KR20010085051A - Resume searching method and apparatus for internet recruitment system - Google Patents

Abstract

PURPOSE: A resume searching system and method is provided to display a similarity rate on a searched resume result such as a value from 0 to 1 according to a similarity degree between a search word and data included in the resume so that it can enhance search efficiency. CONSTITUTION: The method comprises the steps of a job seeker writing in a resume form including key words and a job and worker seeking system storing the resume data at a database, the job and worker seeking system collecting the key words, the system calculating a numerical value on a similarity among the key words, the system adjusting similar key words, the system constructing a similarity matrix database, a company member inputting a search word in a search engine window(101), the search engine referring a similarity between the search word and the stored key word to the similarity matrix database(103), the search engine calculating a relation degree of a resume from a similarity between the key word and the search word(105), the search engine comparing the relation degree with a set value(107) and the search engine outputting the resumes having a relation value more than the set value(109).

Description

Resume searching method and apparatus for internet recruitment system

The present invention relates to an internet online job system. More specifically, the present invention relates to a method and apparatus for more accurately finding talent in corporate job search for a resume in an Internet job system. The method of searching for a resume and the fitness of the resume with mathematical techniques The present invention relates to a method and an apparatus for calculating and allowing a corporate member to know which of the resumes is most suitable.

Recently, with the spread of the Internet, various information is provided or services which are conventionally performed offline are made online. These Internet services utilize the latest rapidly developed communication protocols, image processing technology, software, multimedia, and multitasking technology, and do not require the conventional wide store, space, and management personnel. In addition, it is widely used in the fields of education, advertising, and learning.

The Internet is a collection of computer networks using the TCP / IP protocol, which is usually operated by a client / web server architecture using the HTML (Hypertext) language. The client and web server communicate with each other using the functionality provided by HTTP (hypertext transfer protocol).

The web server stores web pages made of HTML and receives data from various content providers. The client can view the web page through the URL assigned to the web page of the web server through the web browser. The remote server can use the contents provided by the web server through the Internet.

By using the Internet, a plurality of clients can use the contents of the web server at the same time, thereby easily performing services such as job posting and job search. In the conventional online job system using the Internet, a corporate member registered as a member on the service site is made by selecting the required personnel by searching the resume of the job seeker.

However, conventionally, the retrieval of a resume merely means that the resume is electronically documented, and thus, more than a service of simply comparing the contents thereof is not provided. In other words, if you perform the search in accordance with the search word entered into the search item with the power of the computer for the inconvenience of the search in the off-line, and output the corresponding resume, a person examines the resumes and compares them. In the related art, furthermore, no search is performed on a resume that is not identical to the search word. In addition, if there are many search results, it does not show which resume matches what you want to find most, but it is output only in random or in order by resume creation date.

For example, if a corporate member wants to search for job applicants that match the search term "Seoul, 25-year-old, male, four-year college graduate, Java programmer," the existing method will only print resumes that match these terms exactly. Therefore, there are many cases where there is no search result, and in fact, a resume having similar conditions may be excluded from the search. In other words, the same condition occurs, but the search is not possible because the region is the Gyeonggi region, and there is a hassle that the corporate member has to modify the search word several times in order to overcome this problem.

The present invention has been improved to solve the problems of the conventional search method in the job system. The object of the present invention is to introduce a concept of "fit" to a search system, to give a goodness of fit for a result that exactly matches a search term, and to 0 to 1 depending on the degree of flexibility in the case of not matching the search term. It is characterized by providing a compatibility between. Accordingly, the present invention provides a method and apparatus for retrieving a resume that enables retrieval of a resume having a similar condition as well as a resume completely matching the condition to be searched.

In order to achieve the above object, according to the present invention, a similarity matrix (similarity matrix) that can measure the similarity between search keywords is prepared in advance. The resume is retrieved from the resume DB by the keyword (hereinafter referred to as a search keyword) entered by the corporate member through the search engine to search for the resume. At this time, unlike the conventional method, the resume is not searched for only the resume matching the search keyword. The similarity between the keyword (resume keyword) and the search keyword recorded in the resume stored in the DB is extracted from the similarity table. For example, the similarity between Seoul and Gyeonggi-do will be shown as 0.9, and this will be included in the search results. The preparation of the similarity table can be accomplished by collecting the search keyword combinations of corporate members, quantifying the similarities between them and adjusting the managers.

1 is a process flow diagram of a method according to the invention.

2 is a functional block diagram of an apparatus according to the present invention.

3 is an exemplary view for each search item of a search keyword and a resume keyword.

4 illustrates two examples of affinity table.

5 is an exemplary view of a fitness calculation method.

<Overview of invention>

Definition of Terms

Corporate member: A company that wants to join as a member in the Internet job system according to the present invention. Job applicants are not necessarily limited to companies using this system, but the term "company member" is used in this specification for convenience.

Search keyword: Keyword entered by the corporate member to search for a resume with the desired conditions.

Keyword variations: Keywords that are not the same as the search keyword entered by the company member. Similarity values are numbered according to the degree of similarity with the search keywords and stored in the DB as a similarity matrix.

