KR102449806B1 - Method, apparatus and system of providing feedback service for member evaluating based on artificial intelligence - Google Patents

Abstract

Disclosed are a method, an apparatus, and a system of providing a feedback service for member evaluation based on artificial intelligence, which have an effect of providing an optimal member evaluation and management service for users. According to one aspect of the present disclosure, the method of providing a feedback service for member evaluation based on artificial intelligence, which is performed by a computing device, comprises: a step of receiving member evaluation content of an evaluator; a step of generating feedback for the evaluation content including suitability for a character of the evaluator and the evaluation content through the received member evaluation of the evaluator; a step of correcting the received evaluation content of the evaluator step by step according to the generated feedback for the evaluation content; and a step of providing the generated feedback for the evaluation content and the corrected evaluation content. The step-by-step correction for the evaluation content is performed through quantitative correction and qualitative correction.

Description

Method, device, and system for providing AI-based member evaluation feedback service

The present disclosure relates to a member evaluation feedback service, and more particularly, to a method, apparatus, and system for providing a feedback service based on artificial intelligence for previously performed member evaluation information.

In organizations with multiple members, such as public institutions or companies, evaluation of members is performed on a regular or irregular basis.

In relation to the evaluation of these members, it has been developed in the performance measurement system for evaluating the performance or ability of the conventional members.

However, since the conventional performance measurement system is mainly focused on measuring the performance or evaluating the performance of the current member, there is a problem in that the diversity of the organization or the member is not considered, so there is a limitation in the member evaluation.

Therefore, there is a need to develop a system that can properly evaluate members in consideration of various matters including organizational and member information.

Korean Patent Publication No. 10-2009-0012944 (2007.07.31)

The problem to be solved by the present disclosure is to provide a member evaluation and management service providing method, apparatus and system capable of properly evaluating a member in consideration of various matters including information of an organization or member.

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

A method of providing an AI-based member evaluation feedback service performed by a computing device according to an aspect of the present disclosure for solving the above-described problem, the method comprising: receiving evaluation contents of an evaluator for a member; generating evaluation content feedback including the appraiser's disposition and suitability for the evaluation content from the received evaluator's evaluation content; step-by-step correcting the received evaluation contents of the evaluator according to the generated evaluation contents feedback; and providing the generated evaluation content feedback and corrected evaluation content, wherein step-by-step correction of the evaluation content may be performed through quantitative correction and qualitative correction.

According to the method for providing an AI-based member evaluation feedback service according to the present disclosure, the method may further include the step of storing the received evaluation content of the evaluator for the member.

According to the method for providing an AI-based member evaluation feedback service according to the present disclosure, the propensity of the evaluator may be determined by comparing the evaluation score of the evaluator by comparing the evaluation score of at least one or more other evaluators for the same member.

According to the AI-based member evaluation feedback service providing method according to the present disclosure, the suitability for the evaluation content may be calculated based on a keyword included in the AI-based evaluation content.

According to the AI-based member evaluation feedback service providing method according to the present disclosure, the suitability for the evaluation content may be calculated based on the evaluation reliability and fidelity through the analysis of the evaluation writing pattern included in the evaluation content.

According to a method for providing an AI-based member evaluation feedback service according to the present disclosure, the method comprising: extracting evaluation guide data from a database based on the generated evaluation content feedback; and providing the extracted evaluation guide data.

According to the method for providing an AI-based member evaluation feedback service according to the present disclosure, quantitative correction of the evaluation content of the evaluator may be made by correcting the evaluation score for each evaluation item included in the evaluation content based on the propensity of the evaluator. have.

According to the method for providing an AI-based member evaluation feedback service according to the present disclosure, the qualitative correction of the evaluation contents of the evaluator may include at least one of words and sentences included in the evaluation contents of the evaluator based on the generated evaluation feedback. However, when converting the sentence, it may be made by converting it into a representative tone of the corresponding company or companies similar to the corresponding company.

An apparatus for providing an AI-based member evaluation feedback service according to an embodiment of the present disclosure includes: a memory; and a processor in communication with the memory, wherein the processor, when the evaluator's evaluation of the member is received, generates evaluation content feedback including the evaluator's disposition and suitability to the evaluation content from the received evaluator's evaluation content Thus, it is possible to correct the received evaluation contents of the evaluator step by step according to the generated evaluation contents feedback, and control to provide the generated evaluation contents feedback and the corrected evaluation contents.

