KR102222991B1 - Method and system for managing specialist pool on specialist platform based on specialist assessment result - Google Patents

Abstract

A method performed by a computer, comprising: collecting information on a plurality of experts, classifying the information on the plurality of experts according to one or more classification criteria, generating one or more expert pools based on the classification result Disclosed is a method for managing an expert pool of an expert platform, including the step of performing and allocating the plurality of experts to the created expert pool.

Description

Method and system for managing the expert pool of expert platform based on the results of expert evaluation {METHOD AND SYSTEM FOR MANAGING SPECIALIST POOL ON SPECIALIST PLATFORM BASED ON SPECIALIST ASSESSMENT RESULT}

The present invention relates to a method and system for managing an expert pool of an expert platform based on a result of expert evaluation.

In the era of the 4th industrial revolution, it is expected that the depth, speed, and scope will be greatly expanded based on super-connectivity and superintelligence. As a result, not only various industrial fields arise, but also numerous new fields such as the fusion of existing industrial fields and new industrial fields with each other, and it is expected that there will be no or very few experts in the field. Therefore, there is a need for a platform where experts in each field can be found and recruited, and experts in each field can cooperate.

In addition, the period of production activities will be extended due to the extended life expectancy in the future, and it is necessary for each expert to make the relevant field a lifelong “business” rather than simply a job.

In addition, it is desirable for experts to be able to act and be treated based on their ability in each field, but in the case of an organization such as an existing company, it is difficult to make use of the individuality of each expert as it pursues the equity between employees and the stability of the entire organization. However, there were problems such as being unable to be treated according to ability, and becoming a manager rather than using it as careers accumulate.

Therefore, it is necessary to ensure the autonomy of experts based on the Gig economy, draw maximum efficiency, but ensure minimum stability and sustainability.

Unexamined Patent Publication No. 10-2010-0026834, published on 2010.03.10

The problem to be solved by the present invention is to provide a method and system for managing the expert pool of the expert platform based on the expert evaluation result.

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

An expert pool management method of an expert platform according to an aspect of the present invention for solving the above-described problem includes the steps of collecting a plurality of expert information for a plurality of experts (S110); Classifying the collected expert information according to at least one classification criterion (S120); Generating at least one expert pool based on the classification criteria (S130); And allocating the plurality of experts to the created expert pool (S140). Includes.

In this case, the step of classifying (S120) may include classifying information on the plurality of experts based on specialized fields, careers, and background information of each of the plurality of experts.

At this time, the step of allocating the experts (S140), with respect to the specialized field corresponding to each of the generated expert pool, evaluating the expertise of the plurality of experts (S210); And determining whether to assign the expert based on the expert evaluation result (S220). It may include.

In this case, the step of evaluating the expertise (S210) may include classifying the collected expert information into structured information and unstructured information (S410); Evaluating the structured information and the unstructured information to obtain weights for the structured information and the unstructured information (S420); Evaluating expertise based on the weight and the structured information and the unstructured information (S430); It may include.

At this time, the collecting step (S110), collecting and obtaining information input from the plurality of experts or crawled information about the plurality of experts;

Classifying expert information corresponding to information stored in the database as structured information based on a previously stored database; And classifying the expert information excluding the structured information into the unstructured information. It may include.

At this time, the generating step (S130), the step of obtaining a weight for the structured information and the unstructured information by evaluating the structured information and the unstructured information (S510); Generating at least one expert pool based on the obtained weight and the expert information (S520); It may include.

In this case, the allocating step (S140), when the expert evaluation result is greater than or equal to a preset value, allocating the expert to an expert pool corresponding to the expert evaluation result (S610); If a plurality of expertise evaluation results corresponding to a plurality of expert pools are equal to or greater than a preset value, allocating the experts to the plurality of expert pools according to the priority of the plurality of expert evaluation results (S620); It may include.

In this case, the step of classifying (S120), clustering the collected plurality of expert information based on an artificial intelligence model (S710); Obtaining at least one cluster based on the clustering result (S720); Obtaining characteristics of the at least one obtained cluster (S730); Including, and generating the expert pool (S130), the step of generating an expert pool corresponding to each of the one or more clusters based on the acquired characteristics (S740); It may include.

An expert pool management method of an expert platform according to an aspect of the present invention for solving the above-described problem includes a memory storing one or more instructions and a processor executing the one or more instructions stored in the memory, wherein the processor is By executing the one or more instructions, the expert pool management method according to the disclosed embodiment is performed.

The expert pool management program of the expert platform according to an aspect of the present invention for solving the above-described problems is combined with a computer as hardware, and a recording medium readable by a computer to perform the expert pool management method according to the disclosed embodiment. Is stored in.

Other specific details of the present invention are included in the detailed description and drawings.

According to the disclosed embodiment, information on a plurality of experts is easily managed using a pool of experts, and it is advantageous in that it can be easily used for matching various organizations and projects such as the government, local governments, public institutions, associations, and companies. have.

In addition, based on the information of each expert, multiple expert pools can be automatically generated, and various criteria for classifying multiple experts can be derived, making it easy to mix and match experts in different fields according to the purpose. & Match).

In addition, there is an advantage of maintaining and developing the expertise of each expert pool by providing necessary training means for experts belonging to each expert pool.

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

1 is a conceptual diagram illustrating an operation of an expert pool management system according to an embodiment.
2 is a flowchart illustrating a method of managing an expert pool according to an exemplary embodiment.
3 is a flowchart illustrating a method for evaluating expert expertise and a method for assigning experts according to an exemplary embodiment.
4 is a flowchart illustrating a method of classifying expert information according to an exemplary embodiment.
5 is a flowchart specifically illustrating a method of evaluating expert expertise according to weights according to an exemplary embodiment.
6 is a flowchart illustrating a method of generating an expert pool using weights.
7 is a flowchart illustrating a method of assigning experts according to an exemplary embodiment.
8 is a flowchart illustrating a method of generating an artificial intelligence-based expert pool according to an exemplary embodiment.
9 is a block diagram of an apparatus according to an exemplary embodiment.

