KR20210133899A - Apparatus and method for recommending job - Google Patents

Abstract

The present invention relates to a method for recommending a customized appointment and a method for recommending a military service experience-based job. More specifically, the present invention relates to the method for recommending a military position or recommending a civilian job, a qualification, or an education to a person with military service experience using a filter generated based on a dictionary comprising a plurality of job competency units. According to an embodiment of the present invention, the military position to a target may be recommended or a job suitable for the person with military service experience may be recommended based on the filter prepared based on an objective, reasonable, and consistent criteria. The method comprises a step of recommending a job to the target based on the competency-related data and filter of the target by a job recommendation part.

Description

Military service experience-based job recommendation device and method, customized position recommendation device {APPARATUS AND METHOD FOR RECOMMENDING JOB}

The present invention relates to a customized position recommendation method and a military service experience-based job recommendation method, and more particularly, to recommend a military position using a filter generated based on a dictionary including a plurality of job competency units, or to a person with military service experience How to recommend private jobs, certifications, or training to people.

Among young soldiers who are discharged from military service, those who do not return to college go on to enter society without any preparation for employment. According to the current situation, even if military veterans acquire enough experience necessary in society through military life, they are not properly evaluated in the civilian society, nor are they recognized as careers. Even though they studied the same field of study at university and worked as an operational officer, personnel officer, and non-commissioned officer for many years as a professional soldier in the military, they are not recognized as the same professional activity in society, nor are they used as a social career. Of course, the job performance of an enlisted soldier who has completed his compulsory service is also not evaluated as a vocational activity in society.

Officers who performed vocational activities as professional soldiers in the military perform their duties for quite a long time based on their job abilities or specialties. Therefore, they have considerable expertise in their field of work. Nevertheless, these capabilities are not circulated with society. This is because military service experience is not validly recognized as a vocational activity in relation to social experience.

Republic of Korea Patent Publication: KR 2018-0041478 A (published on April 24, 2018)

An object of the present invention is to provide a job recommendation device, a job recommendation method, and a recording medium for recommending a civilian job to a person with military service experience using a filter generated based on a dictionary including a plurality of job competency units.

Another object of the present invention is to provide a job recommendation method capable of recommending a military position based on a filter prepared on an objective, reasonable and consistent basis, or recommending a job suitable for a person with military service experience.

In addition, the present invention is to provide a job recommendation method that can improve job competency during military service life as it is possible to recommend a customized position based on the subject's major, social experience, qualifications, etc. prior to military service.

The problems to be solved by the present invention are not limited to the problems mentioned above. Other technical problems not mentioned will be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description.

The present invention provides a job recommendation method. In an embodiment, the job recommendation method includes recommending a job to the target subject based on the target's competency-related data and a filter by the job recommendation unit, wherein the filter is data related to the required ability according to the job It is generated based on a dictionary including a plurality of job competency units composed of Including at least one, the job competency units include a plurality of military position units composed of data related to the necessary capabilities according to the military position, and the subject's competency-related data includes the subject's military service career.

In an embodiment, in the filter, a word extracted from data constituting the job competency units in the first unit of any one of a row unit and a column unit, and the job competency units are arranged in the other second unit. may include a two-dimensional matrix in which they are arranged, and the two-dimensional matrix may have a degree of association between each job competency unit of the first unit and each word of the second unit as each component value.

In one embodiment, the filter includes a two-dimensional matrix in which job competency units are arranged in a first unit of any one of a row unit and a column unit, and the job competency units are arranged in another second unit, , the two-dimensional matrix has a degree of relevance between a first job capability unit associated with each job capability unit of the first unit and a second job capability unit associated with each job capability unit of the second unit as each component value. can

In an embodiment, the step of recommending the job may include: extracting words related to the military service career of the subject from the competency-related data; and determining one or more job competency units related to the subject from the extracted words. and determining the one or more job competency units as recommended jobs.

In an embodiment, using the one or more job competency units as a first job competency unit, based on the filter and based on the similarity with the first job competency unit, at least one second job among a plurality of job competency units It may include the step of determining the competency unit as the recommended job.

The present invention includes a method for recommending a military position. In an embodiment, the method for recommending a position in the military position includes the step of recommending a position to the subject based on the subject's competency-related data and a filter by the job recommendation unit, comprising data related to the required ability according to the job Doedoe generated based on a dictionary including a plurality of job competency units, at least one of a degree of relevance between the job competency units and words extracted from data constituting the job competency units and a degree of relevance between the job competency units Including, wherein the job competency units include a plurality of military position units consisting of data related to the required competency according to the military position, and the subject's competency-related data includes the subject's education information, holding qualification information, portfolio, resume, Self-introduction, SNS data, data collected by a chatbot that collects history information in the form of conversation with the subject, social relationship information of the subject, interests on SNS of the subject, content viewing record of the subject, specific of the subject and at least one of information retrieval history.

In an embodiment, in the filter, a word extracted from data constituting the job competency units in the first unit of any one of a row unit and a column unit, and the job competency units are arranged in the other second unit. may include a two-dimensional matrix in which they are arranged, and the two-dimensional matrix may have a degree of association between each job competency unit of the first unit and each word of the second unit as each component value.

In one embodiment, the filter includes a two-dimensional matrix in which job competency units are arranged in a first unit of any one of a row unit and a column unit, and the job competency units are arranged in another second unit, , the two-dimensional matrix has a degree of relevance between a first job capability unit associated with each job capability unit of the first unit and a second job capability unit associated with each job capability unit of the second unit as each component value. can

In an embodiment, the step of recommending the job may include: extracting words related to the job ability of the subject from the competency related data; and determining one or more military positions related to the subject from the extracted words; and determining the one or more military positions as recommended military positions.

According to an embodiment of the present invention, a job recommendation method for recommending a military position or recommending a job to a person with military service experience is provided using a filter generated based on a dictionary including a plurality of job competency units.

In addition, according to an embodiment of the present invention, it is possible to recommend a military position to a target or a job suitable for a person with military service experience based on a filter prepared based on an objective, reasonable and consistent standard.

According to an embodiment of the present invention, job competency can be improved even during military service life as a customized position can be recommended based on the subject's major, social experience, qualifications, etc. before enlisting in the military.

The effects of the present invention are not limited to the effects described above. Effects not mentioned will be clearly understood by those of ordinary skill in the art to which the present invention pertains from this specification and the accompanying drawings.

