KR102365659B1 - Apparatus, method and program for providing labor management services - Google Patents

Abstract

The present invention relates to a device for providing a labor management service, which includes: a dispute possibility prediction unit receiving, from a customer terminal, labor information including labor contract information, pay slip information, and employment information and calculating an expected value with respect to the possibility of a labor dispute based on the received labor information; an expected damage prediction unit calculating first expected damage based on the pay slip information, calculating second expected damage based on the employment information, and calculating the total expected damage based on the first expected damage and the second expected damage; an appropriate person count calculation unit receiving customer information from the customer terminal, determining at least one first matching company matching the customer terminal among competitors based on the customer information and pre-stored competitor information, and calculating an appropriate person count based on the customer information and the average productivity of the at least one first matching company; and an appropriate labor cost calculation unit calculating an appropriate productivity based on the customer information and the appropriate person count, determining at least one second matching company matching the customer terminal based on the customer information, the appropriate productivity, the competitor information, and a pre-designated competitor productivity, and calculating an appropriate labor cost based on the labor cost of the at least one second matching company.

Description

APPARATUS, METHOD AND PROGRAM FOR PROVIDING LABOR MANAGEMENT SERVICES

The present invention relates to a labor management service providing apparatus, method and program.

Unless otherwise indicated herein, the material described in this section is not prior art to the claims of this application, and inclusion in this section is not an admission that it is prior art.

In operating a self-employment business, labor management tasks such as manpower management, number of manpower calculations, wage setting, and report preparation are tasks that consume a lot of time and money. In particular, as the size of the business is small, the cost of labor management can be a relatively large burden to the business owner.

It is an object of the present invention to solve the above problems, to provide a labor management service providing apparatus, method, and program capable of providing to customers by calculating an expected value for the possibility of a dispute based on employee employment information.

Another object of the present invention for solving the above problems is to provide a labor management service providing apparatus, method and program that can provide customers by calculating an appropriate number of people and an appropriate labor cost for each industry.

Another object of the present invention for solving the above problems is to provide a labor management service providing apparatus, method and program that can match a customer and a government support project, and provide a matched government support project to a high rank. there is.

In one aspect of the present invention for achieving the above object, there is provided an apparatus for providing a labor management service.

The apparatus may include: a dispute possibility prediction unit configured to receive labor information including labor contract information, pay stub information, and employment information from a customer terminal, and calculate an expected value for the possibility of a labor dispute based on the received labor information; Calculating a first expected amount of damage based on the pay stub information, calculating a second expected amount of damage based on the employment information, and calculating a total expected amount of damage based on the first expected amount of damage and the second expected amount of damage damage prediction unit; receiving customer information from the customer terminal, determining at least one first matching company matching the customer terminal from among a plurality of competitors based on the customer information and pre-stored competitor information, the customer information and the an appropriate number of people calculating unit for calculating an appropriate number of people based on the average productivity of at least one first matching company; and calculating an appropriate productivity based on the customer information and the appropriate number of people, and matching at least one second with the customer terminal based on the customer information, the appropriate productivity, the competitor information, and the productivity of the specified competitor. and an appropriate labor cost calculation unit for determining a matching company and calculating an appropriate labor cost based on the labor cost of the at least one second matching company.

In addition, the dispute possibility prediction unit classifies an item matching a plurality of preset essential clauses among a plurality of clauses included in the labor contract information into a first clause, and the plurality of essential clauses among the plurality of clauses Items that do not match the above are classified into the second clause, and the expected value of the possibility of a dispute corresponding to the labor contract information is based on the number classified in the first and second clauses and the frequency of occurrence of labor disputes. can be calculated by

)에 의하여 결정하고, 상기의 수학식 1에서, PL1은 상기 근로 계약 정보에 대응하는 분쟁 가능성에 대한 기대값을 의미하고, K1은 고객의 직종에서 근로 계약 정보로 인해 노무분쟁이 발생된 빈도에 기초하여 산출된 직종가중치를 의미하며, n은 상기 제1 조항으로 분류된 조항들의 개수를 의미하고, m은 상기 제2 조항으로 분류된 조항들의 개수를 의미하며, P1k는 제1 조항으로 분류된 조항들 각각에 대응하는 무효 확률을 의미할 수 있다.In addition, the expected value for the possibility of a dispute corresponding to the labor contract information is Equation 1 (

), and in Equation 1 above, PL1 means the expected value of the possibility of a dispute corresponding to the labor contract information, and K1 is the frequency of labor disputes occurring due to the labor contract information in the customer's occupation. It means the occupational weight calculated based on the occupational weight, n means the number of articles classified as the first clause, m means the number of clauses classified as the second clause, and P1k is the number of articles classified as the first clause. It may mean an invalid probability corresponding to each of the clauses.

)에 의하여 결정하고, 상기 수학식 2에서, PL2는 급여 명세서 정보에 대응하는 분쟁 가능성에 대한 기대값을 의미하고, K2은 고객의 직종에서 근로 명세서 정보로 인해 노무분쟁이 발생된 빈도에 기초하여 산출된 직종가중치를 의미하며, n은 상기 제1 항목의 개수를 의미하고, m은 상기 제2 항목의 개수를 의미하고, P2k는 상기 제1 항목으로 분류된 항목들 각각에 대응하는 위반 가능성을 의미할 수 있다.In addition, the dispute possibility prediction unit classifies an item matching a plurality of preset mandatory specification items among a plurality of items included in the pay stub information as a first item, and the plurality of mandatory specification items among the plurality of items An item that does not match the item is classified as a second item, and the expected value of the possibility of a dispute corresponding to the pay stub information is calculated by Equation 2 (

( It means the calculated occupational weight, n means the number of the first item, m means the number of the second item, and P2k is the probability of violation corresponding to each of the items classified as the first item can mean

)에 의하여 결정하고, 상기 수학식 3에서, PL3는 상기 고용 정보와 대응되는 분쟁 가능성에 대한 기대값을 의미하고, K3은 고객의 직종에서 상기 고용 정보로 인해 노무분쟁이 발생된 빈도에 기초하여 산출된 직종가중치를 의미하며, WNK는 4대보험의 각각에 대응되는 미가입인원수를 의미하고, TN은 고객이 고용한 직원의 총수를 의미할 수 있다.In addition, the possibility of dispute prediction unit calculates the expected value of the possibility of dispute corresponding to the employment information in Equation 3 (

), and in Equation 3, PL3 means the expected value of the possibility of a dispute corresponding to the employment information, and K3 is the frequency of occurrence of labor disputes due to the employment information in the customer's occupation. It means the calculated job type weight, WNK means the number of uninsured persons corresponding to each of the four major insurances, and TN may mean the total number of employees hired by the customer.