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 together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms. It is provided to fully inform the skilled person of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terms used in the present specification are for describing exemplary embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated in the phrase. As used herein, “comprises” and/or “comprising” do not exclude the presence or addition of one or more other elements other than the mentioned elements. Throughout the specification, the same reference numerals refer to the same elements, and "and/or" includes each and all combinations of one or more of the mentioned elements. Although "first", "second", and the like are used to describe various elements, it goes without saying that these elements are not limited by these terms. These terms are only used to distinguish one component from another component. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical idea of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used with meanings that can be commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not interpreted ideally or excessively unless explicitly defined specifically.

The term "unit" or "module" used in the specification refers to a hardware component such as software, FPGA or ASIC, and the "unit" or "module" performs certain roles. However, "unit" or "module" is not meant to be limited to software or hardware. The “unit” or “module” may be configured to be in an addressable storage medium, or may be configured to reproduce one or more processors. Thus, as an example, "sub" or "module" refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, Includes procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. Components and functions provided within "sub" or "modules" may be combined into a smaller number of components and "sub" or "modules" or into additional components and "sub" or "modules". Can be further separated.

Spatially relative terms "below", "beneath", "lower", "above", "upper", etc. It can be used to easily describe the correlation between a component and other components. Spatially relative terms should be understood as terms including different directions of components during use or operation in addition to the directions shown in the drawings. For example, if a component shown in a drawing is turned over, a component described as "below" or "beneath" of another component will be placed "above" the other component. I can. Accordingly, the exemplary term “below” may include both directions below and above. Components may be oriented in other directions, and thus spatially relative terms may be interpreted according to the orientation.

In the present specification, a computer means all kinds of hardware devices including at least one processor, and may be understood as encompassing a software configuration operating in the corresponding hardware device according to embodiments. For example, the computer may be understood as including all of a smartphone, a tablet PC, a desktop, a laptop, and a user client and application running on each device, and is not limited thereto.

Meanwhile, before describing various embodiments according to the present disclosure, information or terms used in the present disclosure are summarized below.

In the present specification, an expert may be understood to mean all kinds of people having a predetermined expertise in a specific field, and the criteria are not limited thereto. All kinds of experienced persons, trained persons, or talented persons capable of performing work in a specific field can all be referred to as experts, but it is preferably understood to mean persons with a predetermined experience in a specific field. Can be.

Accordingly, experts according to the present disclosure may mean experts who meet the requirements required by a specific group, as well as experts of the existing conceptual, common, and general criteria.

Meanwhile, the term “expert” in the present disclosure refers to an individual individual expert, but is not limited thereto. In other words, a specific group itself can be viewed as an expert.

In particular, an expert according to the present disclosure may mean a person who satisfies various evaluation criteria to be described later.

In this specification, when a specific person produces a specific result in a certain field of work, expert information may mean information related to the produced result.

Therefore, expert information can include all kinds of information that can evaluate the expertise of each expert, and information that can evaluate expert expertise means the relationship between the expert's expertise in a specific area of expertise and a predetermined reference value or more. It may mean information having, but is not limited thereto.

Furthermore, the expert information may mean information on the career of a corresponding expert, and the career may include various information according to embodiments. For example, information such as the workplace where each expert worked, the academic background of each expert, the thesis published by each expert, and the project each expert participated in may be included. Furthermore, the record of each expert's activities in online communities or clubs, etc. It can be understood to mean that it encompasses all various formal and unstructured information, such as projects performed as a hobby or material created as a hobby.

Meanwhile, in the present specification, expert information may be divided into structured information and unstructured information.

In an embodiment, the formal information may refer to information (or information that can be quantified) that has already been quantitatively proven to have a relationship with the expertise of a specific specialty field. For example, as described above, objective data that can prove an expert may correspond to formal information.

For example, the formal information may include, but is not limited to, experience working in a company or department in a specific field, experience in participating in a project or assignment in a specific field, academic background in a specific field, or thesis presentation information, etc.

In an embodiment, the unstructured information does not belong to the structured information, but may mean all types of career information having a relationship with a specific specialized field. For example, as described above, career information according to individual efforts may belong to unstructured information.

For example, the unstructured information may include, but is not limited to, records of activities in online communities or clubs, materials created for personal projects or hobbies, and information on taking online educational content.

In an embodiment, expert information on a new specialized field in which the previously structured information has not been stored may be obtained. In this case, since no formal information, which is a criterion for evaluating the relevant specialized field, is not stored at all, all expert information can be evaluated as unstructured information.

For example, if there is no pre-evaluated or stored information for the relevant specialty even if it is information that has worked in a job or department in a specific specialty field, or even information that has performed a project or task, it is evaluated as unstructured information and It can be used to assess the expertise of each expert.

In addition, expert information may include advisory (committee) activities, screening and evaluator activities, lectures and lecture activities, mentoring activities, and outside experience, etc., which may belong to formal or unstructured information. For example, the classification criterion may be whether or not a database has been established by quantitatively verifying the relationship with the expertise of a specific specialized field, but is not limited thereto.

In addition, data that can verify the morality of experts may also belong to expert information.For example, various information such as awareness and reputation of activities in the relevant industry or industry, SNS, etc., rebates, criminal records, and fraud records can be used. In addition, even if all the information is not available in reality, information that can be collected by being disclosed to the outside can be used as expert information classified as formal information or unstructured information.

Meanwhile, some of the above-described structured information and unstructured information may be difficult to quantify as data. When it is difficult to quantify data, it may mean that data cannot be quantified because the standard is not established. In this case, it goes without saying that data that are difficult to quantify can be directly input based on the surrounding evaluation and can be quantified through various types of algorithms.