Resume Keywords: Keywords that are included in the resume by a person seeking a job.

Similarity figures: The degree of similarity between search keywords and similar keywords.

Composition of the Invention

In accordance with the present invention, the resume search method on the Internet job search system basically searches for a desired resume from the resume DB based on a resume DB for storing a resume created by a job seeker including a resume keyword and a search keyword inputted by a company member. This is done in an internet job search system that includes a search engine.

The method of the present invention comprises the steps of creating a similarity table representing the similarity between the search keyword and the similar keyword and storing it in the similarity table DB; When a corporate member inputs a search keyword into a search engine, retrieving a similarity relationship between the resume keyword included in each resume stored in the resume DB and the corresponding search keyword from the similarity table DB; Calculating a relevance of the resume from a similar relationship between the search keyword inputted by the corporate member and the resume keyword included in the resume; And outputting a resume corresponding to the case where the calculated goodness of fit is equal to or greater than a predetermined set value.

On the other hand, the resume search apparatus according to the present invention relates to a system for corporate members to access the job system through the Internet to search for the desired resume with a search keyword,

A resume DB for storing a resume created by a job seeker including a keyword (hereinafter, referred to as a resume keyword); A similarity table generating means for creating a similarity table representing a similarity between the search keyword and the similar keyword; A search information management system including a similarity table DB for storing the similarity tables;

A similarity table query means for inquiring a similarity relationship between a resume keyword included in each resume stored in the resume DB and the corresponding search keyword when the corporate member inputs a search keyword into a search engine; Goodness-of-fit calculation means for calculating a goodness of fit of the resume from a similar relationship between the search keyword inputted by the corporate member and the resume keyword included in the resume; And a search engine including search result output means for outputting a resume corresponding to a case where the calculated goodness of fit is equal to or greater than a predetermined set value.

In the method and apparatus of the present invention, the similarity table is assigned a similarity value according to the degree of similarity between the search keyword and the similar keyword, and the resume suitability is based on the average value of the similarity values assigned to the resume keywords included in the resume. It is characterized in that it is calculated.

<Description of Example>

Hereinafter, specific embodiments of the present invention will be described with reference to the accompanying drawings.

1 is a process flowchart of a resume searching method in the job system according to the present invention. In FIG. 1, before the process related to the method of the present invention proceeds, a job searcher stores a resume created by including a resume keyword in a resume DB 13, and a manager of the job system collects a combination of search keywords and The step of quantifying the similarity between the two and adjusting the similar keyword to create a similarity table and storing it in the similarity table DB 11 is preceded.

After the similarity table DB 11 and the resume DB 13 are prepared, when a corporate member inputs a search keyword into a search engine [101], resume keywords included in each resume stored in the resume DB 13 and Retrieving a similarity relationship with the corresponding search keyword from the similarity table DB 11 [103]; Calculating a fitness of the resume from the similarity between the search keyword inputted by the corporate member and the resume keyword included in the resume [105]; Comparing the calculated goodness of fit with a predetermined set value [107], if the goodness of fit is equal to or greater than the set value, the step of outputting a resume corresponding thereto is followed.

Meanwhile, the process flow as shown in FIG. 1 is performed by the device as shown in FIG. 2 is a functional block diagram of a resume retrieval apparatus according to the present invention, and is largely comprised of a retrieval information management system and a retrieval engine.

The search information management system includes a resume DB (13) for storing a resume created by a job seeker including a keyword (hereinafter, referred to as a resume keyword); A similarity table generation unit 15 for creating a similarity table indicating a similarity relationship between the search keyword and the similar keyword by the manager 10 of the job system; It is configured to include a similarity table DB (11) for storing a similarity table,

When the search engine enters a search keyword in the search engine, the search engine searches the similarity table DB 11 for the similarity between the resume keyword included in each resume stored in the resume DB 13 and the corresponding search keyword. A similarity table inquiry unit 17; A fitness calculation unit 19 for calculating a fitness of the resume from a similar relationship between the search keyword inputted by the corporate member 23 and the resume keyword included in the resume; And a search result output unit 21 for outputting a resume corresponding to the case where the calculated goodness of fit is equal to or greater than a predetermined set value.

The operation of the apparatus shown in FIG. 2 is accomplished by the process flow diagram shown in FIG. Hereinafter, each step will be described in more detail.

FIG. 3 shows search keywords (a) of FIG. 3 and resume keywords (refer to FIG. 3) recorded in the resume DB for each search word item that can be included in the resume. In FIG. 3, as a search item, region, age, gender, specialty, career, salary, and final education level are taken as an example. According to such a search item, the corporate member defines an object to search for and inputs a search keyword to compare with the resume keywords recorded in the resume DB. Conventionally, since the resume is output only when the search keyword and the resume keyword match, the search result does not come out if any one of the search items does not match. In (b) of FIG. 3, it can be seen that various resume keywords exist. In fact, the similarities in Seoul / Gyeonggi area or age 25/24 or annual salary of 20 million won / 18 million won in FIG. 3 (b) may be considered to be consistent in terms of job placement. However, conventionally, even in this case, the search result was excluded from the search object because of inconsistency.