)에 의하여 결정하고, 상기 수학식 5에서, D1은 상기 제1 예상피해액을 의미하고, K2는 고객의 직종에서 근로 명세서 정보로 인해 노무분쟁이 발생된 빈도에 기초하여 산출된 직종가중치를 의미하며, n은 상기 제1 항목의 개수를 의미하고, P2k는 상기 제1 항목으로 분류된 항목들 각각에 대응하는 위반 가능성을 의미하고, PD1k는 상기 제1 항목으로 분류된 항목들 각각에 대응하는 제1 피해액을 의미할 수 있다.In addition, the expected damage amount prediction unit calculates the first expected damage amount by Equation 5 (

), and in Equation 5, D1 means the first expected amount of damage, and K2 means the job type weight calculated based on the frequency of labor disputes occurring due to the information on the labor specification in the customer's job type, , n denotes the number of the first items, P2k denotes the probability of violation corresponding to each of the items classified as the first item, and PD1k denotes the first item corresponding to each of the items classified as the first item. 1 can mean the amount of damage.

)에 의하여 결정하고, 상기 수학식 6에서, D2는 상기 제2 예상피해액을 의미하고, K3은 고객의 직종에서 고용 정보로 인해 노무분쟁이 발생된 빈도에 기초하여 산출된 직종가중치를 의미하며, WNK는 4대보험 각각에 대응되는 미가입인원수를 의미하고, PD2K는 상기 4대보험 각각에 대응되는 제2 피해액을 의미할 수 있다.In addition, the expected damage amount predicting unit, Equation 6 (

), and in Equation 6, D2 means the second expected amount of damage, and K3 means the job type weight calculated based on the frequency of labor disputes occurring due to employment information in the customer's job type, WNK may mean the number of uninsured persons corresponding to each of the four major insurances, and PD2K may mean the second amount of damage corresponding to each of the four major insurances.

)에 기초하여 제1 유사도를 산출하고, 상기 제1 유사도에 기초하여 상기 적어도 하나의 제1 매칭업체를 특정하고, 상기 수학식 7에서, C1은 고객이 운영하는 업체와 경쟁업체 사이의 상권과 매출액의 유사성을 보여주는 상기 제1 유사도를 의미하고, TS1은 고객의 업체의 매출액을 의미하며, TS2는 경쟁업체의 매출액을 의미하고, A는 상기 고객 정보를 변환한 벡터를 의미하고, B는 상기 경쟁업체 정보를 변환한 벡터를 의미하며, 상기 고객 정보 및 상기 경쟁업체 정보 각각에는, 고객 및 경쟁업체 각각에 대한 영업장 면적, 영업장 층수, 영업장에 인접한 기 설정된 영역의 유동인구, 영업장에 인접한 기 설정된 영역의 동종업체밀도 및 영업장에 인접한 기 설정된 영역의 긍정적 인접업종의 수를 포함할 수 있다.In addition, the appropriate number of people calculation unit, Equation 7 (

), calculate the first similarity, specify the at least one first matching company based on the first similarity, and in Equation 7, C1 is the commercial area between the company operated by the customer and the competitor means the first similarity showing the similarity of sales, TS1 means the sales of the customer's company, TS2 means the sales of the competitor, A means the vector converted from the customer information, and B is the It means a vector converted from competitor information, and in each of the customer information and the competitor information, the business area for each customer and competitor, the number of floors of the business, floating population in a preset area adjacent to the business place, and a preset adjacent to the business place It may include the density of peers in the area and the number of positive adjacent businesses in the preset area adjacent to the business site.

)에 기초하여 제2 유사도를 산출하고, 상기 제2 유사도에 기초하여 상기 적어도 하나의 제2 매칭업체를 특정하며, 상기 수학식 8에서, C2은 고객이 운영하는 업체와 경쟁업체 사이의 상권과 생산성의 유사성을 보여주는 상기 제2 유사도를 의미하고, PR1은 상기 적정 생산성을 의미하며, PR2는 상기 경쟁업체의 생산성을 의미하고, A는 고객 정보를 변환한 벡터를 의미하고, B는 경쟁업체 정보를 변환한 벡터를 의미하며, 상기 고객 정보 및 상기 경쟁업체 정보 각각에는, 고객 및 경쟁업체 각각에 대한 영업장 면적, 영업장 층수, 영업장에 인접한 기 설정된 영역의 유동인구, 영업장에 인접한 기 설정된 영역의 동종업체밀도 및 영업장에 인접한 기 설정된 영역의 긍정적 인접업종의 수를 포함할 수 있다.In addition, the appropriate labor cost calculation unit, Equation 8 (

), calculate a second similarity, specify the at least one second matching company based on the second similarity, and in Equation 8, C2 is the commercial area between the company operated by the customer and the competitor means the second degree of similarity showing the similarity of productivity, PR1 means the appropriate productivity, PR2 means the productivity of the competitor, A means a vector converted from customer information, and B is the competitor information is a vector converted from It may include the number of positive adjacent businesses in the preset area adjacent to the business density and business place.

Another aspect of the present invention provides a method of operation for providing a labor management service.