Meanwhile, in the present disclosure, a weight means a weight obtained in relation to expert information. In this case, expressions that a weight is assigned (to expert information), that a weight is matched (to expert information), or that a weight is added (to expert information) can all be interpreted in a similar meaning.

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

Each of the steps described herein is described as being performed by a computer, but the subject of each step is not limited thereto, and at least some of the steps may be performed by different devices according to embodiments.

1 is a conceptual diagram illustrating an operation of an expert pool management system according to an embodiment.

Each component shown in FIG. 1 does not necessarily represent a specific device or module, and may mean a database including data and data, or a step corresponding to a specific operation, and is not limited thereto.

According to the system 1 of the disclosed embodiment, expert information 10 is disclosed. The expert information 10 includes information on an expert in a specific field, and may include, for example, career information of an expert, but is not limited thereto. Expert information can include all kinds of information that can evaluate the expertise of each expert, and the information that can evaluate expert expertise means information that has a correlation with the expert's expertise in a specific field of expertise and a predetermined reference value or more. It can be, but is not limited thereto.

In recent years, due to the development of online communities, etc., there are many cases where projects are carried out alone or by several people outside of the existing regular educational institutions or workplaces, and as the information is relatively easy to access, the information imbalance has been resolved. Even if they are not educated or experienced experts, many geek experts at the amateur level or beyond the level of amateurs have arisen, and their expertise is sometimes compared to or rather beyond the traditional experts.

Also, traditional experts and geek experts are not dichotomous, and traditional experts may also have a geek expert element, and traditional experts in a specific field may be recognized as geek experts in other fields or similar fields. have.

In particular, in the era of the 4th industrial revolution, the depth, speed, and range are expected to be greatly expanded based on hyperconnectivity and superintelligence. As a result, not only various industrial fields arise, but also numerous new fields such as the fusion of existing industrial fields and new industrial fields with each other, and it is expected that there will be no or very few experts in the field. Therefore, there is a need for a platform where experts in each field can be found and recruited, and experts in each field can cooperate.

In addition, the period of production activity will be extended due to the extended life expectancy in the future, and there is a need for each expert to make the field a lifelong "business" rather than simply a job.

In addition, it is desirable for experts to be able to act and be treated based on their ability in each field, but in the case of an organization such as an existing company, it is difficult to make use of the individuality of each expert as it pursues the equity between employees and the stability of the entire organization. However, there were problems such as being unable to be treated according to ability, and becoming a manager rather than using it as careers accumulate.

Therefore, it is necessary to ensure the autonomy of experts based on the Gig economy, draw maximum efficiency, but ensure minimum stability and sustainability.

The platform built on this basis must have an infrastructure to discover, recruit, and educate experts, provide work and generate profits, and an expert ecosystem that enables experts to network, educate, grow, and collaborate with each other. Building it is more important than anything else.

In order to build such an expert ecosystem, an expert pool that classifies each expert is created, experts are assigned to the generated expert pool, and furthermore, the assigned expert is verified, and the level of expertise of each expert is evaluated to play an appropriate role. It is important to give.

Therefore, a method of determining whether a specific person is an expert or not, determining whether or not a specific person is an expert, or determining whether he or she is an expert in a certain field, and assigning it to an appropriate expert pool has become more necessary.

In an embodiment disclosed in the present specification, an expert may mean an expert in the MICE field. MICE is a term after English letters such as Meeting, Incentive trip, Convention, Exhibition, Exhibition & Event, etc.In a narrow sense, it is mainly based on international conferences and exhibitions. It refers to a promising industry and, in a broader concept, refers to a convergence industry that includes reward tourism and mega events centered on participants.

Since MICE is a convergence industry in various fields, experts in various fields are required, and collaboration between these experts is required. Accordingly, the expert platform and the expert pool management method for the expert platform according to the disclosed embodiment may be applied to MICE experts, but are not limited thereto.

Therefore, the system 1 according to the disclosed embodiment extracts various structured and unstructured information from the expert information 10 using the model 30 that is learned or constructed based on the previously stored database 20, as well as It is structured so that experts in different fields can be classified, and various career information, which was previously difficult to be included as a career, is also constructed so that it can be used in various ways to prove expertise in the field.

In addition, the expert pool classification system is operated based on the classification result 40, but if the classification result for the expert classification is actually obtained, a feedback 50 is generated based on this, and the classification of the corresponding expert can be updated accordingly. In addition, it is possible to update the model 30 based on this.

2 is a flowchart illustrating a method of managing an expert pool according to an exemplary embodiment.

Each of the steps described herein is described as being performed by a computer, but the subject of each step is not limited thereto, and at least some of the steps may be performed by different devices according to embodiments.

In the present specification, a computer means all kinds of hardware devices including at least one processor, and may be understood as encompassing a software configuration operating in the corresponding hardware device according to embodiments. For example, the computer may be understood as including all of a smartphone, a tablet PC, a desktop, a laptop, and a user client and application running on each device, and is not limited thereto.

In step S110, the computer collects a plurality of expert information about the plurality of experts.

In the present specification, an expert may be understood to mean all kinds of people having a predetermined expertise in a specific field, and the criteria are not limited thereto. All kinds of experienced persons, trained persons, or talented persons capable of performing work in a specific field can all be referred to as experts, but it is preferably understood to mean persons with a predetermined experience in a specific field. Can be.

Accordingly, expert information may mean information on the career of a corresponding expert, and the career may include various information according to embodiments. For example, information such as the workplace where each expert worked, the academic background of each expert, the thesis published by each expert, and the project each expert participated in may be included. Furthermore, the record of each expert's activities in online communities or clubs, etc. It can be understood to mean that it encompasses all various formal and unstructured information, such as projects performed as a hobby or material created as a hobby.