1 is a block diagram of a job recommendation apparatus according to an embodiment of the present invention.
2 is a block diagram of a job recommendation server constituting a job recommendation device according to an embodiment of the present invention.
3 is a block diagram of a filter generating unit constituting a job recommendation apparatus according to an embodiment of the present invention.
4 is a configuration diagram of a job recommendation unit constituting a job recommendation apparatus according to an embodiment of the present invention.
5 is a block diagram of a first filter generating unit constituting the job recommendation apparatus according to an embodiment of the present invention.
6A and 6B are diagrams illustrating job competency unit documents of a dictionary including a plurality of job competency units.
7 is an exemplary diagram of a first filter according to a first embodiment of the present invention.
8 is a configuration diagram of a first job recommendation unit constituting a job recommendation apparatus according to an embodiment of the present invention.
9 is an exemplary view of a first filter according to a second embodiment of the present invention.
10 is a block diagram of a first job recommendation unit constituting a job recommendation device according to a second exemplary embodiment of the present invention.
11 is a configuration diagram of a first job recommendation unit constituting a job recommendation apparatus according to another embodiment of the present invention.
12 is a block diagram of a second filter generating unit constituting the job recommendation apparatus according to an embodiment of the present invention.
13 is an exemplary diagram of a second filter according to the first embodiment of the present invention.
14 is an exemplary diagram of a second filter according to a second embodiment of the present invention.
15 is a configuration diagram of a second job recommendation unit according to the first embodiment constituting the job recommendation apparatus according to the embodiment of the present invention.
16 is a configuration diagram of a second position recommendation unit according to a second embodiment constituting the job recommendation apparatus of the present invention.
17 is a configuration diagram of a second job recommendation unit constituting a job recommendation device according to another embodiment of the present invention.

Other advantages and features of the present invention, and a method of achieving them, will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, and the present invention is only defined by the scope of the claims. Unless defined, all terms (including technical or scientific terms) used herein have the same meaning as commonly accepted by common skill in the prior art to which this invention belongs. A general description of known configurations may be omitted so as not to obscure the gist of the present invention. In the drawings of the present invention, the same reference numerals are used as far as possible for the same or corresponding components.

The terms used in the present application are only used to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present application, terms such as "comprise", "have" or "have" are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one It should be understood that it does not preclude the possibility of the presence or addition of or more other features or numbers, steps, operations, components, parts, or combinations thereof.

As used throughout this specification, '~ unit' is a unit that processes at least one function or operation, and may refer to, for example, a hardware component such as software, FPGA, or ASIC. However, '~part' is not meant to be limited to software or hardware. The '~ unit' may be configured to reside on an addressable storage medium or may be configured to refresh one or more processors. As an example, '~ part' denotes components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, procedures, and sub It may include routines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. A function provided by a component and '~ unit' may be performed separately by a plurality of components and '~ unit', or may be integrated with other additional components.

In this specification, the term 'job competency' is used. As used herein, the term 'job competency' refers to knowledge, skills, and attitudes required for job performance in an industrial field. For example, it may be defined as a competency unit according to the concept systematized by the National Competency Standards (NCS) by the Human Resources Development Service of Korea or the Korea Vocational Competency Development Institute. Specifically, regarding NCS, information can be obtained from https://www.ncs.go.kr/, and the main classification (24) classified by referring to the Korean Employment Classification of Occupations (KECO), etc. According to the sub-category (79), sub-category (253), and sub-category (1,001), the knowledge, skill, attitude can be However, the above-described example is only an example, and job competency should not be construed as a concept limited to the above-described capability unit or capability unit element. For example, job competency may refer to all descriptions such as the name, definition, job basic competency, evaluation items, and performance criteria for each competency unit element defined by the NCS. In addition, job competency may be defined as the ability to perform work for a certain job, not the definition by NCS. In addition, job competency may be a concept classified according to a certain standard to include knowledge, skills, attitudes, etc. required for job performance in an industrial field not described above.

As used throughout this specification, the definition of job competency unit (consisting of the dictionary collection included in the dictionary) does not simply mean the job ability itself, but data related to the ability required by each job competency. It should be understood as a concept of a data set containing it should be

Similar to the definition of the job competency unit for the military position unit, which is a term used throughout this specification, it does not simply mean the position itself, but is a data set including data related to the ability required for each position. It should be understood as a concept, and it should be interpreted as a concept including a glossary or sentence-defined commentary, position description, position commentary, etc., which contains a plurality of words for each position, and indexes thereof.

1 is a block diagram of a job recommendation apparatus according to an embodiment of the present invention. Referring to FIG. 1 , a job recommendation apparatus 10 according to an embodiment of the present invention may include a job recommendation server 30 and a database 40 .

The job recommendation server 30 may collect competency-related data from the target terminal 20 and recommend a job suitable for the target using a filter generated based on a dictionary including a plurality of job competency units. In addition, based on the recommended job, it is possible to recommend related qualifications and training necessary to improve job competency.

A filter according to an embodiment of the present invention will be described in more detail later with reference to drawings such as FIGS. 7, 9, 13, and 14 .

The subject terminal 20 is a terminal used by the subject, and may be implemented as, for example, a smartphone, tablet PC, video phone, desktop PC, laptop PC, netbook computer, workstation, PDA, PMP, and wearable device, etc. However, the present invention is not limited thereto.

The database 40 is a dictionary collection including a plurality of job competency units (eg, a report by competency unit according to NCS classification, NCS classification) It is possible to store information such as a vocabulary extracted from the report for each capability unit) and filters.

The database 40 is not necessarily interpreted as being provided separately from the job recommendation server 30 , and should be understood as a concept including a memory provided in the job recommendation server 30 . The job recommendation server 30 may access the target terminal, the recruiter terminal, and the database 40 through various types of communication. Here, various types of communication may be implemented by a network interface capable of wired and/or wireless communication, for example, short-range communication (eg, Bluetooth, WiFi direct, or infrared data association (IrDA)) or long-distance communication (eg, cellular). network, Internet, or computer network (eg, LAN or WAN)). Wired/wireless network interfaces communicate according to various communication standards, such as, for example, IEEE, Zigbee, 3rd Generation (3G), 3rd Generation Partnership Project (3GPP), Long Term Evoloution (LTE), and 5th Generation (5G). It may be configured as a network interface that performs, but is not limited thereto.

2 is a block diagram of a job recommendation server constituting a job recommendation device according to an embodiment of the present invention. 1 and 2 , the job recommendation server 30 may include a filter generating unit 100 , a job classification filter 200 , and a job recommendation unit 300 .

The filter generating unit 100 may generate one or a plurality of different filters 200 based on a dictionary including a plurality of job competency units.

A dictionary containing a plurality of job competency units (hereinafter referred to as a dictionary collection) includes job competency and It refers to a dataized set in which related data (eg, words, sentences, etc. related to job competency) are classified and organized according to each defined job competency unit. Each job competency unit may include an index representing job competency and data related to job competency.

The filter 200 derives an output value related to the input value when an input value is provided. For example, the input value may be a word related to job ability or job ability. For example, the output value may be a word related to job ability or job ability. In an embodiment, the filter 200 may be calculated based on a dictionary including a plurality of job competency units. The filter 200 generated by the filter generator 100 may be stored in the database 40 . The job recommendation unit 300 may recommend a job suitable for the target based on the target-related data and the filter 200 generated by the filter generator 100 .