An AI-based member evaluation feedback service providing system according to an embodiment of the present disclosure includes: at least one terminal for uploading evaluation content information for a member; and a computing device for processing the uploaded evaluation content information for the member, wherein the computing device is configured to, when the evaluator's evaluation of the member is received, from the received evaluator's evaluation content, the evaluator's disposition and evaluation contents A processor that generates evaluation content feedback including suitability, corrects the received evaluator's evaluation content step-by-step according to the generated evaluation content feedback, and controls to provide the generated evaluation content feedback and corrected evaluation content; may include

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

According to the present disclosure, it may have the following effects.

According to the present disclosure, it is possible to provide an optimal configuration evaluation and management service to the user through the member evaluation and management service in consideration of various matters regarding the organization and the members.

Effects of the present disclosure 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 diagram schematically illustrating a member evaluation and management service system according to an embodiment of the present disclosure.
FIG. 2 is a configuration block diagram of the computing device shown in FIG. 1 .
3 is a configuration block diagram of a processing unit according to an embodiment of the present disclosure.
4 is a block diagram of a processing unit according to another embodiment of the present disclosure.
5 is a block diagram of a processing unit according to another embodiment of the present disclosure.
6 is a diagram illustrating a detailed configuration block diagram of a processing unit according to an embodiment of the present disclosure.
7 is a flowchart illustrating a member evaluation and management service providing method according to an embodiment of the present disclosure.
8 to 10 are diagrams illustrating a user interface for providing member evaluation and management services according to an embodiment of the present disclosure.
11 is a flowchart illustrating a member evaluation and management service providing method according to another embodiment of the present disclosure.
12 is a diagram illustrating a user interface for providing member evaluation and management services according to another embodiment of the present disclosure.
13 is a flowchart illustrating a member evaluation and management service providing method according to another embodiment of the present disclosure.
14 is a diagram illustrating a user interface for providing member evaluation and management services according to an embodiment of the present disclosure.

Advantages and features of the present disclosure and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only the present embodiments allow the disclosure of the present disclosure to be complete, and those of ordinary skill in the art to which the present disclosure pertains. It is provided to fully understand the scope of the present disclosure to those skilled in the art, and the present disclosure is only defined by the scope of the claims.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present disclosure. In this specification, the singular also includes the plural, unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components. Like reference numerals refer to like elements throughout, and "and/or" includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first component mentioned below may be the second component within the spirit of the present disclosure.

Unless otherwise defined, all terms (including technical and scientific terms) used herein may have the meaning commonly understood by those of ordinary skill in the art to which this disclosure belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless clearly specifically defined.

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

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

In the present specification, the member evaluation and management service providing apparatus according to the present disclosure includes all of various apparatuses capable of providing a result by performing arithmetic processing. For example, the member evaluation and management service providing apparatus according to the present disclosure includes at least one computer or computing device, a server device, a terminal, and the like, or may be in any one form.

In the above, the computer may include, for example, a laptop, a desktop, a laptop, a tablet PC, a slate PC, etc. equipped with a web browser.

In the above, the server device is a server that processes information by communicating with an external device, and includes an application server, a computing server, a database (DB) server, a file server, a game server, a mail server, a proxy server, a web server, etc. may be included.

In the above, the portable terminal includes, for example, as a wireless communication device that guarantees portability and mobility, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System) , PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminal, smart phone ( All kinds of handheld-based wireless communication devices such as smart phones, watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or head-mounted-devices A wearable device may be included.

In this specification, in relation to the member evaluation and management service providing apparatus according to the present disclosure, an information provision control model may be defined or a related service platform may be built, which includes big data and artificial intelligence (Artificial Intelligence). ) can be created and provided by a technology-based computer, and is a virtual convergence technology (XR, eXtended) that collectively refers to virtual reality (VR, Virtual Reality), augmented reality (AR, Augmented Reality) and mixed reality (MR, Mixed Reality). Reality), information and communication technology (ICT) technology such as block-chain technology may be used or implemented by reference for security of personal information of a user using an information providing device. However, in the present specification, detailed descriptions of such ICT technologies will be omitted with reference to known technologies.

As used herein, the term “member” may refer to a member belonging to a company, organization, or group using a service platform provided by the service system or service providing device according to the present disclosure for convenience. Such members may be evaluators or evaluators who are the subject of evaluation.

In the present disclosure, the following contents may be referenced or used for member evaluation and management services.

1 is a diagram schematically illustrating a member evaluation and management service system according to an embodiment of the present disclosure.