In step S120, the computer classifies the plurality of expert information according to one or more classification criteria.

In an embodiment, the computer may classify information on the plurality of experts based on specialty fields, careers, and background information of each of the plurality of experts.

In an embodiment, the classification criterion may be classified based on formal information among expert information acquired in step S110. In this case, even if the information is structured, the importance of the information may be different. When a weight is obtained for the formal information, the computer may classify the information on the expert based on the formal information in which the weight is equal to or greater than a preset value.

For example, when the structured information for classification is a specialized field of an expert, the computer may classify the expert information by acquiring structured and unstructured data related to the specialized field. That is, the computer may obtain a category for a specialized field (for example, a legal category, an accounting category, etc.), and determine which category of the acquired category the collected expert information is related to classify the expert information. .

In an embodiment, when there are two or more formal information for classification, the computer may classify expert information in various ways. When the formal information for classification is specialized field and career information, the computer may classify expert information in consideration of both the specialized field and career. In this case, more detailed classification may be possible than when there is only one formal information for classification. For example, if the formal information for classification is one specialized field, the category is limited to a specialized field such as a legal category and an accounting category, but if the formal information for classification is specialized field and career information, the category is a legal field. Expert information can be classified according to subdivided categories such as experts with less than 10 years in the field of law, experts with more than 10 years in the field of law, experts with less than 10 years in the field of accounting, and experts with more than 10 years in the field of accounting.

That is, the computer can classify the expert pool according to various criteria as needed.

In one embodiment, if the expert pool according to the present invention is a pool of experts for classifying MICE industry experts, the computer is used for meetings, incentive trips, conventions, exhibitions and events (Exhibition & Event). ), you can classify the expert pool based on each industry field.

In another embodiment, the computer may classify a pool of experts by functional field such as planning, directing, production, operation, sales, and administration.

In another embodiment, the computer may classify the expert pool according to the degree of experience retention of experts according to the type of project. Specifically, even if an expert with the same experience in the same field, there may be a difference in project execution experience (know-how). For example, even a planning or directing expert of the same annual project experience in domestic or overseas projects and project execution experiences for each organizer or orderer such as government, public institutions, local governments, companies, etc. It is possible to have different degrees of know-how depending on the experience of performing projects by scale, such as billions of dollars. Thus, the computer can classify the expert pool based on the degree of expert experience.

In another embodiment, the computer may classify the expert pool according to the level of the expert's reputation. Expert information on the reputation of an expert may be obtained through various information such as awareness of activities in the industry or industry, SNS, etc., rebates, criminal records, and fraud records, or through letters of recommendation and supporting documents for the expert.

It goes without saying that the various embodiments described above are not individually used for classifying expert pools, but may be used in combination. For example, it goes without saying that the computer can classify the expert pool by using at least two criteria among experts' field, function, experience and possession, and reputation.

In step S130, the computer generates one or more expert pools based on the classification result.

In step S140, the computer allocates the plurality of experts to the generated expert pool.

Specifically, the computer may include acquiring an expert pool having the highest correlation with the expert, and determining whether to assign an expert according to a degree of expertise set in the acquired expert pool.

In an embodiment, the computer may extract expert information from the expert, compare the extracted expert information with a plurality of expert pools, and allocate the expert to the expert pool having the highest degree of association.

Meanwhile, in the method of allocating experts, the computer may evaluate the expertise of each expert and assign the expert to the expert pool only when the evaluated expertise is at least a certain level. For example, even if you are an expert in the field of law, some experts may have only completed a law degree (the result of the expert evaluation will be low), while others may have a variety of practical experience (the result of the expert evaluation will be high). . Thus, the computer can decide whether to allocate a specific expert to the expert pool based on the evaluated expertise.

In this case, the computer may match a plurality of experts assigned to each expert pool, allocate a project or task, and perform matching with a recruiting company according to an embodiment.

3 is a flowchart illustrating a method for evaluating expert expertise and a method for assigning experts according to an exemplary embodiment.

In step S210, the computer may evaluate the expertise of the plurality of experts with respect to specialized fields corresponding to each of the generated expert pools.

In evaluating the professionalism of an expert, in the past, an already certified expert was used to evaluate the expertise through documents or interviews, or to prove the expertise through a predetermined test. For example, in evaluating experts, it has been common to evaluate their expertise through documents such as qualifications, certificates of completion, and diplomas, and tests such as evaluation and interview screening.

However, in this case, there is no means for evaluating the expertise in heterogeneous fields that do not have enough information and manpower to evaluate the expertise, and it is difficult to objectively and accurately evaluate the expertise, and the expertise of each expert updated in real time every time. There is a problem in that it is difficult to evaluate, and it is difficult to provide organic feedback even when there is an incorrect evaluation.

To evaluate the expertise of various experts, it is necessary to classify expert information according to its nature.

In an embodiment, the expert information may include information that has already been quantitatively proven to have a relationship with the expertise of a specific specialty field. For example, this may be objective data that can prove an expert.

For example, expert information may include, but is not limited to, experience working in a company or department in a specific field, experience in participating in a project or assignment in a specific field, academic background in a specific field, or thesis presentation information, etc.

In addition, expert information may include advisory (committee) activities, screening and evaluator activities, lectures and lectures, mentoring experience, outside experience, and awards.

In addition, data that can verify the morality of experts may also belong to expert information.For example, various information such as awareness and reputation of activities in the relevant industry or industry, SNS, etc., rebates, criminal records, and fraud records can be used. And, even if all the information is not available in reality, any information that can be collected by being disclosed to the outside can be used for verification by experts as well.

