receiving unit 121, it is connected to a corporate terminal (not shown) of the job offer company 2 by networking to receive the request document 21 uploaded by the corporate terminal through the network.

In the case of the keyword extracting unit 122 , the request receiving unit 121 analyzes the received request document 21 and extracts the required keyword written by the job offer company 2 in the request document 21 .

The keyword extraction unit 122 also extracts keywords from the request document by using the OCR method or a commercialized electronic keyword extraction method.

In the case of the keyword arranging unit 123, it is a configuration in which needs keywords are arranged in a predetermined order.

In the case of the predetermined order, it can be set differently according to the order in which the frequency appears high among the keywords of the need, or the required major keyword, the required career keyword, age, gender, industry, and also keywords representing the required career.

The matching code unit 130 may include a keyword matching unit 131 and a matching ordering unit 132 as shown in FIG. 6 .

First, in the case of the keyword matching unit 131, the specification keywords are sorted according to a preset order, and the needs keywords are sorted according to the preset order. Thereafter, the keyword matching unit 131 is configured to mutually match the specification keyword and the need keyword according to the mutually arranged order.

In addition, in the case of the mapping ordering unit 132, the specification keywords and the needs keywords matched according to the keyword matching unit 131 are rearranged in a predetermined order.

Here, the predetermined order may be a form in which a specification keyword and a need keyword are arranged in order, and then a matching order that is matched to the order (importance) is arranged, and the keywords are mutually matched according to the frequency with which they are frequently matched. You can also decide the order.

However, as described above, the predetermined order is i) the order of high frequency among the specification keywords, or a word indicating a major, a word indicating a specific position or number of years in a career, a word representing a specific company in a career, etc. Among the industry, business type, and other criteria, ii) among the keywords required, the order in which they appeared with the highest frequency, or the required major keyword, the required career keyword, age, gender, industry, and, in addition, set differently according to keywords that represent the required career can be

More preferably, the code matching unit 130 may further include a feedback reflecting unit 133 as shown in FIG. 6 .

In the case of the feedback reflection unit 133, after the record unit 140 mutually recommends them according to a predetermined order rearranged by the matching ordering unit 132, the expert 1 and the recruiting company 2 actually interact with each other. The matching rate is calculated, and a predetermined order is re-arranged according to the actual matching rate. This repetitive feedback is performed every time, reflected in the keyword analysis and matching of the next expert (1) and the recruiting company (2), and continuously performed, and computational deep learning to increase the accuracy. do.

As shown in FIG. 6 , the code matching unit 130 may further include a decision tree unit 135 .

In the case of the decision tree unit 135, as shown in FIG. 7 , a hierarchical decision list (refer to tree 1 - tree n) is provided, and a previously presented decision list provided on the hierarchical decision list (decision 1 - tree n) is provided. According to decision n), the predetermined order is to be rearranged.

The hierarchical decision list is a variable data area, which can be arbitrarily created, changed, and modified.

As shown in FIG. 7 , these hierarchical decision lists go through the sequential list from top to bottom. As described above, each of these hierarchical decision lists is selectively provided with decision lists.

In the case of decision lists, as a question about the requirements that require a final check from the expert (1) or the hiring company (2), more preferably, the hiring company (2) is a consumer for a specific job or professional service, and Since the information on the work to be done is provided more and more accurately, it is preferable to correspond to the check requirement obtained from the recruiting company (2).

In the case of the decision list, it may be determined differently depending on the industry and business type of the recruiting company (2) and the service to be performed or the professional knowledge required. A check on whether a specific job has been performed in the past, a personality question required by a recruiting company, or other professional service experience may be detailed.

That is, in the case of the decision tree unit 135, the separate check list collected from the expert 1 and the separate check list collected from the recruiting company 2 can be individually reflected in the hierarchical decision list.

In the case of the code matching unit 130 , as shown in FIG. 6 , it may include a project setting unit 136 and a project recommendation unit 137 .

The project setting unit 136 is configured to obtain project information that is obtained from the recruiting company 2 and has a predetermined period and a predetermined mission.

The predetermined period may mean a period in which a predetermined mission requested by the recruiting company 2 is performed, a period in which a predetermined mission must be completed, or a contract period with the expert 1 .

The predetermined mission may be a specific task requested by the recruiting company 2, a specific performance, a specific action, a specific job, a program that requires submission after completion, an object to be provided after completion, and the like.

In the case of the project recommendation unit 137, the needs keywords of the recruiting company that have transmitted the project information and the expert's specification keywords are matched and ordered, and the hierarchy of the decision tree unit 135 is provided to the decision tree unit 135. It is a configuration for selectively recommending the expert (1) to the project information of the recruiting company (2) by rearranging a predetermined order through the appropriate decision list.

In the case of the project recommendation unit 137, as shown in FIG. 8 , it may include a stream arranging unit 137-1, a bit correcting unit 137-2, and a parallel transmission unit 137-3.

As shown in Fig. 9(a), in the case of the stream arranging unit 137-1, the project information as described above is arranged as a bit stream (data stream) in units of bits, so that the expert terminal (10) is transmitted.

In the case of the bit omitting unit, predetermined sections (refer to omitted codes) are set in the bit stream, and the bit stream is transmitted by selectively omitting the predetermined sections.

In addition, in the case of the parallel transmission unit 137-3, it is a configuration that transmits the omitted bit information (omitted codes) of a predetermined section (refer to the committed codes) selectively omitted from the bit stream independently of the non-omitted bit stream. .