In an embodiment, the dictionary including a plurality of job competency units includes a document related to a competency unit according to a job classification defined according to the National Competency Standards (NCS) classification, and a military position (job) in a system similar to the NCS. A job description for each military position defining a military position classification constructed by defining It may include a word collection including words extracted from

The classification of military positions defined in this specification standardizes and systematizes the job competencies required for each military position. Classification of military positions classifies responsibilities by position, and the capabilities required according to the responsibilities of each position are defined. Responsibilities for each position and required competencies are created as a dictionary that defines the duties according to each military position. More specifically, the definition of the classification of military positions can be defined in a method and system similar to the NCS classification. According to the classification of military positions, a military position definition document defining a military position may be included for each classified military position. The military position definition document systematizes and includes responsibilities for each position and required competencies. For example, the military position definition document may include the contents of knowledge, skills, and abilities required to perform duties in the military by subdividing them into 'competence unit elements', 'performance standards', and 'knowledge/skill/attitude'. . The military position definition document may exist as data.

For example, the classification of military positions may be collected and defined from data defined on the Homepage of the Military Manpower Administration of the Republic of Korea (https://www.mma.go.kr/). For example, in defining military service positions, the upper classification is divided into 'Army', 'Navy', 'Marine Corps', and 'Air Force', and 'Army' is 'Technical Administrative Soldier', 'Employment Custom Specialist', 'Paid Volunteer Soldier' ', 'professional special forces soldier', 'language soldier', 'KATUSA Soldier', etc. For example, among them, 'technical administrator' is 'K-series tank control', 'tank communication maintenance', 'K-21 infantry combat' by specialty. It can be classified into 'Vehicle Control', 'Armoured Vehicle Control', 'Balkan Maintenance', 'Field Construction', 'Survey', 'Crane Operation', 'Password Operation', etc. A ‘specialty’ defined by the Military Manpower Administration of the Republic of Korea may be defined as a ‘military position unit’ as defined in this specification. The Military Manpower Administration website provides information related to ‘jobs’, ‘missions’, and ‘related fields’ defined for each ‘specialty’. For example, information related to ‘K-series tank control’ can be viewed at https://www.mma.go.kr/conscription/recruit_service/procedure/army/S_board_text.do?mc=mma0000388&gun_gbcd=1&mojip_gbcd=1. Information on ‘jobs’, ‘missions’, and ‘related fields’ for each ‘specialty’ provided by the Military Manpower Administration can be defined as the necessary capabilities related to the special skill. Alternatively, the data related to the required capability according to the military job is defined as the 'competence unit element', 'performance standard', ' Knowledge, skill, attitude, etc. can be subdivided and defined and stored in the dictionary.

In the embodiment, the filter generating unit 100 is a dictionary collection of each job competency unit (hereinafter, when referred to as a 'job competency unit' without special mention, the job competency unit is provided as a concept including a military position unit) After extracting data according to each job competency unit from can create For example, the data for each job competency unit includes 'classification number', 'name', 'definition', 'evaluation guidelines', 'basic job competency', 'evaluation items', and job competency belonging to the job competency unit. It may be a data-formatted document including information such as 'name', 'performance standard', 'knowledge', 'skill', 'attitude' for each unit element.

3 is a block diagram of a filter generating unit constituting a job recommendation apparatus according to an embodiment of the present invention. 4 is a configuration diagram of a job recommendation unit constituting a job recommendation apparatus according to an embodiment of the present invention. 3 and 4 , the filter generator 100 may include a first filter generator 120 , a second filter generator 140 , and a filter updater 160 . The filter 200 may include a first filter 220 and a second filter 240 . The job recommendation unit 300 may include a first job recommendation unit 320 and a second job recommendation unit 340 .

The first filter generating unit 120 may generate the first filter 220 of a two-dimensional matrix representing a relationship between each job competency unit and words extracted from data according to each job competency unit. The second filter generator 120 may generate the second filter 240 of a two-dimensional matrix indicating the degree of association between each job competency unit.

The first job recommendation unit 320 may recommend a job to the target based on the first filter 220 generated by the first filter generation unit 120 . The second job recommendation unit 340 may recommend a job to the target based on the second filter 240 generated by the second filter generation unit 140 . Hereinafter, the first filter 220 will be described first, and then the second filter 240 will be described.

5 is a block diagram of a first filter generating unit constituting the job recommendation apparatus according to an embodiment of the present invention. 3 and 5 , the first filter generation unit 120 includes a dictionary collection building unit 122 , a word extraction unit 126 , a relevance analysis unit 130 , and a normalization unit 132 . may include The dictionary collection building unit 122 may construct the dictionary collection 125 by dividing and extracting data according to job competency units from the job competency-related data 400 and integrating the extracted data according to the job competency units. In an embodiment, the dictionary collection building unit 122 may extract pdf documents defining a job competency unit for each of the 24 NCS major categories. The number n of job competency units 124 may be different, and in an embodiment, n represents the number of job competency units 124 belonging to a corresponding classification. The pdf documents according to the n job competency units 124 belonging to the same classification can be utilized to calculate the first filter 220 for the corresponding classification, and a plurality of as many as the number of classifications (eg, of the NCS classification system) According to the large classification, 24) first filters 220 may be calculated.

6A and 6B are diagrams illustrating a document defining the job competency units 124 of the NCS as an example of a dictionary including a plurality of job competency units. The dictionary collection building unit 122 includes 'classification number', 'name', 'level', 'definition', 'basic job competency', 'evaluation matters', 'performance criteria' for each job competency unit element, Data related to job competency units can be extracted from documents containing information such as 'knowledge', 'skill', and 'attitude'. The word extraction unit 126 may extract the words 128 from all n job competency units. In an embodiment, the word extractor 126 may include a morpheme analyzer for dividing sentences in a document defining the job competency units 124 into words of a morpheme unit and tagging the part-of-speech. In an embodiment, the word extraction unit 126 excludes stop words from sentences, and N-grams such as Uni-grams and/or bi-grams (N is an integer of 2 or more) ) to extract words.

The relevance analysis unit 130 may analyze the degree of association between the job capability units 124 defined in the dictionary collection 125 and the words 128 extracted from the job capability units 124 . In an embodiment, the relevance analysis unit 130 is based on TF-IDF (Term Frequency Inverse Document Frequency) n job competency units 124 and m words extracted from n job competency units 124 . The degree of association between the 128 may be analyzed. The degree of association between the job competency units 124 and the words 128 may be calculated according to, for example, Equations 1 to 3, or Equations 1, 2, and 4 below.