In addition, information related to the current affiliation of the expert may also belong to expert information. For example, information on whether an expert is a freelancer or an expert belonging to a specific group may also be included in the expert information. When an expert is included in a specific group, the expert can be verified by considering both the expert and the group to which the expert belongs. At this time, whether or not the expert is a freelancer can be checked through the four major insurance coverage. However, it is not limited thereto, and of course, it is possible to directly receive input from each expert or a specific group.

That is, the expert verification method may include the aspect of verifying the expert's expertise, may include the aspect of verifying the authenticity of expert information, and may include the aspect of making an overall evaluation of the expert.

In step S220, the computer may determine whether to assign the expert based on the expert evaluation result.

In an embodiment, the computer may classify expert information into structured information and unstructured information, and may evaluate each of the classified structured information and unstructured information. The computer can evaluate the expert's expertise based on the evaluation results of both structured and unstructured information.

Once the expert's expertise is evaluated, the computer can determine the level of expertise set in the expert pool.

In other words, even if a specific person has information on a specific field, if he does not have expertise, he should not enter the expert pool. For example, a law student may be classified as an expert in the field of law according to the expert pool generated in FIG. 2, but since they do not have a certain level of expertise, they should not be assigned to the expert pool. Accordingly, the computer may evaluate the expertise of the expert and assign the expert to the expert pool only when the expertise is equal to or greater than a preset value.

At this time, the degree of expertise (higher level) that the expert pool must have can be determined through various methods.

In one embodiment, the degree of expertise may be directly input by the user. However, since the measure of the degree of professionalism is specifically quantified and not an intuitively understandable measure, computers are based on the expertise of multiple experts associated with a specific expert pool. The expertise of the company can be classified and provided. For example, assuming that 100 experts are involved in the first expert pool, the computer uses the top 5% experts as the top experts, the top 15% experts as the advanced experts, and the top 80% experts. Can be set as a general expert. In this case, the remaining 20% of experts may not be assigned to the expert pool.

In another embodiment, the degree of expertise may be determined in relation to the difficulty of the task to be performed. In other words, the experts assigned to the expert pool are not fixed, but can be flexibly configured according to individual tasks. Considering that the purpose of establishing an expert pool is to match experts, for tasks with a clear direction and purpose, it is necessary to provide an expert pool suitable for the task. Therefore, the computer is provided with information on the task to be provided (e.g., field of work, order amount, task category, level of expertise required, quality of previous work results, etc.), and the expertise to be applied to the pool of experts based on the degree of work provided. You can determine the degree of.

The degree of expertise described above may be directly set by the system 1, but is not limited thereto. In other words, the level of expertise can be directly set by users of the platform requiring experts (for example, various organizations such as government, local governments, public institutions, associations, companies, and experts who need experts for collaboration). Of course.

4 is a flowchart illustrating a method of classifying expert information according to an exemplary embodiment.

Specifically, FIG. 4 is for explaining a specific method of classifying expert information for evaluating expert expertise into structured information and unstructured information.

In step S310, the computer may acquire a plurality of expert information.

As described above, expert information refers to a plurality of expert information about a plurality of experts.

In step S320, the computer may classify the collected expert information into structured information and unstructured information based on the database.

Specifically, in classifying expert information, the computer may classify expert information corresponding to the information stored in the database as structured information based on a previously stored database.

In an embodiment, information on each specialty field and career information corresponding to each specialty field (ie, career information having a correlation with the corresponding specialty field) may be stored in the database in advance.

The computer may use the database to determine that information already stored in the database is formal information. In an embodiment, the computer may perform a process of matching information stored in a corresponding database with expert information. For example, since expert information entered or crawled by an expert and information previously stored in the database may not exactly match, a process of determining whether or not structured information may be performed by matching similar information with each other. For example, it may be determined that information stored in the database and information showing a matching rate equal to or greater than a predetermined reference value are formal information.

In an embodiment, the computer may filter the structured information in the input step by displaying the matching database information in the input step by the expert and allowing the expert to directly match the corresponding database information.

In an embodiment, the computer may classify expert information excluding the structured information classified based on the database as unstructured information.

That is, information that does not match information previously stored in the database may be classified as unstructured information, and model-based expertise evaluation may be performed.

In an embodiment, a plurality of databases may be hierarchically configured. For example, information on each specialized field and career information corresponding to each specialized field are stored in the first database, and expert information corresponding to the information stored in the first database may be classified as formal information. .

In addition, information on each specialized field and career information corresponding to each specialized field may be stored in the second database, but expert information corresponding to the information stored in the database is classified as unstructured information, and the first data Information that is not stored in both the base and the second database can be classified as unstructured information likewise.

In addition, according to an embodiment, expert information corresponding to information stored in the second database is classified as unstructured information, but information stored in the second database may be used for evaluation of the corresponding unstructured information.

That is, the information stored in the second database may correspond to an intermediate step between the unstructured information and the structured information, but the information stored in the second database may not be evaluated as structured information. However, this only corresponds to an exemplary embodiment, and the configuration of the database and a method of using the database are not limited.

5 is a flowchart specifically illustrating a method of evaluating expert expertise according to weights according to an exemplary embodiment.

In step S410, the computer may classify the collected expert information into structured information and unstructured information. According to the various methods described above, the computer may classify the collected expert information into structured information and unstructured information.

In step S420, the computer may acquire weights for the structured information and the unstructured information.

Specifically, the computer may acquire a high weight for important or meaningful information and a low weight for less important or less meaningful information.

For example, the computer may set the weight of the structured information higher than the weight of the unstructured information. Alternatively, the computer may set the weight of the information stored in the first database to be higher than the weight of the information stored in the second database.

As another example, the computer may set different weights according to the expert pool to which even the same information belongs. For example, the weight for information related to the'law degree' may be set higher in the pool of law experts, but the weight may be set lower in the pool of accounting experts than in the pool of law experts.