The operation method may include: calculating the possibility of a dispute corresponding to each of the employment contract information, the pay stub information and the employment information from the customer terminal; Calculating a first expected amount of damage based on the pay stub information, calculating a second expected amount of damage based on the employment information, and calculating a total estimated amount of damage based on the first expected amount of damage and the second expected amount of damage ; receiving customer information from the customer terminal, and specifying at least one first matching company matching the customer terminal among a plurality of competitors based on the customer information and pre-stored competitor information, wherein the customer information and the at least one calculating an appropriate number of people based on the average productivity of one first matching company; and calculating appropriate productivity based on the customer information and the appropriate number of people, and matching at least one product with the customer terminal based on the customer information, the appropriate productivity, the competitor information, and the pre-stored productivity of the competitor 2 specifying a matching company, and calculating an appropriate labor cost based on the labor cost of the at least one second matching company.

Further, in another aspect of the present invention, a computer program recorded in a non-transitory recording medium to execute the method of operation may be provided.

Through the labor management service according to an embodiment of the present invention, a customer who uses the service can be provided with an accurate possibility of a labor dispute and an estimated amount of damage quickly.

In addition, through the labor management service according to the embodiment of the present invention, the customer using the service can be provided with the number of people and labor costs optimized for the customer's business quickly.

In addition, through the labor management service according to an embodiment of the present invention, a recommendation degree for a government support project is provided along with a government support project that can be supported to a customer using the service. You can choose government subsidies that can be beneficial.

In addition, through the labor management service according to an embodiment of the present invention, it provides a function to create, deliver, and store online documents that the business owner must legally prepare, deliver, and store to customers who use the service. It can prevent labor disputes between workers.

1 is a schematic diagram of a system for providing a labor management service according to an embodiment.
FIG. 2 is a block diagram exemplarily showing a functional module of the apparatus for providing a labor management service according to FIG. 1 .
3 is a diagram conceptually illustrating a process in which the dispute possibility prediction unit according to FIG. 2 calculates an expected value for the possibility of a dispute based on labor contract information.
FIG. 4 is a diagram conceptually illustrating a process in which the dispute possibility prediction unit according to FIG. 2 calculates an expected value for the possibility of dispute based on pay stub information.
5 is a diagram conceptually illustrating a process in which the dispute possibility prediction unit according to FIG. 2 calculates an expected value for the possibility of a dispute based on employment information.
FIG. 6 is a diagram conceptually illustrating a process in which the expected damage amount predicting unit 102 according to FIG. 2 derives the expected damage amount based on pay stub information and employment information.
FIG. 7 is a diagram conceptually illustrating a process of calculating the appropriate number of people and the appropriate labor cost by the appropriate number of people calculating unit and the appropriate labor cost calculating unit according to FIG. 2 .
8 is a diagram conceptually illustrating a process of matching the maintenance support project based on customer information by the government support project matching unit according to FIG. 2 .
9 is a diagram conceptually illustrating an operation process of the statistical data providing unit according to FIG. 2 .
FIG. 10 is a diagram conceptually illustrating an operation process of the educational material providing unit according to FIG. 2 .
11 is a diagram conceptually illustrating an operation process of a labor management service application according to the present invention.
12 is a diagram conceptually illustrating an operation process of a labor management service application according to the present invention.
13 is a diagram conceptually illustrating an operation process of a labor management service application according to the present invention.
14 is a diagram conceptually illustrating an operation process of a labor management service application according to the present invention.
15 is a diagram conceptually illustrating an operation process of a labor management service application according to the present invention.
16 is a diagram conceptually illustrating an operation process of the report management unit according to FIG. 2 .
FIG. 17 is a diagram conceptually illustrating an operation process of the report management unit according to FIG. 2 .
18 is a diagram exemplarily illustrating a hardware configuration of the providing apparatus according to FIG. 1 .

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include all modifications, equivalents and substitutes included in the spirit and scope of the present invention. In describing each figure, like reference numerals have been used for like elements.

Terms such as first, second, A, and B may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.

When an element is referred to as being “connected” or “connected” to another element, it is understood that it may be directly connected or connected to the other element, but other elements may exist in between. it should be On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that the other element does not exist in the middle.

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” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It should be understood that this does not preclude the existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical and scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a schematic diagram of a system for providing a labor management service according to an embodiment.

Referring to FIG. 1 , the labor management service providing system may include a device 100 , a customer terminal 200 , a training server 300 , an external server 400 , and a terminal in charge 500 .

The customer terminal 200 is a terminal of a customer who wants to use a labor management service, and registers customer information in the device 100, and stores labor contract information, pay stub information, and employment information for employees hired by the customer. By sending to 100 , various functions of the labor management service are available through the device 100 . In this case, the labor management service may be provided in the form of a platform.

In one embodiment, the customer information includes the business site area, the number of floors of the business site, the floating population of the preset area adjacent to the business site, the density of similar businesses in the preset area adjacent to the business site, the number of positive neighboring businesses in the preset area adjacent to the business site, sales , the number of people, salary information, industry information, and the like. In an embodiment, the labor contract information may include information about the labor contract. In an embodiment, the pay stub information may include information about pay stubs that are regularly paid to employees. In an embodiment, the employment information may include information on the age, gender, work type, employment period, monthly working days, monthly working hours, and whether to subscribe to four major insurances of the employees.

The education server 300 may provide educational materials for the legal compulsory education of employees to the device 100 .

The external server 400 may provide labor statistics data, such as average labor cost, average number of people by sales scale, average working hours of workers, and labor shortage rate, to the device 100 . In one embodiment, the external server 400 may be a server of a public institution such as the Ministry of Employment and Labor and employment and labor statistics.

The person in charge terminal 500 receives the new entrant information and the labor cost report information from the device 100 , creates a wage book based on the received new entrant information and the labor cost report information, and reports the labor cost.

Customer terminal 200 , training server 300 , external server 400 , for example, a communicationable desktop computer (desktop computer), laptop computer (laptop computer), notebook (notebook), smart Smart phone, tablet PC, mobile phone, smart watch, smart glass, e-book reader, PMP (portable multimedia player), portable game machine, navigation (navigation) device, digital camera, DMB (digital multimedia broadcasting) player, digital audio recorder, digital audio player, digital video recorder, digital video player (digital video player), PDA (Personal Digital Assistant), and the like.

The apparatus 100 may be a server for providing a labor management service to the customer terminal 200 and the person in charge terminal 500 . The apparatus 100 may provide various services to the customer terminal 200 and the person in charge terminal 500 by operating a labor management service.