The member evaluation and management service system according to the present disclosure may be configured to include a first terminal 100 , a second terminal 150 , and a computing device 200 .

The member evaluation and management service system according to the present disclosure may provide a service platform to provide services related to evaluation methods, evaluation relationships, evaluation questions, evaluation feedback, meeting data recommendation and management, etc. to the service platform users. In this case, the system may determine and use the suitability for each service item in order to provide an optimal member evaluation and management service.

The member evaluation and management service system according to the present disclosure is software for a member evaluation creation service based on member information, an evaluation feedback service based on analysis of the evaluator's disposition and evaluation content for members, and a meeting agenda recommendation and management service based on the member's evaluation record , hardware, etc.

According to an embodiment, the member evaluation and management service system may be configured by adding one or more components in relation to performing an operation according to the present disclosure in addition to the components shown in FIG. 1 .

The provision of the member evaluation and management service according to the present disclosure may be provided in the form of a web service through an application or the web.

In the above, the application is provided, for example, by the computing device 200, downloaded from the terminals 100 and 150, installed and then executed, user information is input through a user interface (UI) or member evaluation and management Service information (eg, member evaluation results, etc.) may be provided. At this time, when user information is input according to the execution of the application, the terminals 100 and 150 may directly process it and output result data, that is, member evaluation and management service result information. According to an embodiment, the terminals 100 and 150 transmit the user information input through the application to the computing device 200 without directly processing, and return and output the result data generated by the computing device 200 . You may. On the other hand, in order to support the direct processing of user information, the computing device 200 is an algorithm or logic for providing or processing member evaluation and management service according to the present disclosure, and/or related API (Application Programming Interface) ) or a plug-in may be provided to the terminals 100 and 150 .

However, in the following for convenience of description, the terminals 100 and 150 transmit the user information input through the executed application to the computing device 200 without directly processing, and the computing device 200 processes it to evaluate and evaluate members. Although it is described as an example that the management service information is generated, returned to the terminals 100 and 150 and provided to the user by the terminals 100 and 150, the present disclosure is not limited thereto. Meanwhile, in the above, at least some of the operation(s) according to the present disclosure performed by the computing device 200 may be performed through resources of the terminals 100 and 150 .

The terminals 100 and 150 may download and install an application for member evaluation and management service provision as described above, and perform various operations such as user information and various search requests through the application after installation.

Referring to Figure 1, the first terminal 100 is a user involved in the evaluation, for example, a terminal of an evaluator or an evaluation target (ie, an evaluator) or a customer who performs evaluation management (eg, a performance management act). It can represent a terminal.

The computing device 200 may collect various information about the user's personal information, performed tasks, past evaluation history, and the like through the first terminal 100 or an internal database.

The second terminal 150 may collect external data about the user of the first terminal 100 , that is, an evaluator/evaluator or a company to which the user belongs.

In this case, the external data on the company may include the size (eg, number of full-time employees), duration, industry, industry, etc. of the company.

In the above, the external data for the first terminal user may mean, for example, all data except for data accumulated using the first terminal 100 . Such data may include, for example, your social media history, whether you use a job search platform, and the like. The job search platform may include a typical job search site, for example, linked-in.

The computing device 200 may recommend or suggest an evaluation method for evaluating a member according to the present disclosure based on information collected through the first terminal 100 and/or the second terminal 150 , and the evaluator may perform evaluation according to a recommended or proposed evaluation method through the first terminal 100 .

According to an embodiment, the computing device 200 may diagnose the propensity of the evaluator based on the evaluation data, and may provide feedback on the evaluation contents performed by the evaluator based on the diagnosis.

According to an embodiment, the computing device 200 automatically recommends and suggests an evaluation method with reference to information collected through the first and second terminals 100 and 150 , and based on the proposed result, the meeting participant can be set, proceeded and recorded.

Even if it is only referred to as the computing device 200 in the present specification, since it is an entity that provides and manages the service platform according to the present disclosure, it may indicate that a service based on such a service platform is made.

Meanwhile, although the first terminal 100 and the second terminal 150 have been separately described in FIG. 1 , the two terminals may be implemented as one terminal. According to an embodiment, on the contrary, more terminals may be used for the member evaluation and management service according to the present disclosure.

In addition, even if it is just the first terminal 100, it is used in a symbolic meaning and may be used as a meaning indicating a plurality of terminals without necessarily indicating only one terminal.

Meanwhile, the computing device 200 may be referred to by various names such as a server, a computer, and a control unit, and is not limited thereto.