On the other hand, it goes without saying that even the same information may change the weight according to circumstances. In an embodiment, the computer may obtain different weights depending on when expert information is generated or updated. That is, expert information that has been collected but has not been updated after a time elapses may be obtained with a low weight.

In another embodiment, the weight may be changed according to other expert information associated with even the same information. For example, when the information for which the weight is to be obtained is career information, the weight may be obtained differently according to the career disconnection period between the career and the career. For example, if the number of career interruptions is more than a preset number or the career interruption period is longer than a preset period, the computer may set a lower weight for the corresponding career.

In step S430, the computer may evaluate the expert's expertise.

Professional evaluation can be performed on the basis of both formal and informal information. In an embodiment, the computer may evaluate the structured information and the unstructured information, and may evaluate expertise based on the evaluated structured information and the unstructured information. In this case, the evaluation of professionalism may be evaluated using the weight added to the structured information and the unstructured information.

Structured and unstructured information can be evaluated in a variety of ways. In one embodiment, the computer may evaluate the formal information based on the evaluation information stored in the database. For example, expertise evaluation information corresponding to each structured information may already be stored in the database, and the computer may quantitatively calculate a score corresponding to each structured information based on the information stored in the database.

For example, in the database, information on the degree of expertise can be recognized based on the relationship between specific career information for a specific field of expertise, the period of the relevant career information, and the role played by the expert in the relevant career information. Can be saved in advance.

Information stored in advance may be input by an expert in the relevant field, or may be determined based on information calculated based on previously collected big data. For example, if there are a lot of reference data for the relevant field, you can use it to build big data, and use learning methods such as machine learning to determine the correlation between each field and career information. According to detailed information, information that can quantitatively evaluate professionalism can be obtained.

Such information may be stored in advance in a database and used to evaluate expert information classified as formal information.

Meanwhile, computers can evaluate unstructured information through various methods. In one embodiment, in evaluating the unstructured information, the computer may estimate the expertise of the expert in the specialized field from the unstructured information by using a model trained to derive a relationship between the career information and the expertise of each specialized field. have.

For example, the model 30 as shown in FIG. 1 may be a model learned using a machine learning method, but is not limited thereto.

The trained model may be a model trained to derive a relationship between specific career information and expertise in each specialized field, and estimate the expertise of a corresponding expert from unstructured information.

In an embodiment, when new unstructured information is acquired for a specific specialized field previously stored in a database, the learned model may estimate a correlation between the unstructured information and the specialized field.

In addition, the trained model can estimate the correlation between a new specialty field that is not stored in the database and new unstructured information of the relevant specialty field.

In an embodiment, the trained model may be trained based on information stored in a database. The computer can generate a variety of learning data based on the information stored in the database.For example, a special field previously stored in the database can be used to estimate the association between a specific specialty field previously stored in the database and new unstructured information. It is possible to generate training data including career information for which the association is previously labeled, and train a model based on this.

As another example, in order to estimate the association between a specific specialty field not previously stored in the database and new unstructured information, feature information extracted for a specific specialty field, feature information extracted for specific career information, and each feature information It is possible to generate training data labeled with correlation in advance, and train a model based on this.

In conclusion, the computer can evaluate the expertise of each expert by classifying the expert's structured and unstructured information using the database and the model learned based on the database, and evaluating the expertise of each classified information. .

As described above, the evaluated expertise can be used when assigning experts to the expert pool.

6 is a flowchart illustrating a method of generating an expert pool using weights.

As described above, the weights added to the structured information and the unstructured information may be used for evaluation of expertise, but, of course, they may be used to create a pool of experts as described below.

In step S510, the computer may acquire weights for the structured information and the unstructured information. In an embodiment, weights for structured information and unstructured information may be obtained by using the method described with reference to FIG. 5.

In step S520, the computer may generate an expert pool based on the weight and expert information.

In an embodiment, the computer may generate an expert pool using only expert information having a weight equal to or greater than a preset value. Specifically, the computer classifies expert information into structured and unstructured information, acquires weights for the classified structured and unstructured information, and is structured information whose weight value is equal to or greater than a preset value among the structured and unstructured information obtained by weight. And unstructured information can be selected. The computer may generate an expert pool by classifying only the selected structured information and unstructured information according to classification criteria.

That is, even if it is expert information collected directly by an expert or through crawling, some information may be information that is not useful. Therefore, before the step of creating an expert pool by classifying expert information, meaningful information can be obtained, and an expert pool can be created with only meaningful information.

In another embodiment, the computer may generate an expert pool based on a result of multiplying the weight obtained for each expert information by the corresponding expert information (ie, expert information reflecting the weight). That is, the computer creates an expert pool using all the collected expert information, but can adjust the importance contributing to the expert pool creation by taking weights into account.

7 is a flowchart illustrating a method of assigning experts according to an exemplary embodiment.

In step S610, the computer may allocate an expert according to the expert evaluation result. As described above, experts with a predetermined level or higher may be allocated to a suitable expert pool.

In step S620, the computer may allocate experts to a plurality of expert pools. That is, it goes without saying that one expert does not belong to one expert pool, but may belong to a plurality of expert pools.

Specifically, in order for a specific expert to be assigned to the expert pool, it is necessary to provide expert information. As described above, expert information may be directly input from an expert or obtained by collecting public information.

In this case, the computer may compare expert information on a specific expert with expert information included in the generated plurality of expert pools, and allocate experts to the plurality of expert pools.

Specifically, when a plurality of expert evaluation results corresponding to the plurality of expert pools are equal to or greater than a preset value, the computer may allocate experts to the plurality of expert pools according to the priority of the plurality of expert evaluation results.

That is, the expert may have both the first expertise evaluation result for the first expert pool and the second expertise evaluation result for the second expert pool equal to or greater than a preset value. In this case, the experts may be assigned to both the first expert pool and the second expert pool. However, the computer may determine a priority by comparing the results of the first expertise evaluation and the results of the second expertise evaluation, and allocate experts to the expert pool according to the priority. That is, even when experts are assigned to a plurality of expert pools, the computer may preferentially allocate experts to expert pools with higher expertise.