Specifically, the device 100 may calculate the possibility of a labor dispute based on the labor contract information, the pay stub information, and the employment information and provide it to the customer terminal 200 .

In addition, the device 100 may calculate an expected amount of damage due to the occurrence of a labor dispute based on the pay stub information and the employment information and provide it to the customer terminal 200 .

Also, the device 100 may calculate an appropriate number of people and an appropriate labor cost based on the customer information and provide it to the customer terminal 200 .

Also, the device 100 may provide the matched government support project and the recommendation degree of the matched government support project to the customer terminal 200 based on the customer information.

Also, the device 100 may provide various statistical data received from the external server 400 and education data received from the education server 300 to the customer terminal 200 .

In addition, the device 100 may request labor cost report information and new manpower information from the customer terminal 200 , and provide the labor cost report information and new manpower information received from the customer terminal 200 to the person in charge terminal 500 . .

FIG. 2 is a block diagram exemplarily showing a functional module of the device according to FIG. 1 ;

Referring to FIG. 2 , the apparatus 100 includes a dispute possibility prediction unit 101 that calculates the possibility of a dispute due to a labor problem based on labor contract information, pay stub information, and employment information.

In addition, the apparatus 100 includes a predicted damage amount predicting unit 102 that calculates an expected damage amount due to a labor dispute based on the pay stub information and the employment information.

In addition, the device 100 specifies at least one first matching company matching the customer terminal 200 from among a plurality of competitors based on the customer information and the competitor information, and the specified at least one first matching company It calculates the average productivity of the company, and includes an appropriate number of people calculation unit 103 for calculating the appropriate number of people based on the customer sales and the calculated average productivity.

In an embodiment, the competitor information includes: a business area for the competitor, the number of floors of the business, floating population of a preset area adjacent to the business site, density of peers in a preset area adjacent to the business site, positive adjacency of a preset area adjacent to the business site It may include the number of industries, sales, number of people, salary information, and industry information. Competitor information may be pre-stored in the database of the device 100 , and may be received from the external server 400 .

In addition, the device calculates appropriate productivity based on customer sales and an appropriate number of people, and at least one second matching company matched with the customer terminal 200 based on customer information, appropriate productivity, competitor information, and competitor productivity and an appropriate labor cost calculation unit 104 for calculating an appropriate labor cost based on the specified labor cost of at least one second matching company.

In addition, the device matches the customer terminal 200 and the government support project based on the customer information, calculates a recommendation degree of the matched government support project, and provides the matched government support project and the recommendation degree to the customer terminal 200 . Includes a government support project matching unit 105 to provide.

In addition, the apparatus includes a statistical data providing unit 106 that processes the data received from the external server 400 according to the customer's needs and provides it to the customer terminal 200 .

In addition, the apparatus includes an education material providing unit 107 that provides the education material received from the education server 300 to the customer terminal 200 .

In addition, the device includes a report management unit 108 that requests labor cost report information and new manpower information from the customer terminal 200 and provides the labor cost report information and new manpower information received from the customer to the person in charge terminal 500 . .

3 to 17, the conflict possibility prediction unit 101, the expected damage prediction unit 102, the appropriate number of people calculation unit 103, the appropriate labor cost calculation unit 104, the government support project matching unit 105 ), the statistical data providing unit 106, the educational data providing unit 107, and the operation process of the report management unit 108 will be described as an example.

FIG. 3 is a diagram conceptually illustrating a process in which the dispute possibility prediction unit 101 according to FIG. 2 calculates an expected value for the possibility of a dispute based on labor contract information.

In order to calculate the expected value of the possibility of dispute based on the labor contract information, the dispute possibility prediction unit 101 may match a plurality of clauses included in the labor contract information with a first clause matching a plurality of preset essential clauses and matching It is classified as Article 2 that does not work.

In one embodiment, the plurality of mandatory provisions include the labor contract period, probation/trial period, reduction ratio/amount/period, work place and duties, prescribed working hours, wages, calculation methods of wages, specific payment methods, etc. may include However, the present invention is not limited thereto, and provisions other than the above-described essential provisions may be included.

The conflict possibility prediction unit 101 may match the clauses included in the labor contract information with a plurality of essential clauses through a first artificial neural network supervised by learning data. The first artificial neural network is a predictive model implemented in software or hardware that mimics the computational power of a biological system using a large number of artificial neurons (or nodes). According to an embodiment, the artificial neural network may be an artificial neural network according to a Recurrent Neural Network (RNN), Long ShortTerm Memory models (LSTM), Gated Recurrent Unit (GRU), or Word2Vec algorithm.

The learning data used for learning of the first artificial neural network consists of a first feature vector (input value) converted from clauses included in the training labor contract information, and a second feature vector (output value) converted from a plurality of preset essential clauses. do. Since supervised learning is performed in a state where the correct answer is known, the first feature vector (input value) and the second feature vector (output value), which are the correct answer relationship, are labeled and used for learning.

The first artificial neural network calculates a degree of similarity between a first feature vector converted from a plurality of clauses included in the labor contract information and a second feature vector converted from a plurality of essential clauses, and a first matched second feature vector based on the calculated similarity. A second clause that does not match clause 1 can be classified.

When the classification of the plurality of clauses is completed, the potential conflict prediction unit 101 may calculate the possibility of invalidation of the first clause through a second artificial neural network supervised by the learning data. . The second artificial neural network is a predictive model implemented in software or hardware that mimics the computational power of a biological system using a large number of artificial neurons (or nodes). According to an embodiment, the artificial neural network may be an artificial neural network according to a Recurrent Neural Network (RNN), Long ShortTerm Memory models (LSTM), Gated Recurrent Unit (GRU), or Word2Vec algorithm.