4 shows a regional similarity table and an age similarity table as examples of a similarity matrix that is the core of the present invention. All of the similarity tables can be defined for each search item, and there can be two types of similarity tables according to the characteristics of the values included in the search items. As shown in (a) of FIG. 4, when the value to be entered in the search item is a nominal scale, it is composed of n × n square matrix, and when the value to be entered in the search item is a ratio scale, the difference between the search item and the resume information is expressed as a percentage. Can be expressed as n × 1 matrices. For example, in FIG. 4B, when the age is searched for 25 years old, if the value of the resume keyword is 20 years old, the difference is -20%, and the corresponding similarity becomes 0.4.

5 shows an example of the step [105] of calculating the relevance of the resume during the retrieval process. After retrieving the similarity by comparing the values of the search keyword and the resume keyword for each search item in the similarity table, that is, extracting the similarity values [103], the arithmetic mean is calculated by calculating the arithmetic suitability. In FIG. 5, the items matching the search keyword and the resume keyword are gender, specialty, and final education. The similarity value is 1, and the similarity value of Seoul / match is 0.9, the similarity value of age is 0.8, the similarity value of career is 0.9, and the annual salary is 0.9. Since the similarity value of is 0.9, the arithmetic mean of all the similarity values divided by the number of items is "0.93".

Thus, if the fitness is calculated [105] if the resume having a resume keyword with a similarity value smaller than the threshold set by the corporate member is output without outputting a resume equal to or greater than the set value [109] according to the present invention. Resume retrieval method is made. When printing resumes, printing resumes from the highest relevance (that is, resumes with the highest arithmetic mean of similarity values) to the relevance in descending order will provide a more convenient search method for corporate members.

One embodiment in which the technical spirit of the present invention has been described above has been described, but it is obvious that the technical idea of the present invention does not correspond to the above embodiment. The technical spirit of the present invention is not limited to the above embodiments, but is determined by the technical scope determined by the appended claims.

According to the present invention, the similarity between the resume information (resume keyword) and the search keyword is obtained by using the similarity table, and the enterprise member can search several times by outputting the results that are not exactly matched with the appropriateness of the search. There is no need and the search results are listed in order of relevance, which has the advantage of knowing which resume the company member is looking for most.

Claims (10)

Internet job search system that includes a resume DB that stores a resume created by job seekers including keywords (hereafter, resume keywords) and a search engine that searches for the desired resume from the resume DB by inputting desired keywords (hereafter, search keywords). As a resume search method in Creating a similarity table representing the similarity between the search keyword and the similar keyword and storing the similarity table in the similarity table DB; When a corporate member inputs a search keyword in a search engine, retrieving a similar relationship between a resume keyword included in each resume stored in the resume DB and the corresponding search keyword from the similarity table DB, Calculating a relevance of the resume from the similarity between the search keyword entered by the company member and the resume keyword included in the resume; And outputting a resume corresponding to the case where the calculated goodness of fit is equal to or greater than a predetermined set value. The method of claim 1, wherein the similarity table is assigned a numerical value (hereinafter, similarity value) according to the similarity between the search keyword and the similar keyword. The method of claim 2, wherein the similarity value is in the range of 0 to 1, wherein "1" is assigned to the same keyword as the search keyword, and "0" is assigned to the keyword having the closest relation with the search keyword. Resume retrieval method on Internet job offer system to do. The method of claim 2, wherein the resume suitability is an average value of similarity values assigned to resume keywords included in the resume. The method of claim 1, wherein in the resume outputting step, the resume is output from the highest suitability resume to the lowest suitability order. This is a system where corporate members access the job search system through the Internet and search for the desired resume with search keywords. A resume DB for storing a resume created by a job seeker including a keyword (hereinafter, referred to as a resume keyword); A similarity table generating means for creating a similarity table representing a similarity between the search keyword and the similar keyword; A search information management system including a similarity table DB for storing the similarity tables; A similarity table query means for inquiring a similarity relationship between a resume keyword included in each resume stored in the resume DB and the corresponding search keyword when the corporate member inputs a search keyword into a search engine; Goodness-of-fit calculation means for calculating a goodness of fit of the resume from a similar relationship between the search keyword inputted by the corporate member and the resume keyword included in the resume; A search engine including search result output means for outputting a resume corresponding to a case where the calculated goodness of fit is equal to or greater than a predetermined set value; A resume retrieval apparatus on an internet job system. The apparatus of claim 6, wherein the similarity table is assigned a numerical value (hereinafter, similarity value) according to the similarity between the search keyword and the similar keyword. The method of claim 7, wherein the similarity value is in a range of 0 to 1, wherein “1” is assigned to the same keyword as the search keyword, and “0” is assigned to the keyword having the closest relation with the search keyword. Resume retrieval apparatus on the Internet job system which do. The apparatus of claim 6, wherein the fitness calculation unit calculates the fitness by calculating an average value of similarity values assigned to resume keywords included in the resume. The resume search apparatus according to claim 6, wherein the search result output means outputs the resume from the highest suitability resume to the lowest suitability order.