[Formula 1]

[Equation 2]

[Equation 3]

[Equation 4]

In Equations 1 to 4, 'tfidf(t,d,D)' is a word frequency inverse document frequency, 't' is a word, 'd' is a document, D is a set of all documents, and tf(t,d) is a word frequency, idf(t,D) is the inverse document frequency, 'ft,d' is the frequency of the word t in document d, and 'df(t,d)' is the number of documents including the word t among all documents indicates

Relevance analysis unit 130 applies not only TF-IDF, but also word embedding technology such as Non-negative Matrix Factorization (NMF), Word2Vec/Doc2Vec, and job competency unit documents. It is also possible to analyze the degree of association between words extracted from the job competency unit documents.

When the number of words 128 extracted from the n job competency units 124 is m, the degree of association between the job competency units 124 and the words 128 can be represented by a matrix of size n×m. have. The normalizer 132 may generate a normalized matrix by dividing each component of the n×m matrix by a set value so that the size of the n×m matrix is normalized to 1. The set value may be different according to each component of the n×m matrix. Also, the set value may be different according to a normalization method. Meanwhile, in the present embodiment, the size of the matrix is normalized to be 1, but it may be transformed to an appropriate size such as 10 or 100. Accordingly, the normalized first filter 220 may be generated for each major classification. By normalizing the first filter 220, it is possible to prevent the priority of recommended tasks belonging to different large categories from being influenced by different matrix sizes between the large classifications, and recommended tasks as a unit of competency within a specific large classification (a large classification with a large matrix size) It is possible to prevent the concentration of jobs and recommend suitable jobs (competency units) to job seekers not only in large classifications with large matrix sizes but also in large classifications with small matrix sizes.

7 is an exemplary diagram of a first filter according to a first embodiment of the present invention. 5 and 7, the first filter 220 is a row (row) unit (first unit) job competency units 124 (a1, a2, ..., an-1, an) Words 128 (v1, v2, . This can be expressed as a two-dimensional matrix that is arranged. Each component value of the two-dimensional matrix of the first filter 220 may be a value obtained by dividing a TF-IDF value of a word by a matrix size and normalizing the value. The first filter 220 may be a set of row vectors having a size of 1×m of each of the n job unit capabilities. The number m of words 128 may be approximately tens to hundreds of thousands of words.

In the above, for the embodiment of the first filter 220 in which job competency units are sequentially arranged in a row unit (first unit), and words extracted from job competency unit documents are sequentially arranged in a column unit (second unit) Although described, in the first filter 220, job competency units may be arranged in a column unit, and words extracted from job competency unit documents may be arranged in a row unit. In this case, each job competency unit vector may be provided as a column vector.

Referring back to FIG. 3 , the filter updater 160 collects update information of a dictionary collection including a plurality of job competency units through a periodic crawling technique, and collects the dictionary collection including a plurality of job competency units. The filter can be updated based on the update information of . The dictionary collection including a plurality of job competency units can be changed and updated to reflect future social changes and the demands of industrial sites, and when new or disappearing jobs exist, job competency units can be added or deleted. In addition, words or required abilities included in the job competency unit may be changed or updated to include more specific terms. For example, when the word ��powerpoint��, which is a program name used in the industrial field, is collected from the self-introduction, which is an example of the job ability-related data 400, powerpoint provides job ability When a word is not included in the unit, the dictionary including the job competency unit may be updated to include it. The filter update unit 160 may update the filter when a dictionary including a plurality of job competency units is updated.

For example, the filter update unit 160 may detect a new classification number or a new name in a dictionary including a plurality of job competency units collected by crawling, or may use classification numbers of existing job competency units or If there is a classification number or name deleted among the names, or by crawling technology, 'job basic competency', 'evaluation items' of job competency units, 'performance standards', 'knowledge' by job competency unit elements When it is determined that information such as , 'skill', 'attitude' has been changed, it is determined that the dictionary book including a plurality of job competency units has been changed or updated, and the filter may be updated.

8 is a configuration diagram of a first job recommendation unit constituting the job recommendation apparatus according to the first embodiment of the present invention. 4, 7, and 8 , the first job recommendation unit 320 may recommend one or more jobs to the target based on the target's competency-related data 500 and the first filter 220 . In an embodiment, the competency-related data 500 includes military service experience. Military service experience refers to the main skills at the time of military service, a portfolio containing data related to military duties, the duties performed at the time of military service, the name of the position, and special skills. Military service period, service period by military position, service period by special skill, etc. may include military assignment history, which is a career related to the position at the time of military service. In addition, competency-related data 500 is education information, possessed certification information, portfolio, resume, self-introduction, SNS data, data collected by a chatbot that collects history information in the form of conversation with the subject, and collected through the subject's voice The data may include digitized personal information such as social relationship information of the subject, SNS interests, content viewing records, and search information. The competency-related data 500 may be collected from the target terminal, or collected from the target's SNS or email, a job search relay server, a school or national administrative agency server that stores and manages information about the target. The first job recommendation unit 320 may include a word extraction unit 322 , a vectorization unit 324 , a similarity calculation unit 326 , and a recommended job determination unit 328 .

The word extraction unit 322 may extract competency-related words from the subject's capability-related data 500 . The word extractor 322 may include a morpheme analyzer that divides sentences in the competency-related data 500 into words of a morpheme unit and tags parts of speech. In an embodiment, the word extraction unit 322 may extract a noun-signing morpheme after morpheme analysis, remove stop words, and then apply N-grams such as bi-grams in the case of a noun continuation. . The word extraction unit 322 may extract only words corresponding to the words 128 defined in the first filter 220 from among the words of the capability-related data 500 .

The vectorization unit 324 may vectorize the words extracted from the capability-related data 500 by the word extraction unit 322 according to the arrangement order of the words 128 defined in the first filter 220 . In the first filter 220, the order of the words 128 is arranged in a column unit. The vectorization unit 324 extracts k words (k is an integer greater than or equal to 0 and less than or equal to m) extracted from the competency-related data 500 according to the arrangement order of the words 128 defined in the first filter 220 . can be converted into row vectors of size 1×m. In this case, in the row vector transformed from the words extracted from the capability-related data 500 , k components may have a size greater than 0, and the remaining (m-k) components may have a value of 0. The component value of each of the k words may be determined as a value corresponding to the term frequency (TF) of the word in the competency-related data 500, and is a value irrespective of the frequency by applying a binary score. may be decided. When the word frequency (TF) is used, the normalized term frequency may be used to calculate similarity (eg, cosine similarity). When using a binary score, it can be applied by taking the square and then dot product.