The training data used for training of the second artificial neural network is composed of a first feature vector (input value) converted from invalid clauses for learning and a second feature vector (output value) converted from a plurality of preset invalid clauses. Since supervised learning is performed in a state where the correct answer is known, the first feature vector (input value) and the second feature vector (output value), which are the correct answer relationship, are labeled and used for learning. In one embodiment, the plurality of preset invalid clauses are, the labor contract period with no time limit, the probation/trial period of the unacceptable range, the reduction ratio/amount/period of the unacceptable range, and It may include changes to the place of work and duties, prescribed working hours within an unacceptable range, basic pay, non-stating legal allowances for overtime/night/holidays, unclear description of wage calculation method, unclear description of payment method, etc. . However, the present invention is not limited thereto, and clauses other than the above-mentioned invalid clauses may be included.

The second artificial neural network calculates a similarity between the first feature vector in which the classified first clause is transformed and the second feature vector in which the plurality of invalid clauses are transformed, and the invalidity of the classified first clause is calculated based on the calculated similarity. possibility can be calculated.

When the possibility of invalidity calculation is completed, the potential dispute prediction unit 101 calculates an expected value for the possibility of disputes in the labor contract based on the labor contract information based on the possibility of invalidity of the first clause, the number of the second clauses, and the weight of the job type. can

In an embodiment, the expected value for the possibility of a labor contract dispute based on the labor contract information may be defined as in Equation 1.

In Equation 1, PL1 represents the expected value of the possibility of a labor contract dispute that will occur based on the labor contract information, and K1 is calculated based on the frequency of labor disputes occurring due to the labor contract information in the customer’s occupation. Shows the weighted occupation. In an embodiment, K1 may be calculated based on the average number of occurrences of labor disputes in the customer's occupation compared to the number of occurrences of labor disputes per year related to the labor contract information. n denotes the number of first clauses and m denotes the number of second clauses. P1k represents the invalidity probability corresponding to each of the clauses classified as the first clause. A higher expected value for the possibility of a labor contract dispute indicates a greater likelihood of a dispute, and a lower expected value for the possibility of a labor contract dispute indicates a lower likelihood of a dispute.

4 is a diagram conceptually illustrating a process in which the dispute possibility prediction unit 101 according to FIG. 2 calculates an expected value for the possibility of a dispute based on pay stub information.

In order to calculate an expected value of the possibility of dispute based on the pay stub information, the dispute possibility predicting unit 101 sets a plurality of items included in the pay stub information to a first item matching a plurality of preset essential specification items and It is classified as a second item that does not match.

In one embodiment, the plurality of mandatory specification items may include basic salary, bonus, performance pay, overtime allowance, night allowance, holiday allowance, etc., salary calculation method, deduction details, and the like. However, the present invention is not limited thereto, and provisions other than the above-mentioned essential specification items may be included.

The dispute possibility prediction unit 101 may match the items included in the pay stub information with a plurality of essential specification items through a third artificial neural network supervised by the learning data. The third artificial neural network is a predictive model implemented in software or hardware that mimics the computational power of a biological system using a large number of artificial neurons (or nodes). According to an embodiment, the artificial neural network may be an artificial neural network according to a Recurrent Neural Network (RNN), Long ShortTerm Memory models (LSTM), Gated Recurrent Unit (GRU), or Word2Vec algorithm.

The learning data used for learning of the third artificial neural network consists of a first feature vector (input value) converted from items included in the learning pay stub information and a second feature vector (output value) converted from a plurality of preset essential specification items. is composed Since supervised learning is performed in a state where the correct answer is known, the first feature vector (input value) and the second feature vector (output value), which are the correct answer relationship, are labeled and used for learning.

The third artificial neural network calculates a similarity between a first feature vector converted from a plurality of items included in the pay stub information and a second feature vector converted from a plurality of essential specification items, and is matched based on the calculated similarity. A second item that does not match the first item may be classified.

When the classification of the plurality of items is completed, the conflict possibility prediction unit 101 may calculate the possibility of violation of the first item through a fourth artificial neural network supervised by the learning data. . The fourth artificial neural network is a predictive model implemented in software or hardware that mimics the computational power of a biological system using a large number of artificial neurons (or nodes). According to an embodiment, the fourth artificial neural network may be an artificial neural network according to a Recurrent Neural Network (RNN), Long ShortTerm Memory models (LSTM), Gated Recurrent Unit (GRU), or Word2Vec algorithm.

The training data used for learning of the fourth artificial neural network is composed of a first feature vector (input value) converted from violation items for learning and a second feature vector (output value) converted from a plurality of preset violation items. Since supervised learning is performed in a state where the correct answer is known, the first feature vector (input value) and the second feature vector (output value), which are the correct answer relationship, are labeled and used for learning. In one embodiment, the plurality of violation items are not specified/misrepresented, such as basic salary that does not exceed the minimum hourly wage, bonus/performance pay/extended allowance/night allowance/holiday allowance, etc. It may include dog/ogi materials, etc. However, the present invention is not limited thereto, and items other than the plurality of violation items described above may be included.

The fourth artificial neural network calculates a similarity between a first feature vector in which the classified first item is converted and a second feature vector in which a plurality of violation items are converted, and violation of the classified first item based on the calculated similarity possibility can be calculated.

When the possibility of violation calculation is completed, the dispute possibility prediction unit 101 calculates an expected value for the possibility of a dispute possibility based on the payslip information based on the possibility of violation of the first item, the number of the second item, and the weight of the job type. can

In an embodiment, an expected value for the possibility of a pay stub dispute based on pay stub information may be defined as Equation (2).

In Equation 2, PL2 represents the expected value of the possibility of a pay stub dispute due to a labor dispute based on the pay stub information, and K2 is calculated based on the frequency of labor disputes occurring due to the labor stub information in the customer's job. Shows the weighted occupation. In an embodiment, K2 may be calculated based on the average number of occurrences of labor disputes in the customer's occupation compared to the number of occurrences of labor disputes per year related to the pay stub information. n represents the number of first items, and m represents the number of second items. P2k indicates the possibility of violation corresponding to each of the items classified as the first item. A higher expected value for the possibility of a payslip dispute indicates a greater likelihood of a dispute, and a lower expected value for the possibility of a payslip dispute indicates a lower likelihood of a dispute.

In an embodiment, the first to fourth artificial neural networks described above may include an input layer, a hidden layer, and an output layer.