Accordingly, in the case of the record unit 140, the parallel transmission unit 137-3 selectively transmits the missing bit information separately transmitted when the ID of the expert terminal 10 is confirmed as the expert 1 of the allowed identity. do.

Thereafter, in the case of the record unit 140, after the missing bit information (refer to the omitted codes) is transmitted, the missing bit information of a predetermined section of the bit stream is merged and the project information is recovered, that is, restored.

The scope of the present invention is determined by the matters described in the claims, and parentheses used in the claims are not described for selective limitation, but are used for clear components, and descriptions in parentheses are also interpreted as essential components. should be

1: Professional (Freelancer) 2: Hiring Companies
10: expert terminal 11: resume
21: Recruit Document 100: System
110: Resume review unit 111: Resume receiving unit
112: keyword extraction unit 113: keyword ordering unit
120: Recruit review unit 121: Recruit receiving unit
122: keyword extraction unit 123: keyword arranging unit
130: code matching unit 131: keyword matching unit
132: matching ordering unit 133: feedback reflecting unit
135: decision tree unit 136: project setting unit
137: project recommendation section 137-1: stream listing section
137-2: bit omission unit 137-3: parallel transmission unit
140: record unit

Claims (10)

a resume review unit that receives a resume uploaded through an expert terminal, reviews the resume, and extracts a specification keyword;
a recruit review unit that receives a recruit document uploaded through a corporate terminal of a recruiting company, reviews the recruit document, and extracts a keyword of a need;
a code matching unit for rearranging the predetermined order by matching the specification keyword and the needs keyword with each other, arranging them in a predetermined order, and assigning a matching rate weight according to a past actual matching rate; and
Human resources matching system through selective weighting of keyword matching, characterized in that it comprises a recommendation unit (recommend unit) that provides a mutual recommendation between the expert and the recruitment company according to the rearranged predetermined order.
According to claim 1, wherein the rejume review unit,
a resume receiving unit connected to the expert terminal through networking to receive the resume uploaded by the expert terminal;
a keyword extraction unit that analyzes the received resume, and recognizes and extracts the specification keyword written by the expert in the resume; and
Human resources matching system through selective weighting of keyword matching, characterized in that it comprises a keyword ordering unit for arranging the specification keywords in a preset order.
The method of claim 2, wherein the recruit review unit,
a request receiving unit that is networked with the corporate terminal of the recruiting company and receives the request document uploaded by the corporate terminal;
a keyword extracting unit that analyzes the received request document and extracts the necessary keywords written by the recruiting company in the recruit document; and
Human resource matching system through selective weighting of keyword matching, characterized in that it comprises a keyword arranging unit for arranging the need keywords in a predetermined order.
The method of claim 3, wherein the code matching unit comprises:
a keyword matching unit for arranging the specification keywords according to the preset order, aligning the need keywords according to the predetermined order, and matching the specification keywords with the need keywords according to the mutually sorted order; and
Human resources matching system through selective weighting of keyword matching, characterized in that it comprises a matching ordering unit for rearranging the specification keyword and the need keyword matched according to the keyword matching unit in the predetermined order.
The method of claim 4, wherein the code matching unit comprises:
After the record unit performs mutual recommendation in accordance with the predetermined order rearranged by the matching ordering unit, the real matching rate is calculated between the expert and the recruiting company,
Human resource matching system through selective weighting of keyword matching, characterized in that it further comprises a feedback reflecting unit for re-arranging the predetermined order according to the matching rate weight according to the actual matching rate.
The method of claim 5, wherein the code matching unit comprises:
Humans through selective weighting of keyword matching, characterized in that it further comprises a decision tree unit that includes a hierarchical decision list and rearranges the predetermined order according to a decision list previously presented on the hierarchical decision list. resource matching system.
The method of claim 6, wherein the decision tree unit,
A human resource matching system through selective weighting of keyword matching, characterized in that the separate check list collected from the expert and the separate check list collected from the recruitment company are individually reflected in the hierarchical decision list.
The method of claim 7, wherein the code matching unit comprises:
a project setting unit for obtaining project information obtained from the recruitment company and having a predetermined period and a predetermined mission; and
The required keyword of the recruiting company that has transmitted the project information and the specification keyword of the expert are matched to each other and ordered, provided to the decision tree, through the hierarchical decision list of the decision tree Human resource matching system through selective weighting of keyword matching, characterized in that it further comprises a project recommendation unit for selectively recommending the expert to the project information of the recruiting company by rearranging the order.
The method of claim 8, wherein the project recommendation unit,
a stream arranging unit for arranging the project information in a bit stream of a bit unit and transmitting it to the expert terminal;
a bit omission unit configured to set a predetermined section in the bit stream and selectively omit the predetermined section to transmit the bit stream; and
Human resource matching system through selective weighting of keyword matching, characterized in that it comprises a parallel transmission unit for independently processing the missing bit information of the predetermined section selectively omitted from the bit stream.
The method of claim 9, wherein the record unit,
When the ID of the expert terminal is confirmed, selectively transmitting the missing bit information separately transmitted by the parallel transmission unit,
Human resources matching system through selective weighting of keyword matching, characterized in that after the missing bit information is transmitted, the project information is recovered by merging the missing bit information in the predetermined section of the bit stream .

Images