For example, assuming that the word 'architecture design' is included twice in the competency-related data 500 and the word corresponding to 'architecture design' in the first filter 220 is defined as the 12145th column, the word When the frequency is not considered, the value of the 12145th component of the primary vector generated by the vectorization unit 324 is determined to be '1'. When the frequency of words is considered, the primary vector generated by the vectorization unit 324 is The value of the 12145th component of '2' may be determined as '2'. Thereafter, a normalized row vector (feature vector) may be generated by a vector normalization operation. The similarity calculator 326 calculates the similarity between the feature vector and the capability units of the job seeker based on the row vector (feature vector) generated from the capability-related data 500 by the vectorization unit 324 and the first filter 220 . can be calculated. In an embodiment, the similarity calculating unit 326 may calculate a similarity (eg, cosine similarity) between the job seeker's feature vector and a row vector (a competency unit vector) of each of the capability units.

In an embodiment, the cosine similarity may be calculated as a value obtained by dividing the feature vector of the subject and the row vector of each of m job competency units by the size of the feature vector and the row vector size of the job competency unit after dot product. When the feature vector is normalized, the operation of dividing by the size of the feature vector may be omitted. The similarity calculating unit 326 may calculate the similarity with the feature vector of the subject for each of n job competency units belonging to one major category. The recommended job determination unit 328 determines the priority of the job competency units in the order of the highest similarity with the target's feature vector among n similarity values between the target's feature vector and the job competency units, and selects a recommended job to be recommended to the target. can decide In addition, it is possible to provide certification information necessary for the recommended job or education information necessary to improve competency.

In an embodiment, the recommended job determination unit 328 may determine a recommended job from one or a plurality of major categories in an order of high similarity between the feature vector of the subject and the job competency units for each major category. Alternatively, when the subject makes an input for selecting a specific large classification using the subject interface, a job (job competency unit) belonging to the large classification may be recommended based on the filter only for the selected large classification. In the above, the case where various filters are created for each major category and job recommendation is described as an example, and it is also possible to recommend a job by creating one filter for all job competency unit documents belonging to all major categories.

9 is an exemplary view of a first filter according to a second embodiment of the present invention. 5 and 9, the first filter 220 is a row (row) unit (first unit) in the military position units 1124 (b1, b2, ..., bn-1, bn) It is arranged and expressed as a two-dimensional matrix in which the military position unit and the previously extracted words 128 (v1, v2, ..., vm-1, vm) are arranged in a column unit (the second unit). can Each component value of the two-dimensional matrix of the first filter 220 may be a value obtained by dividing a TF-IDF value of a word by a matrix size and normalizing the value. The first filter 220 may be a set of row vectors having a size of 1×m of each of the n job unit capabilities. The number m of words 128 may be approximately tens to hundreds of thousands of words.

With respect to the embodiment of the first filter 220 in which military position units are sequentially arranged in a row unit (first unit), and words extracted from job competency unit documents are sequentially arranged in a column unit (second unit) Although described, in the first filter 220, military position units may be arranged in a column unit, and words extracted from job competency unit documents may be arranged in a row unit. In this case, each group direct unit vector may be provided as a column vector.

10 is a configuration diagram of a first position recommendation unit constituting the position recommendation apparatus of the present invention. 4, 9, and 10 , the first position recommendation unit 1320 may recommend one or more positions to the subject based on the subject's competency-related data 500 and the first filter 220 . In an embodiment, the competency-related data 500 includes data collected by a chatbot that collects education information, holding certificate information, portfolio, resume, self-introduction, SNS data, and history information in the form of conversation with the subject, and the subject's voice. It may include digitized personal information such as data collected through, social relationship information of the subject, interests of SNS, content viewing records, and search information. The competency-related data 500 may be collected from the target terminal, or collected from the target's SNS or email, a job search relay server, a school or national administrative agency server that stores and manages information about the target. The first position recommendation unit 1320 may include a word extraction unit 322 , a vectorization unit 324 , a similarity calculation unit 326 , and a recommended position determination unit 1328 .

The word extraction unit 322 may extract competency-related words from the subject's capability-related data 500 . The word extractor 322 may include a morpheme analyzer that divides sentences in the competency-related data 500 into words of a morpheme unit and tags parts of speech. In an embodiment, the word extraction unit 322 may extract a noun-signing morpheme after morpheme analysis, remove stop words, and then apply N-grams such as bi-grams in the case of a noun continuation. . The word extraction unit 322 may extract only words corresponding to the words 128 defined in the first filter 220 from among the words of the capability-related data 500 .

The vectorization unit 324 may vectorize the words extracted from the capability-related data 500 by the word extraction unit 322 according to the arrangement order of the words 128 defined in the first filter 220 . In the first filter 220, the order of the words 128 is arranged in a column unit. The vectorization unit 324 extracts k words (k is an integer greater than or equal to 0 and less than or equal to m) extracted from the competency-related data 500 according to the arrangement order of the words 128 defined in the first filter 220 . can be converted into row vectors of size 1×m. In this case, in the row vector transformed from the words extracted from the capability-related data 500 , k components may have a size greater than 0, and the remaining (m-k) components may have a value of 0. The component value of each of the k words may be determined as a value corresponding to the term frequency (TF) of the word in the competency-related data 500, and is a value irrespective of the frequency by applying a binary score. may be decided. When the word frequency (TF) is used, the normalized term frequency may be used to calculate similarity (eg, cosine similarity). When using a binary score, it can be applied by taking the square and then dot product.

The similarity calculator 326 compares the row vector (feature vector) generated from the capability-related data 500 by the vectorization unit 324 and the first filter 1220 between the subject's feature vector and the military rank units. similarity can be calculated. In an embodiment, the similarity calculating unit 326 may calculate a similarity (eg, cosine similarity) between a feature vector of the subject and a row vector (job capability unit vector) of each of the military position units.

In an embodiment, the cosine similarity may be calculated as a value obtained by dividing the feature vector of the subject and the row vector of each of m job competency units by the size of the feature vector and the row vector size of the job competency unit after dot product. When the feature vector is normalized, the operation of dividing by the size of the feature vector may be omitted. The similarity calculating unit 326 may calculate the similarity with the feature vector of the subject for each of n job competency units belonging to one major category. The recommended job determination unit 328 determines the priority of the job competency units in the order of the highest similarity with the target's feature vector among n similarity values between the target's feature vector and the job competency units, and selects a recommended job to be recommended to the target. can decide

In an embodiment, the recommended position determining unit 1328 may determine the recommended position in the order of a high degree of similarity between the target's feature vector and military position units.

According to the embodiment of the present invention, the filter is not created by learning using the subject's resume, portfolio, self-introduction, etc. (past job experience), but only a dictionary containing a plurality of job competency units. By creating a filter based on the filter and recommending a job or military position to the subject, it is possible to prevent an error factor from occurring in the filter due to uncertain learning data. You may recommend a job or military position.

According to an embodiment of the present invention, job competency can be improved even during military service life as a customized position can be recommended based on the subject's major, social experience, qualifications, etc. before enlisting in the military.