In an embodiment, the first to fourth artificial neural networks receive an input value and multiply the connection strength (or weight) for each of the output values of the input layer and the input layer having nodes corresponding to the number of components of the first feature vector, , one or more hidden layers outputting by adding a bias; and an output layer that multiplies each output value of the hidden layer by a connection strength (or a weight) and outputs the result using an activation function. For example, the activation function may be a LeRU function or a Softmax function, but is not limited thereto. Connection strength and bias can be continuously updated by supervised learning.

Specifically, the first to fourth artificial neural networks may be supervised so that an output value of a loss function according to a given input value (first feature vector) and output value (second feature vector) is minimized. For example, the loss function H(Y, Y`) may be defined as in Equation 3 below.

In Equation 3, Y _m may be the m-th component of the second feature vector, and Y′ _m may be the m-th component of the output vector output by receiving the first feature vector from the foreign product recommendation model 105b.

FIG. 5 is a diagram conceptually illustrating a process in which the dispute possibility prediction unit 101 according to FIG. 2 calculates an expected value for the possibility of a dispute based on employment information.

In order to calculate the expected value of the possibility of dispute based on the employment information, the conflict possibility prediction unit 101 includes the age, gender, work type, employment period, monthly working day, monthly working hours, and 4 generations of the employees included in the employment information. Based on whether or not insurance is purchased, the number of uninsured persons corresponding to each of the four major insurances is calculated. The dispute possibility prediction unit 101 compares the pre-stored subscription requirements of each of the four major insurances with the employment information of the employees, searches for the four major insurances that each employee must subscribe to, and does not subscribe corresponding to each of the four major insurances. Calculate the number of people

The possibility of dispute based on employment information may be defined as in Equation (4).

In Equation 4, PL3 represents the expected value of the possibility of four major insurance disputes that will occur based on employment information, and K3 is calculated based on the frequency of occurrence of labor disputes due to employment information in the customer's occupation. Shows occupational weight. In an embodiment, K3 may be calculated based on the average number of occurrences of labor disputes in the customer's occupation compared to the number of occurrences of labor disputes per year related to employment information. WNK represents the number of uninsured persons corresponding to each of the four major insurances, and TN represents the total number of employees hired by the customer. A higher expected value for the possibility of four major insurance disputes indicates a greater likelihood of a dispute, and a lower expected value for the possibility of four major insurance disputes indicates a lower likelihood of a dispute.

FIG. 6 is a diagram conceptually illustrating a process in which the expected damage amount prediction unit 102 according to FIG. 2 derives the expected damage amount based on pay stub information and employment information.

The expected damage prediction unit 102 calculates the expected damage based on the sum of the first expected damage D1 calculated based on the pay stub information and the second expected damage D2 calculated based on the employment information.

The first expected damage D1 calculated based on the pay stub information may be defined by Equation 5 below.

In Equation 5, D1 represents the first expected amount of damage calculated based on the pay stub information, and K2 represents the job type weight calculated based on the frequency of occurrence of labor disputes due to the information on the labor statement in the customer's occupation. In an embodiment, K2 may be calculated based on the average number of occurrences of labor disputes in the customer's occupation compared to the number of occurrences of labor disputes per year related to the pay stub information. n represents the number of the first item. P2k indicates the possibility of violation corresponding to each of the items classified as the first item. PD1k indicates a first damage amount corresponding to each of the items classified as the first item. In an embodiment, the first damage amount may be preset to correspond to each of the items classified as the first item. In one embodiment, the first amount of damage is the age, residential area, gender, The cosine similarity of the vector converted to the occupation may be extracted, and the cosine similarity may be calculated based on the average of the winning amount of the complainant above a preset level.

The second expected damage D2 calculated based on the employment information may be defined by Equation 6 below.

In Equation 6, D2 represents the expected damage amount calculated based on the employment information, and K3 represents the job type weight calculated based on the frequency of occurrence of labor disputes due to the employment information in the customer's occupation. In an embodiment, K3 may be calculated based on the average number of occurrences of labor disputes in the customer's occupation compared to the number of occurrences of labor disputes per year related to employment information. WNK represents the number of uninsured persons corresponding to each of the four major insurances. PD2K represents the second damage amount corresponding to each of the four major insurances. In an embodiment, the second damage amount may be preset to correspond to each of the four major insurances. In one embodiment, the second amount of damage may be determined based on the amount of fine and the number of violations corresponding to each of the four major insurances.

7 is a diagram conceptually illustrating a process in which the appropriate number of people calculating unit 103 and the appropriate labor cost calculating unit 104 according to FIG. 2 calculate the appropriate number of people and the appropriate labor cost.

In one embodiment, the customer information includes the business site area, the number of floors of the business site, the floating population of the preset area adjacent to the business site, the density of similar businesses in the preset area adjacent to the business site, the number of positive neighboring businesses in the preset area adjacent to the business site, sales , the number of people, salary information, industry information, and the like.

In an embodiment, the competitor information includes: a business area for the competitor, the number of floors of the business, floating population of a preset area adjacent to the business site, density of peers in a preset area adjacent to the business site, positive adjacency of a preset area adjacent to the business site It may include the number of industries, sales, number of people, salary information, and industry information.

The appropriate number of people calculating unit 103 calculates a first degree of similarity indicating the similarity between the commercial area and sales between the company operated by the customer and the competitor, and based on the first degree of similarity, at least one similar business area and sales volume to the company operated by the customer determine the first matching company of

The first similarity may be defined by Equation 7 below.

In Equation 7, C1 represents the first degree of similarity showing the similarity of the commercial area and sales between the company operated by the customer and the competitor. TS1 represents the sales of the customer's company, and TS2 represents the sales of the competitor. A represents a vector converted from customer information, and B represents a vector converted from competitor information.

In the customer information and competitor information converted into vectors, the business site area, the number of floors, the floating population in the preset area adjacent to the business site, the density of the same type of business in the preset area adjacent to the business site, and positive neighboring business types in the preset area adjacent to the business site include number. Positive adjacent industries may be set differently depending on the customer's business. A smaller value of the first similarity indicates that the commercial district and sales are similar, and a larger first similarity indicates that the commercial district and sales are different.