FIG. 2 is a configuration block diagram of the computing device 200 shown in FIG. 1 .

Referring to FIG. 2 , the computing device 200 according to an embodiment of the present disclosure may be largely divided into a memory and a processor 210 , and the memory may correspond to or include a storage unit 260 , and the processor 210 may include a communication unit 220 , a collection unit 230 , a processing unit 240 , a control unit 250 , and the like.

The communication unit 220 may provide an environment for communication with the first terminal 100 , the second terminal 150 , and an external source.

The communication unit 220 may include the following modules.

The communication module supports data communication with the terminals 100 and 150 and may receive user information from the terminals 100 and 150 . The communication module may be connected to various external sources to receive data related to member evaluation and management service according to the present disclosure. In the above, some data may be directly received from the user's terminals 100 and 150 rather than an external source, and the redundant data is a source with high reliability (in this case, the source includes the terminals 100 and 150), the latest Filtering may be performed with reference to information and the like.

Such a communication module may include at least one or more components that enable communication with the terminals 100 and 150 or at least one external source (not shown), for example, a wired communication module, a wireless communication module, and a short-distance communication module. It may include at least one of a communication module, a location information module, and the like.

In the above, the wired communication module includes various wired communication modules such as a Local Area Network (LAN) module, a Wide Area Network (WAN) module or a Value Added Network (VAN) module, as well as a USB ( Universal Serial Bus), HDMI (High Definition Multimedia Interface), DVI (Digital Visual Interface), RS-232 (recommended standard-232), power line communication, or POTS (plain old telephone service) can include various cable communication modules. have.

The wireless communication module includes, in addition to a Wi-Fi module and a wireless broadband module, a global system for mobile communication (GSM), a code division multiple access (CDMA), a wideband code division multiple access (WCDMA), and a UMTS. (Universal mobile telecommunications system), TDMA (Time Division Multiple Access), LTE (Long Term Evolution), 4G (eneration), may include a wireless communication module supporting various wireless communication methods such as 5G, 6G.

The short range communication module is for short range communication, and includes Bluetooth™, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, and Near Field Communication (NFC). Field Communication), Wi-Fi (Wireless-Fidelity), Wi-Fi Direct, and Wireless USB (Wireless Universal Serial Bus) technology may be used to support short-distance communication.

The location information module is, for example, a module for acquiring the location (or current location) of the terminals 100 and 150, and a representative example thereof is a Global Positioning System (GPS) module or a Wi-Fi module. For example, if the GPS module is used, the positions of the terminals 100 and 150 may be obtained using signals transmitted from GPS satellites. As another example, if the Wi-Fi module is used, the location of the terminals 100 and 150 may be acquired based on information of the Wi-Fi module and a wireless access point (AP) that transmits or receives a wireless signal. If necessary, the location information module may perform any function among other modules of the communication module to obtain data on the location of the terminals 100 and 150 in substitution or addition. The location information module is a module used to obtain the location (or current location) of the terminals 100 and 150 , and is not limited to a module that directly calculates or obtains the location of the terminals 100 and 150 . Such a location information module may be embedded in the terminals 100 and 150 to provide location information of the terminals 100 and 150 to the computing device 200 .

The collection unit 230 may collect information on the users of the first terminal 100 and the second terminal 150 and information on the organization to which the terminal user belongs through the communication unit 220 .

The processing unit 240 may process, for example, evaluation and management related data related to various embodiments of the present disclosure.

The controller 250 is involved in overall control of the computing device 200 and may transmit a control signal necessary for each component.

The control unit 250 controls the operation of all components of the computing device 200, but stores various algorithms available in the process or data about a program that reproduces the algorithm storage unit 260 , and data stored in the storage unit 260 may be used to perform various operations for member evaluation and management service information provision according to the present disclosure.

The storage unit 260 may store data collected, calculated, processed, and the like in the computing device 200 .

The member information-based member evaluation and management method, member evaluation content analysis and feedback method, member evaluation-based meeting recommendation and management method, etc. according to the present disclosure will be described in more detail as follows.

3 is a configuration block diagram of the processing unit 240 according to an embodiment of the present disclosure, and a configuration related to member information-based evaluation and management is illustrated, but is not limited thereto.