With respect to the method for defining the dictionary, the job recommendation device and job recommendation method according to the embodiment of the present invention have been described by taking as an example building job competency unit data by crawling a document defined by NCS. The job recommendation device and job recommendation method according to the embodiment include not only NCS, but also National Occupational Standards (NOS), Vocational Ability Assessment Standards (VAAS), National Skill Standard (NSS), National Qualification Framework (NQF), and Korean Qualifications Framework (KQF). ), KECO (Korean Employment Classification of Occupations), or other job competency standards that are not listed may be crawled to construct job competency unit data in the dictionary. In addition, even if it is not defined as a standard, it can be applied to the case where knowledge, skills, attitudes, etc. are defined for job competency.

11 is a configuration diagram of a first job recommendation unit constituting a job recommendation apparatus according to another embodiment of the present invention. 7 and 10 , the first job recommendation unit 320 classifies the recruitment-related jobs based on the hiring related data 600 and the first filter 220 of the recruiter, and based on the job classification information of the recruiter, One or more job postings may be recommended to a job seeker. In an embodiment, the recruitment-related data 600 may include recruitment information, such as career, educational program, academic background, job experience, etc. required by a recruiter such as a company or a public institution. Recruitment-related data 600 may be collected from a recruiter terminal, or may be collected from a recruiter's SNS or email, a job search relay server, a school or national administrative agency server that stores and manages information about the recruiter. In addition, the recruitment-related data 600 may be collected by a chatbot that collects information in the form of a conversation with the recruiter, or may be collected through voice analysis of the recruiter. The first job recommendation unit 320 may include a word extraction unit 330 , a vectorization unit 332 , a similarity calculation unit 334 , and a job classification unit 336 .

The word extraction unit 330 may extract employment-related words from the employment-related data 600 of the recruiter. The word extractor 330 may include a morpheme analyzer that divides sentences in the employment-related data 600 into words of a morpheme unit and tags the part-of-speech. In an embodiment, the word extraction unit 330 extracts noun-signing morphemes after morpheme analysis from sentences, removes stop words, and applies N-grams such as bi-grams in the case of noun continuation. can do. Except for stop words, words may be extracted as N-grams (N is an integer greater than or equal to 2) such as uni-grams and/or bi-grams. The word extraction unit 330 may extract only words corresponding to the words 128 defined in the first filter 220 from among the words of the employment-related data 600 .

The vectorization unit 332 may vectorize the words extracted from the employment-related data 600 by the word extraction unit 330 according to the arrangement order of the words 128 defined in the first filter 220 . The vectorization unit 332 is a p word (p is an integer greater than or equal to 0 and less than or equal to m) extracted from the employment-related data 600 according to the arrangement order of the words 128 defined in the first filter 220 . can be converted into row vectors of size 1×m. In this case, in the row vector (feature vector) transformed from the words extracted from the recruitment-related data 600, p components out of m components have a size greater than 0, and the remaining (mp) components have a value of 0. can have The component value of each of the p words may be determined as a value corresponding to the term frequency (TF) of the word in the employment related data 600, and based on the binary score, this value of 1 regardless of the frequency may be decided. When the word frequency (TF) is used, the normalized term frequency may be used to calculate similarity (eg, cosine similarity). When using a binary score, it can be applied by taking the square and then dot product.

For example, assuming that the word 'architecture design' is included 3 times in the recruitment-related data and the word corresponding to 'architecture design' in the first filter 220 is defined as the 12145th column, the frequency of the word is considered If not, the value of the 12145th component of the primary row vector (feature vector) generated by the vectorization unit 332 is determined to be '1', and when the frequency of words is considered, the primary row generated by the vectorization unit 332 is The value of the 12145th component of the vector (feature vector) may be determined to be '3'. We can then generate normalized row vectors (feature vectors) by vector normalization (when considering the frequency of words) or vector square and dot product (when applying binary scores) operations.

The similarity calculator 334 compares the row vector (feature vector) generated from the recruitment-related data 600 by the vectorization unit 332 and the feature vector of the recruiter and the job competency units based on the first filter 220 . similarity can be calculated. In an embodiment, the similarity calculating unit 334 may calculate a similarity (eg, cosine similarity) between the feature vector of the recruiter and the row vector of each of the job competency units. In an embodiment, the cosine similarity may be calculated as a value obtained by dividing the size of the feature vector of the recruiter by the size of the row vector of the job competency unit after dot product of the feature vector of the recruiter and the row vector of each of m job competency units. The similarity calculating unit 334 may calculate a cosine similarity with a feature vector of the recruiter for each of n job competency units belonging to one major category. The job classification unit 336 determines the priority of the job competency units in the order of the cosine similarity with the recruiter's feature vector among n cosine similarity values between the recruiter's feature vector and the job competency units to determine the priority of job competency units (jobs) related to the recruiter. competency unit) can be classified according to the job competency standard classification.

In an embodiment, the job classification unit 336 may provide a job classification result in one or a plurality of major categories in an order of high similarity between the recruiter's feature vector and the job competency units for each major category. Alternatively, when the recruiter inputs to select a specific large category using the target interface, jobs belonging to the large category may be classified based on the filter only for the selected large category. In the above, the case of classifying the job of the recruiter by creating various filters for each major category has been described as an example.

When one or more job classification information related to the recruiter is determined, the employment information is provided to a job seeker matching the job classification information of the corresponding recruiter, so that job matching between the job seeker and the recruiter may be performed based on the job competency unit. Therefore, a person who meets the job aptitude of a company or a public institution can be recommended to the recruiter, and recruitment information that can perform a suitable job based on the job competency unit is provided to the recruiter. can prevent

Hereinafter, an embodiment of creating the second filter 240 to recommend a job will be described. 12 is a block diagram of a second filter generating unit constituting the job recommendation apparatus according to an embodiment of the present invention. Referring to FIGS. 3 and 12 , the second filter generating unit 140 generates the second filter generator 140 of a two-dimensional matrix representing the degree of association between the job competency units 124 extracted from the dictionary collection 125 including a plurality of job competency units. Two filters 240 may be created.

The second filter generation unit 140 may include a dictionary collection building unit 122 , a word extraction unit 146 , and a relationship analysis unit 148 between classifications. The dictionary collection building unit 122 may extract job competency units 124 from the job capability related data 400 . In an embodiment, the dictionary collection building unit 122 may extract job competency units 124 from job capability-related data. The number of extracted job competency units 124 may be n. The n job competency units 124 may be utilized to calculate the second filter.

The word extraction unit 146 may extract words from all of the n job competency units 124 . The word extractor 146 may include a morpheme analyzer that divides the sentences in the job competency units 124 into words of a morpheme unit and tags the part-of-speech. In an embodiment, the word extraction unit 146 may extract a noun-signing morpheme after morpheme analysis, remove stop words, and then apply N-grams such as bi-grams in the case of a noun continuation. .