The appropriate number of people calculating unit 103 may determine the companies having the first similarity equal to or less than a preset standard as the first matching companies, and calculate the average productivity based on the average of the productivity of the first matching companies. In an embodiment, the average of the productivity of the first matching companies may be the average productivity. The productivity of the first matching company may be a value obtained by dividing the sales of the first matching company by the number of people.

When the average productivity is calculated, the appropriate number of people calculating unit 103 may calculate the appropriate number of people based on a value obtained by dividing the customer's sales by the average productivity. In an embodiment, the appropriate number of people may be determined as a value obtained by dividing a customer's sales by an average productivity.

When the appropriate number of people is calculated, the appropriate labor cost calculation unit 104 calculates the appropriate productivity based on the customer sales and the appropriate number of people. In an embodiment, the appropriate productivity may be determined as a value obtained by dividing customer sales by the appropriate number of people.

When the appropriate productivity is calculated, the appropriate labor cost calculation unit 104 calculates a second degree of similarity indicating the similarity between the commercial area and productivity between the company operated by the customer and the competitor, and based on the second similarity, the business area and the commercial area operated by the customer and at least one second matching company having similar productivity is determined.

The second similarity may be defined by Equation 8 below.

In Equation 8, C2 represents a second degree of similarity showing the similarity of commercial districts and productivity between a company operated by a customer and a competitor. PR1 represents the optimal productivity of the customer, and PR2 represents the productivity of the competitor. A represents a vector converted from customer information, and B represents a vector converted from competitor information.

In the customer information and competitor information converted into vectors, the business site area, the number of floors, the floating population in the preset area adjacent to the business site, the density of the same type of business in the preset area adjacent to the business site, and positive neighboring business types in the preset area adjacent to the business site include number. Positive adjacent industries may be set differently depending on the customer's business. A smaller value of the second similarity indicates that the productivity is similar to the commercial district, and a higher value of the second similarity indicates that the productivity is different from the commercial district.

The appropriate labor cost calculation unit 104 may determine companies having a second degree of similarity equal to or less than a preset standard as the second matching companies, and calculate an average labor cost based on the average of the labor costs of the second matching companies. In an embodiment, an appropriate labor cost may be calculated based on the average of the labor costs of the second matching companies. In an embodiment, the appropriate labor cost may be determined as an average of the labor costs of the second matching companies.

8 is a diagram conceptually illustrating a process of matching the maintenance support project based on customer information by the government support project matching unit according to FIG. 2 .

The government-supported project matching unit 105 matches the customer and the supportable government-supported project based on customer information and pre-stored government-supported project information.

In an embodiment, the customer information may include sales, number of people, salary information, industry information, gender and age of employees, and the like.

In an embodiment, the pre-stored government support project information may include a sales standard, a number of people standard, a salary standard, an industry standard, a gender standard of a person, an age standard of a person, etc. for applying to a government support project.

When a supportable government-supported project is matched, the government-supported project matching unit 105 calculates a recommendation degree of the matched government-supported project.

In one embodiment, the recommendation of the government-supported project is the expected increase in sales based on a value obtained by multiplying the number of additional personnel by the appropriate productivity and employment retention period, government subsidy, and the expected expenditure based on a value obtained by multiplying the number of additional personnel by the appropriate labor cost and employment retention period It can be calculated based on the increase. Specifically, the recommendation degree of the government support project may be determined by dividing the customer's sales by subtracting the expected increase in expenditure from the sum of the expected increase in sales and the government subsidy.

9 is a diagram conceptually illustrating an operation process of the statistical data providing unit 106 according to FIG. 2 .

The statistical data providing unit 106 processes the statistical data provided from the external server 400 and provides it to the customer terminal 200 .

In an embodiment, the external server 400 may provide labor statistics data, such as an average labor cost, an average number of people per sales scale, an average working time of workers, and a labor shortage rate, to the device 100 . In one embodiment, the external server 400 may be a server of a public institution such as the Ministry of Employment and Labor and employment and labor statistics.

In one embodiment, the statistical data providing unit 106 may process the received statistical data to provide unemployment statistics by industry, wage statistics by industry, etc. to the customer terminal 200 .

FIG. 10 is a diagram conceptually illustrating an operation process of the educational material providing unit 107 according to FIG. 2 .

The training data providing unit 107 may receive training materials for statutory compulsory education of employees from the training server 300 , and provide the received training materials to the customer terminal 200 .

11 to 15 , it is a diagram conceptually illustrating an operation process of a labor management service application according to the present invention.

Referring to FIG. 11 , the device 100 may check the name of each company, the business owner, the person in charge, and the received report through the application. Also, the device 100 may transmit a report creation request to at least one of the manager and the holder through the application.

In an embodiment not shown, the customer terminal 200 may transmit a report of the customer's company to the device 100 through the application. In addition, the customer terminal 200 may check the report submitted by the customer through the application.

Referring to FIG. 12 , the device 100 may check information about the customer's name, contact information, owned store, and management store through the application.

Referring to Figure 13, the device 100, the creation date of the report creation request transmitted to the customer terminal 200 through the application, the input date input from the customer, the store name of the customer terminal 200 that requested the report creation, the owner, You can check information about the manager and the base month.

In an embodiment not shown, the customer terminal 200 checks information about the creation date of the report creation request requested by the customer through the application, the input date the customer entered the report, the store name, the owner, the owner, and the base month. can

Referring to FIG. 14 , the device 100 may check information on a report input by the customer through the application. In an embodiment not shown, the customer terminal 200 may check information about the report input by the customer through the application.

Referring to FIG. 14 , the device 100 may check the name of the person in charge, ID, and information on the active state through the application. In an embodiment not shown, the customer terminal 200 may check the name of the person in charge of the customer's business through the application, the ID, and information about the loudspeaker state.

16 and 17 are diagrams conceptually illustrating an operation process of the report management unit according to FIG. 2 .

Referring to FIG. 16 , the report management unit 108 may request the customer terminal 200 to input report information.