The "evaluation method" described herein refers to a method of evaluating a member through a service platform for member evaluation and management service, and this evaluation method is, for example, evaluation direction, evaluation order, job evaluation data It can be determined with reference to the like. In this case, the evaluation direction data may indicate whether the evaluation of the member is, for example, bottom-up, top-down, parallelism, or a combination of the bottom-up and top-down. The parallel expression may represent an evaluation between members having the same or similar job or rank or position. The evaluation order data indicates the number of evaluations used for member evaluation. For example, if the number of evaluations is 1, it is usually evaluated by the direct superior of the target member, and if it is 2 times, the evaluation of the next higher direct superior is also included. can indicate In addition, the job evaluation data may represent data for evaluating a job for a member, for example, competency, performance, skill, and the like. However, the present disclosure is not necessarily limited to the above definitions or numerical values.

The "evaluation relationship" described in this specification may indicate a relationship between an evaluator and an evaluator among members, for example, an evaluation performing entity for a member through the service platform. Such an evaluation relationship may be referred to in determining the optimal evaluator for the subject to be evaluated, and setting the evaluation authority for the determined evaluator. In this case, the evaluator's decision may be made by the suitability calculated in consideration of the relationship with the evaluator, that is, the subject to be evaluated. That is, the evaluator with the highest fitness calculated as the evaluator for the evaluator may be determined as the optimal evaluator. etc. In determining the optimal evaluator, various factors other than the suitability may be further considered. On the other hand, the evaluation authority setting refers to whether it is possible to refer to the evaluation results of the previous year and evaluation results of others, and it is calculated as an optimal evaluation relationship and used as an auxiliary material in the evaluation process according to the hierarchy between the designated evaluation target and the evaluator and related preset policies. It is possible to indicate whether various information about possible subjects can be read. For example, if colleague A is selected as the optimal evaluator according to the evaluation relationship calculation, authority may be set so that past evaluation results and self-evaluation results cannot be viewed because A is a colleague. According to an embodiment, the evaluation authority may be determined according to the suitability of the evaluator determined as the optimal evaluator. For example, even if the evaluator with the highest fitness among the plurality of evaluation candidates is determined as the best evaluator, the evaluation authority of the evaluator may be set differently according to the first critical fitness section, the second critical fitness section, and the like. In another embodiment, when the evaluation suitability calculated for all of the target evaluators is less than the reference value (eg, 50%), the target evaluator not only does not select an optimal evaluator but also reselects the target evaluation candidate. In this case, it may be excluded from the list of candidates for the re-selected evaluators. Meanwhile, in the above, the evaluation authority may be the contents of the data or information access range or authority for the subject to be evaluated, the evaluation method, the evaluation item, and the like, depending on the embodiment.

As used herein, the term “evaluation item” may refer to an evaluation item used when evaluating a subject according to an evaluation method selected by an evaluator in the evaluation relationship. In this case, the evaluation items may include information for defining and judging diversity of organizations and members as well as business problems.

Referring to FIG. 3 , the processing unit 240 may include an evaluation method calculation unit 310 , an evaluation relationship calculation unit 320 , and an evaluation item classification unit 330 , but is not limited thereto.

The evaluation method calculation unit 310 may calculate an evaluation method suitable for member evaluation based on the information collected through the collection unit 230 of FIG. 2 .

The evaluation relationship calculation unit 320 provides relation data for selecting an optimal evaluator for the evaluation of the subject to be evaluated, that is, the subject to be evaluated, from among the evaluation candidates selected based on the evaluation method calculated by the evaluation method calculating unit 310 . can As described above, the evaluation relationship calculating unit 320 may differentially grant the evaluation authority of the corresponding evaluator based on the relation data.

The evaluation item classification unit 330 classifies the evaluation items to be used for evaluation based on the evaluation method calculated by the evaluation method calculation unit 310 and the relationship data between the evaluator and the subject being evaluated by the evaluation relationship calculation unit 320, and Among the classified evaluation items, the optimal evaluation item can be determined and provided.

The member evaluation and management service system (or device) according to the present disclosure may provide evaluation method data determined with reference to evaluation target and organizational data.

In the above, the evaluation target, tissue data, etc. referred to to define the evaluation method may be classified into external data and internal data.

In the above, internal data may be defined as data input by a user of the service platform provided by the device or system according to the present disclosure, or data generated, stored, etc., processed in the course of use, and external data refers to the service platform It can be defined as data that is difficult or impossible to obtain through For example, the external data may be data collected, acquired, and processed through an external source.

The internal data of the evaluation target information may be divided into user input information and user behavior information.

In the above, the user input information may include personnel information, such as job, position, position, year of employment, affiliated department, department, etc. input by the user in the process of using the service platform.