The relationship between classification analysis unit 148 may analyze the degree of correlation between the job competency units 124 based on the similarity of the words extracted from the job competency units 124 . In an embodiment, the relationship analysis unit 148 between classifications may analyze the correlation between the n job competency units 124 based on a Term Frequency Inverse Document Frequency (TF-IDF).

In the embodiment, when m words are extracted from all job competency units 124 by the word extraction unit 146 , the inter-class relationship analysis unit 148 determines that the order of all words is mapped to the column order. A row vector of size ×m can be created. The classification relationship analysis unit 148 generates a row vector for each job competency unit 124 from a set of words extracted from each job competency unit 124 , As many (n) number of row vectors can be generated. Then, for all n × n combinations of the n job competency units 124 each other, for example, the cosine similarity between the row vectors of the two job competency units 124 is calculated according to Equations 1 to 4 above. By calculating, the relationship between the job competency unit documents 124 can be analyzed. Relation values between the job competency unit documents 124 may be represented by a two-dimensional matrix of n×n size, and the two-dimensional matrix may be stored as the second filter 240 .

13 is an exemplary diagram of a second filter according to the first embodiment of the present invention. 12 and 13, the second filter 240 is a row (row) unit (first unit) job capability units 124 (a1, a2, ..., an-1, an) It can be expressed as a two-dimensional matrix in which the job competency units 124 (a1, a2, ..., an-1, an) are arranged in a column unit (a second unit). The job competency units 124 (a1, a2, ..., an-1, an) may be defined in the same arrangement order in a row unit and a column unit. Each component value of the two-dimensional matrix of the second filter 240 may be a similarity value between a plurality of row vectors generated from a plurality of word sets each extracted from a plurality of job competency units. Since the diagonal component of the second filter 240 is a similarity value between a plurality of same row vectors extracted from the same job competency unit, the component value becomes 1. The second filter 240 may be a set of row vectors having a size of 1×n of each of the n job unit capability units. A large component value in the second filter 240 means that the similarity between the two job competency units is large, and in the opposite case, the similarity between the two job competency units is small.

14 is an exemplary diagram of a second filter according to a second embodiment of the present invention. 12 and 14, in the second filter 1240, job competency units (a1, a2, ..., an-1, an) are arranged in a row unit (a first unit), It may be expressed as a two-dimensional matrix in which the military position units (b1, b2, ..., bm-1, bm) are arranged in a column unit (a second unit). Each component value of the two-dimensional matrix of the second filter 1240 may be a similarity value between a plurality of row vectors generated from a plurality of word sets respectively extracted from job competency units and military position units. A large component value in the second filter 240 means that the similarity between the two job competency units is large, and in the opposite case, the similarity between the two job competency units is small.

15 is a configuration diagram of a second job recommendation unit according to the first embodiment constituting the job recommendation apparatus according to the embodiment of the present invention. 4, 13 or 14 and 15 , the second job recommendation unit 340 assigns one or more jobs to the subject based on the subject's competency-related data 500 and the second filters 240 and 1240. can recommend The second job recommendation unit 340 may include a word extraction unit 342 , a job capability analysis unit 344 , and a recommended job determination unit 346 .

The word extractor 342 may extract competency-related words from the subject's competency-related data 500 . The word extractor 342 may include a morpheme analyzer that divides sentences in the competency-related data 500 into words of a morpheme unit and tags parts of speech. In an embodiment, the word extraction unit 342 extracts noun-signing morphemes after morpheme analysis, removes stop words, and applies N-grams such as bi-grams in the case of noun continuation, and the subject Words related to job competency units can be extracted.

The job competency analysis unit 344 determines one or more first job competency units related to the subject from the words extracted from the competency-related data 500 , and based on the second filters 240 and 1240 , the first job competency unit One or more second job competency units having a high degree of similarity to As another example, when a level score for the first job competency unit is input by the subject, or completion information such as career or educational program related to the first job competency unit is input, the job competency analysis unit 344 may perform the second filter Based on the similarity matrix relationship between job competency units of (240, 1240), the level score or completion information for the first job competency unit may be reflected as the level score or completion information for other job competency units.

For example, in the second filter 240 shown in FIG. 13 , the relevance of the an-th job competency unit (the second job competency unit) to the an-1th job competency unit (the first job competency unit) is 0.30. , if the subject's level score related to the an-1st job competency unit is entered as 90 points or measured/evaluated or analyzed, the subject's level score can be reflected as 27 points for the anth job competency unit as well. Accordingly, based on the second filter 240, the job search-related history for the first job competency unit of the subject can be extended to other job competency units, and various job competency units suitable for the subject can be recommended to the subject. .

16 is a configuration diagram of a second position recommendation unit according to a second embodiment constituting the job recommendation apparatus of the present invention. 4, 14 and 16 , the second position recommendation unit 1340 may recommend one or more military positions to the subject based on the subject's competency-related data 500 and the second filter 1240 . . The second position recommendation unit 1340 may include a word extraction unit 342 , a job ability analysis unit 344 , and a recommended position determination unit 1346 .

The word extractor 342 may extract competency-related words from the subject's competency-related data 500 . The word extractor 342 may include a morpheme analyzer that divides sentences in the competency-related data 500 into words of a morpheme unit and tags parts of speech. In an embodiment, the word extraction unit 342 extracts noun-signing morphemes after morpheme analysis, removes stop words, and applies N-grams such as bi-grams in the case of noun continuation, and the subject Words related to job competency units can be extracted.

The job competency analysis unit 344 determines one or more first job competency units related to the subject from the words extracted from the competency-related data 500 , and based on the second filter 1240 , the first job competency unit and the high As one or more second job competency units having a similarity, for example, a military position unit may be extracted.

The recommended position determining unit 1328 may determine the recommended position in the order of a high degree of similarity between the target's feature vector and the military position units.

According to the embodiment of the present invention, the filter is not created by learning using the subject's resume, portfolio, self-introduction, etc. (past job experience), etc., but based on a dictionary including a plurality of job competency units. As a filter is created to recommend a job or military position to the target, it is possible to recommend a job or military position suitable for the target based on the filter prepared based on objective, reasonable, and consistent criteria.

17 is a configuration diagram of a second job recommendation unit constituting a job recommendation device according to another embodiment of the present invention. 15 and 17 , the second job recommendation unit 340 analyzes the employment-related job competency criteria based on the hiring related data 600 and the second filter 240 of the recruiter, and provides the target with one or more job information. can recommend The second job recommendation unit 340 may include a word extraction unit 348 and a job competency standard analysis unit 350 .