Referring to FIG. 17 , the customer terminal 200 receiving the request for input of report information may transmit at least one of new manpower information and report information to the report management unit 108 . In one embodiment, the new manpower information may include the name of the newly joined personnel, a resident registration number, a date of entry, whether to subscribe to four major insurances, and the like. In an embodiment, the report information may include the name of the person, the actual amount of labor cost, and the like.

In an embodiment not shown, the device 100 provides the customer terminal 200 with a function that enables the business owner to prepare and deliver documents that must be legally prepared and delivered online. Also, the device 100 may store documents that the business owner legally must keep for a certain period in the server.

Referring to FIG. 18 , the device 100 includes at least one processor 110 and a memory storing instructions instructing the at least one processor 110 to perform at least one operation. (memory) may be included.

The at least one operation may include the components 101 to 108 of the above-described apparatus 100 or other functions or operation methods.

Here, the at least one processor 110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to embodiments of the present invention are performed. can Each of the memory 120 and the storage device 160 may be configured of at least one of a volatile storage medium and a non-volatile storage medium.

For example, the memory 120 may be one of a read only memory (ROM) and a random access memory (RAM), and the storage device 160 is a flash-memory. , a hard disk drive (HDD), a solid state drive (SSD), or various memory cards (eg, micro SD card).

Also, the device 100 may include a transceiver 130 that performs communication through a wireless network. Also, the device 100 may further include an input interface device 140 , an output interface device 150 , a storage device 160 , and the like. Each of the components included in the device 100 may be connected by a bus 170 to communicate with each other.

For example, the device 100 can communicate with a desktop computer (desktop computer), a laptop computer (laptop computer), a notebook (notebook), a smart phone (smart phone), a tablet PC (tablet PC), a mobile phone (mobile phone) , smart watch, smart glass, e-book reader, PMP (portable multimedia player), portable game console, navigation device, digital camera, DMB (digital multimedia broadcasting) player , a digital audio recorder, a digital audio player, a digital video recorder, a digital video player, a personal digital assistant (PDA), or the like.

The methods according to the present invention may be implemented in the form of program instructions that can be executed by various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the computer-readable medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software.

Examples of computer-readable media may include hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions may include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as at least one software module to perform the operations of the present invention, and vice versa.

In addition, the above-described method or apparatus may be implemented by combining all or part of its configuration or function, or may be implemented separately.

Although the above has been described with reference to the preferred embodiment of the present invention, those skilled in the art can variously modify and change the present invention within the scope without departing from the spirit and scope of the present invention as described in the claims below. You will understand that it can be done.

Claims (5)

A device for providing labor management services, comprising:
a dispute possibility prediction unit for receiving labor information including labor contract information, pay stub information, and employment information from a customer terminal, and calculating an expected value for the possibility of a labor dispute based on the received labor information;
Calculating a first expected amount of damage based on the pay stub information, calculating a second expected amount of damage based on the employment information, and calculating a total estimated amount of damage based on the first and second expected damages damage prediction unit;
receiving customer information from the customer terminal, and determining a plurality of first matching companies matching the customer terminal among a plurality of competitors based on the customer information and pre-stored competitor information, wherein the customer information and the first matching company are determined 1 an appropriate number of people calculating unit that calculates an appropriate number of people based on the average productivity of matching companies; and
A plurality of second matching companies that calculate appropriate productivity based on the customer information and the appropriate number of people, and match the customer terminal based on the customer information, the appropriate productivity, the competitor information, and the productivity of the specified competitor including an appropriate labor cost calculation unit for determining the
The appropriate number of people calculation unit,
calculating a first degree of similarity of each of the competitors, and determining, among the competitors, competitors having the first similarity less than or equal to a preset standard as the first matching companies;
retrieving the sales and number of people previously matched with each of the first matching companies in the database,
determining a value obtained by dividing the sales previously matched with each of the first matching companies by the number of people as the productivity of each of the first matching companies;
Determining the average value of the productivity of each of the first matching companies as the average productivity,
A value obtained by dividing the sales of the customer's company included in the customer information by the average productivity is determined by the appropriate number of people,
The first degree of similarity is calculated based on the following equation,

In the above formula, C1 denotes the first degree of similarity showing the similarity between the commercial area and sales between the company operated by the customer and the competitor, TS1 denotes the sales of the customer's business, and TS2 denotes the sales of the competitor means, A means a vector converted from the customer information, B means a vector converted from the competitor information,
The appropriate labor cost calculation unit,
calculating a second degree of similarity of each of the competitors, and determining, among the competitors, competitors whose second similarity is less than or equal to a preset standard as the second matching companies;
Searching for labor costs matched in advance with each of the second matching companies in the database, and determining the average value of the labor costs of each of the second matching companies as the appropriate labor costs,
The second degree of similarity is calculated based on the following equation,

In the above equation, C2 denotes the second degree of similarity showing the similarity of commercial area and productivity between the company operated by the customer and the competitor, PR1 denotes the appropriate productivity, and PR2 denotes the productivity of the competitor and A means a vector converted from the customer information, B means a vector converted from the competitor information,
In each of the customer information and the competitor information, the business site area for each customer and the competitor, the number of business floors, the floating population in a preset area adjacent to the business site, the density of similar companies in the preset area adjacent to the business site, and a preset adjacent business site including the number of positive contiguous sectors of the domain;
Device. According to claim 1,
The dispute possibility prediction unit,
Classifying an item matching a plurality of preset essential clauses among a plurality of clauses included in the labor contract information into the first clause,
Classifying items that do not match the plurality of essential clauses among the plurality of clauses into a second clause,
calculating an expected value of the possibility of a dispute corresponding to the labor contract information based on the numbers classified into the first and second clauses and the frequency of occurrence of labor disputes;
Device. 3. The method of claim 2,
The expected value for the possibility of a dispute corresponding to the labor contract information is determined by the following formula,

In the above formula, PL1 means the expected value of the possibility of a dispute corresponding to the labor contract information, and K1 means the job type weight calculated based on the frequency of labor disputes occurring due to the labor contract information in the customer's occupation. where n denotes the number of clauses classified as the first clause, m denotes the number of clauses classified as the second clause, and P1k denotes the probability of invalidity corresponding to each of the clauses classified as the first clause means,
Device. delete delete

Images