8 is a flowchart illustrating a method of generating an artificial intelligence-based expert pool according to an exemplary embodiment.

In step S710, the computer may cluster the collected information on the plurality of experts based on the artificial intelligence model. At this time, clustering may be achieved through various methods. According to various embodiments, various techniques such as K-mean clustering, Mean-Shift clustering, Density-Based Spatial clustering, clustering using Gaussian Mixture Models, and Agglomerative Hierarchical clustering techniques may be applied to the clustering according to various embodiments.

In an embodiment, the computer may cluster expert information using a K-mean clustering algorithm. Specifically, 1) When the set of expert information (D) and the number of clusters (k) are input, the computer randomly extracts k objects from the set of expert information D, and extracts k pieces of expert information of each cluster. It can be set as a centroid. 2) The computer can acquire k clusters based on the center point of the cluster. Specifically, the computer obtains a distance for each expert information included in the expert information set (D) and each of the k pieces of expert information extracted, determines which center point and the center point of each expert information has the highest similarity, and establishes a cluster. Can be obtained. 3) The computer can reset the center point of the cluster based on the acquired expert information included in the cluster, and reset k clusters based on the reset center point. That is, the computer obtains a cluster from an arbitrary center point (k center points), determines a new center point by recalculating the center point of the obtained cluster, and obtains a new cluster from the new center point (k center points). You can repeat until the center point of does not change. Through the above method, the computer can obtain a cluster for the input data. Meanwhile, in the above-described embodiment, a method of obtaining a cluster through the K-means clustering algorithm has been described, but it is obvious that the cluster can be obtained by various other methods such as the GMM algorithm.

In step S720, the computer may acquire one or more clusters based on the clustering result.

In an embodiment, clusters having similar characteristics may be located closer to each other. Accordingly, the computer can determine the degree of association between each cluster based on the distance. Specifically, the computer may obtain correlation information between clusters by analyzing the distance between the center points of each cluster.

In step S730, the computer may include acquiring the acquired characteristics of the at least one cluster.

Specifically, the computer can label the acquired clusters. That is, the cluster created by the above-described clustering process is only a collection of data having similar properties, and there is no information on which expert field it is a pool of experts. Accordingly, the computer may analyze the cluster generated by clustering and perform a labeling process corresponding to the expert pool.

In an embodiment, the computer may determine characteristics such as specialty fields, careers, and background information of experts corresponding to each cluster.

In step S740, the computer may generate an expert pool corresponding to each of the one or more clusters based on the acquired characteristics.

Unlike the traditional method of classifying experts by specialty or career, the expert pools generated from this include expert pools classified based on various parameters based on an artificial intelligence model. In this case, one expert is divided into a plurality of expert pools. If there are parts that may belong to each other and that can be fused to each other in different specialization fields, one expert pool including all of them may be created, or an expert pool in which parts overlapping each other (intersection) may be created.

Therefore, according to the purpose, information on experts belonging to different technical fields but mutually common or capable of cooperating can be easily obtained based on the expert pool so that different technical fields can be fused, and matching accordingly. By doing this, it has the effect of enabling a mix & match of various experts who can respond to the era of the 4th industrial revolution.

Meanwhile, when experts are assigned to the expert pool, the computer can allocate tasks to experts based on the expert pool. In one embodiment, the computer may analyze a task to be assigned to select an appropriate expert pool, and assign the task to an expert judged to be best able to perform the task to be assigned among experts included in the expert pool. Specifically, the computer extracts expert information necessary for performing a task, compares the extracted expert information with expert information of each expert included in the expert pool to obtain an appropriate expert, and can allocate the job to the acquired expert. .

Expertise information evaluated for each expert may include information evaluated based on structured information and information estimated based on unstructured information, and thus the actual expertise and error of each expert may occur. In addition, the expert's expertise may be inferior to the career information of each expert, and there may be omissions in the career information, or there may be areas of expertise that cannot be expressed as career information. It could be high.

When a task is assigned to an expert and the expert processes the assigned task, the computer can obtain the task processing result of the expert. In an embodiment, the result of work processing may quantitatively evaluate the result of work performance, or the result of work performance may be qualitatively evaluated by other experts.

The computer may generate feedback on the expert's expertise evaluation result based on the result of the work process, and update the expert's expertise based on the generated feedback.

The computer can update the expert pool based on the expert evaluation results updated for each expert.

By updating the expert's expert evaluation results, it has the effect of systematically managing the experts assigned to the expert pool. For example, if a result of the expert assessment result is inappropriate for the expert pool, the computer may exclude the expert from the expert pool. Conversely, even if the result of the evaluation of expertise was initially low and did not belong to the pool of experts, it may be allocated to the pool of experts as the result of the evaluation of expertise increases.

Meanwhile, in the step of updating the expertise of the expert, the computer may evaluate the estimation result of the learned model based on the feedback.

For example, the expert's expertise may be estimated from unstructured information by using the model learned as described above, and this may be evaluated based on the expert's actual job performance result. The estimated expertise may be overestimated or underestimated, or it may be appropriately estimated.

Also, the computer may update the learned model based on the evaluation information on the estimation result.

The computer may update the trained model by using the unstructured information of the corresponding expert and the information including the actual job performance result as new training data, or by performing training by adding to the existing training data.

Likewise, based on the estimation result and feedback thereon, the computer can also update the model learned through reinforcement learning.

In addition, in the process of updating the trained model described above, when the association between the first unstructured information and the first specialized field exceeds a preset reference value according to the result of updating the learned model, the computer When information on the field is not stored in the database, information on the first specialized field may be stored in the database.