The word extraction unit 348 may extract job-related words related to employment from the employment-related data 600 of the recruiter. The word extractor 348 may include a morpheme analyzer that divides the sentences in the recruitment-related data 600 into words of a morpheme unit and tags the part-of-speech. In an embodiment, the word extraction unit 348 may extract a noun-signing morpheme after morpheme analysis, remove stop words, and then apply N-grams such as bi-grams in the case of a noun continuation. . The word extraction unit 348 may extract words related to job competency units from among the words of the employment related data 600 .

The job competency standard analysis unit 350 determines one or more third job competency units related to the recruiter from the words extracted from the hiring-related data 600 , and based on the second filters 240 and 1240 , the third job competency One or more fourth job competency units having a high degree of similarity to the unit may be extracted. As another example, when the level score required for the third job competency unit is input by the recruiter, or completion information such as military experience or education program required in relation to the third job competency unit is input, the job competency standard analysis unit ( 350) is the level score or completion information required for the third job competency unit based on the similarity matrix relationship between the job competency units of the second filter 240, and the level score or completion information required for other job competency units. can be reflected as Accordingly, based on the second filters 240 and 1240, the employment-related history requested by the recruiter for the first job competency unit can be extended to the level required for other job competency units, so that various job competency units of the recruiter Can recommend candidates who meet the level of demand for employees to the recruiter.

When one or more job competency units related to the recruiter are determined, recruitment information may be provided to a target matching the job competency unit information of the recruiter, so that job matching between the target person and the recruiter may be made based on the job competency unit. Therefore, a person who meets the job aptitude of a company or a public institution can be recommended to the recruiter, and recruitment information that can perform a suitable job based on the job competency unit is provided to the recruiter. As the military career can be converted into a civilian career, the job performance performed in the military can also be recognized as a career.

At least a part of the job recommendation method according to an embodiment of the present invention can be written as a program that can be executed on a computer, and can be implemented in a general-purpose digital computer that operates the program using a computer-readable recording medium. Computer-readable recording media include volatile memory such as SRAM (Static RAM), DRAM (Dynamic RAM), SDRAM (Synchronous DRAM), ROM (Read Only Memory), PROM (Programmable ROM), EPROM (Electrically Programmable ROM), Nonvolatile memory such as Electrically Erasable and Programmable ROM (EEPROM), flash memory device, phase-change RAM (PRAM), magnetic RAM (MRAM), resistive RAM (RRAM), ferroelectric RAM (FRAM), floppy disk, hard disk, or The optical reading medium may be, for example, a storage medium such as a CD-ROM or DVD, but is not limited thereto.

At least a part of the job recommendation method according to an embodiment of the present invention may be implemented as software (eg, a program) including instructions stored in a computer-readable storage medium. At least some of the job recommendation method according to an embodiment of the present invention may be implemented by calling a stored instruction from a storage medium and executing a process according to the called instruction. Instructions may include code generated or executed by a compiler or interpreter. The storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-transitory' means that the storage medium does not include a signal and is tangible, and does not distinguish that data is semi-permanently or temporarily stored in the storage medium.

The above embodiments are presented to help the understanding of the present invention, and do not limit the scope of the present invention, and it should be understood that various modifications are also included in the scope of the present invention. The technical protection scope of the present invention should be defined by the technical spirit of the claims, and the technical protection scope of the present invention is not limited to the literal description of the claims itself, but an invention of a substantially equivalent scope of technical value. It should be understood that it extends to

10: Job Recommendation Device
20: target terminal
30: job recommendation server
40: database
100: filter generating unit
120: first filter generating unit
140: second filter generating unit
160: filter update unit
200: filter
220, 1220: first filter
240, 1240: second filter
300: job recommendation department
320: first job recommendation unit
340: second job recommendation unit
400: job competency related data
500: Competency-related data
600: Recruitment related data

Claims (2)

Recommending, by a job recommendation unit, a job to the subject based on the subject's competency-related data and filters,
The filter is generated based on a dictionary including a plurality of job competency units composed of data related to the required competency according to the job,
The job competency units include a plurality of military position units composed of data related to the required competency according to the military position,
The subject's competency-related data includes the subject's military service career,
The filter is a first filter or a second filter,
The first filter comprises:
A two-dimensional matrix in which job competency units are arranged in a first unit of any one of a row unit and a column unit, and words extracted from data constituting the job competency units are arranged in the other second unit,
The two-dimensional matrix has, as each component value, a degree of association between each job competency unit of the first unit and each word of the second unit,
The second filter for fraud is:
A two-dimensional matrix in which job competency units are arranged in a first unit of any one of a row unit and a column unit, and the job competence units are arranged in another second unit,
The two-dimensional matrix has a degree of relevance between a first job capability unit associated with each job capability unit of the first unit and a second job capability unit associated with each job capability unit of the second unit as each component value,
The steps for recommending the job include:
extracting words related to the military service career of the subject from the competency-related data;
determining a plurality of job competency units related to the subject from the extracted words; and
The one or more job competency units among the plurality of job competency units are used as a first job competency unit, and the second at least one of the plurality of job competency units based on the similarity with the first job competency unit based on the filter. A job recommendation method comprising determining a job competency unit as a recommended job, and determining a plurality of job competency units including the first job competency unit and the second job competency unit as a recommended job. Recommending a position to the subject based on the subject's competency-related data and filters by the job recommendation unit;
It is created based on a dictionary containing a plurality of job competency units composed of data related to the required competency according to the job,
The job competency units include a plurality of military position units composed of data related to the required competency according to the military position,
The subject's competency-related data includes the subject's education information, holding certificate information, portfolio, resume, self-introduction, SNS data, data collected by a chatbot that collects history information in the form of conversation with the subject, and the subject's social relationship information, the subject's interests on SNS, the subject's content viewing record, and at least one of the subject's specific information search history,
The filter is a first filter or a second filter,
The first filter comprises:
A two-dimensional matrix in which job competency units are arranged in a first unit of any one of a row unit and a column unit, and words extracted from data constituting the job competency units are arranged in the other second unit,
The two-dimensional matrix has, as each component value, a degree of association between each job competency unit of the first unit and each word of the second unit,
The second filter comprises:
A two-dimensional matrix in which job competency units are arranged in a first unit of any one of a row unit and a column unit, and the job competence units are arranged in another second unit,
The two-dimensional matrix has a degree of relevance between a first job capability unit associated with each job capability unit of the first unit and a second job capability unit associated with each job capability unit of the second unit as each component value,
The steps for recommending the position are:
extracting words related to the job competency of the subject from the competency-related data; and
determining a plurality of military positions related to the subject from the extracted words; and
The at least one military position unit among the plurality of military position units is a first military position unit, and based on the filter based on the similarity with the first military position unit, the second at least one of the plurality of job competency units A method of recommending a military position, comprising determining a military position unit as a recommended military position position, and determining a plurality of military position unit including the first military position unit and the second military position unit as a recommended military position position.

Images