In addition, when the association between the first unstructured information and the first specialized field exceeds a preset reference value according to the result of updating the learned model, the computer stores the first unstructured information for the first specialized field. It can be stored in the database as information.

In other words, even if it is unstructured information, if references related to the unstructured information are sufficiently accumulated and the association with a specific specialty field can be confirmed by more than a preset reference value, the relevant unstructured information and information on the corresponding specialty field are added to the database. By doing so, the unstructured information can be registered as structured information.

For example, a specific online community activity may be initially treated as unstructured information, but if the expertise of experts actively active in the corresponding online community is proven, the activity information of the corresponding online community may be registered as formal information.

In addition, even in the case of a specialized field that is not previously stored in the database, information on the relevant field and the corresponding unstructured information are used as structured information in the case of securing a correlation between one or more unstructured information and a predetermined reference value or more. It is possible to register at.

9 is a block diagram of an apparatus according to an exemplary embodiment.

The processor 102 may include one or more cores (not shown) and a graphic processing unit (not shown) and/or a connection path (eg, a bus) for transmitting and receiving signals with other components. .

The processor 102 according to an embodiment executes one or more instructions stored in the memory 104 to perform the method described with respect to FIGS. 1 to 8.

For example, the processor 102 acquires new training data by executing one or more instructions stored in the memory, performs a test on the acquired new training data using the learned model, and labels the test result. Extracting first learning data in which the obtained information is obtained with an accuracy equal to or higher than a predetermined first reference value, deleting the extracted first learning data from the new learning data, and removing the new learning data from which the extracted learning data is deleted. The learned model can be retrained by using.

Meanwhile, the processor 102 temporarily and/or permanently stores a signal (or data) processed inside the processor 102, a RAM (Random Access Memory, not shown) and a ROM (Read-Only Memory). , Not shown) may further include. In addition, the processor 102 may be implemented in the form of a system on chip (SoC) including at least one of a graphic processing unit, RAM, and ROM.

The memory 104 may store programs (one or more instructions) for processing and controlling the processor 102. Programs stored in the memory 104 may be divided into a plurality of modules according to functions.

The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. Software modules include Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

Components of the present invention may be implemented as a program (or application) and stored in a medium to be executed by being combined with a computer that is hardware. Components of the present invention may be implemented as software programming or software elements, and similarly, embodiments include various algorithms implemented with a combination of data structures, processes, routines or other programming elements, including C, C++. , Java, assembler, or the like may be implemented in a programming or scripting language. Functional aspects can be implemented with an algorithm running on one or more processors.

In the above, embodiments of the present invention have been described with reference to the accompanying drawings, but those skilled in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features. You will be able to understand. Therefore, the embodiments described above are illustrative in all respects, and should be understood as non-limiting.

Claims (1)

In the method performed by a computer comprising a processor and a memory,
Collecting, by the processor, a plurality of expert information about a plurality of experts and storing them in the memory (S110);
Classifying, by the processor, the collected expert information according to at least one classification criterion (S120);
Generating, by the processor, at least one expert pool based on the classification criterion (S130); And
Allocating, by the processor, the plurality of experts to the generated expert pool (S140); Including,
The collecting step (S110),
Collecting and obtaining, by the processor, information input from the plurality of experts or information crawled with respect to the plurality of experts;
Classifying, by the processor, expert information corresponding to the information stored in the database as structured information based on a previously stored database; And
Classifying, by the processor, the expert information excluding the structured information as unstructured information; Including,
The step of allocating the expert (S140),
Evaluating, by the processor, the expertise of the plurality of experts with respect to specialized fields corresponding to each of the generated expert pools (S210); And
Determining, by the processor, whether to assign the expert based on the expert evaluation result (S220); Including,
The step of evaluating the expertise (S210),
Classifying, by the processor, the collected expert information into structured information and unstructured information (S410);
The processor, evaluating the structured information and the unstructured information to obtain weights for the structured information and the unstructured information (S420);
Evaluating, by the processor, expertise based on the weight and the structured information and the unstructured information (S430); Including,
The classifying step (S120),
Clustering, by the processor, the collected expert information based on an artificial intelligence model (S710);
Obtaining, by the processor, at least one cluster based on the clustering result (S720);
Obtaining, by the processor, characteristics of the obtained at least one cluster (S730); Including,
The step of creating the expert pool (S130),
Generating, by the processor, expert pools respectively corresponding to the one or more clusters based on the acquired characteristics (S740); Including,
The step (S430) of evaluating expertise based on the weight and the structured information and the unstructured information,
Evaluating the formal information based on the evaluation information stored in the database; And
Estimating the expertise of the expert in the specialized field from the unstructured information using a model learned based on information stored in the database to derive a relationship between career information and expertise in each field of expertise; Including,
Allocating a task to the expert based on the expert's expert evaluation result;
Obtaining a result of the expert's work processing;
Generating a feedback on a result of the expert's expertise evaluation based on the result of the work process; And
Updating the expertise of the expert based on the feedback; Including more,
The step of updating the expertise,
Evaluating an estimation result of the learned model based on the feedback; And
Updating the learned model based on evaluation information on the estimation result; Including more,
Updating the trained model,
When the association between the first unstructured information and the first specialized field exceeds a preset reference value according to the result of updating the learned model, the first unstructured information is stored in the database as structured information on the first specialized field. And, when information on the first specialized field is not stored in the database, storing information on the first specialized field in the database; Including more,
The allocating step (S140),
The processor, when the expert evaluation result is greater than or equal to a preset value, allocating the expert to an expert pool corresponding to the expert evaluation result (S610);
When the plurality of expert evaluation results corresponding to the plurality of expert pools are equal to or greater than a preset value, allocating the expert to the plurality of expert pools according to the priority of the plurality of expert evaluation results (S620); Expert pool management method of the expert platform comprising a.

Images