In the above, the user behavior information may represent information extracted based on the user's behavior occurring according to the use of the service platform and log data representing the behavior. In the above, the action may represent, for example, a page inquiry, evaluation writing, interview request, etc. in the process of using the service platform.

In the above, the behavior log data may be composed of a log of the type of behavior of the service platform user, the time difference between each behavior, and the target of the behavior, and each log is collected and the behavior type of the user through machine learning It can also be used as a reference for forecasting, such as additional information. In relation to this, the computing device 200 provides a behavior log sequence vector for vectorizing a sequence of various information including a time difference between various behavior logs and a time difference between various behavior logs of members on a service platform, and a sequence after vectorizing various information based on each behavior log generating at least one of the behavior log vector sequences to be transformed, and calculating the type and time of the user's behavior related to member evaluation based on the behavior log sequence vector and/or the behavior log vector sequence to define the behavior information of the user can

In general, vectorization refers to digitizing a concept so that a machine can learn it. In relation to this, vectorization in the present disclosure may indicate, for example, that the evaluator digitizes and records the user's behavior log data, such as viewing past records of the subject to be evaluated, viewing interview records, starting evaluation, ending evaluation, and the like. On the other hand, in the present disclosure, sequencing may indicate that the behavior log or the vectorized behavior log is time-series arranged to be set as one object. For example, the sequence of the evaluator may be expressed in a chronological order, such as viewing past evaluation records for the subject to be evaluated, viewing previous interview records, starting evaluation, and ending evaluation writing. In this way, performing the sequencing is, for example, by sequencing the behavior vectors of the members, the evaluation process of each member (evaluator), for example, the pattern before and after writing the evaluation, can be grasped, and based on this, the raters can be classified in various ways. because you can

In Table 1 below, evaluator types are classified according to an embodiment, but the present disclosure is not limited thereto.

example Type 1 View past evaluation　->　Type of evaluation Type 2 View past evaluations　->　Read interview records　->　Type of evaluation Type 3 View interview record　->　View past evaluation　->　Type of evaluation Type 4 Interview record reading　->　Type of evaluation Type 5 Types of writing assessments without access to records Type 6 A type of compiling multiple member evaluations in a short period of time

In the present disclosure, as described above, by defining and classifying evaluation types in various ways based on the behavior log vector sequence or the behavior log sequence vector, it is possible to obtain more quantified detailed information about the members (evaluators) for each evaluation type. Also, the computing device 200 may generate or obtain statistical data on an evaluation writing method of each evaluation type or evaluation type group through the quantified classification. In addition, the computing device 200 may provide a service by generating an evaluation guide suitable for each evaluation type based on artificial intelligence by learning the acquired quantification data to an artificial intelligence (computer).

Also, for example, in preparing for an interview (1:1) between members through the service platform, access the service platform in advance to register the agenda for discussion, that is, whether or not to add an opinion on the agenda of the other party From the viewpoint, the computing device 200 may classify the type of the user. Here, the addition time may indicate, for example, addition immediately before an interview, addition one hour before an interview, and the like.

Next, the external data in the above represents data acquired through an external source such as social media (SNS: Social Networking Service) data of the user rather than the service platform, and through this, information on total career, social media activity, etc. is extracted and may be referred to as a user type classification.

According to the present disclosure, the member evaluation and management service system (or device) extracts or calculates a list of evaluation methods suitable for the subject and the suitability of each evaluation method based on the above-described internal and/or external data to obtain an optimal evaluation method can be recommended.

On the other hand, the organization information mentioned in the present disclosure can also be divided into internal data and external data, and the internal data represents the organizational structure, size, etc. It can indicate the period of existence, industry, industry group information, and the degree of media exposure.

According to the present disclosure, a member evaluation and management service system (or device) identifies an organization's tendencies and philosophies based on the various internal and/or external data described above, and the identified organizational tendencies and philosophies include a list of suitable evaluation methods and The suitability of each evaluation method may be extracted or referenced for calculation.

The member evaluation and management service system (or device) according to the present disclosure may extract or calculate an appropriate evaluation time and calculation degree of an evaluation method with reference to a past evaluation history.

The member evaluation and management service system (or device) according to the present disclosure may calculate the overall suitability including the suitability for the organization and the subject, and may recommend and provide an appropriate evaluation method with reference to the past evaluation history.

The member evaluation and management service system (or device) according to the present disclosure may calculate the evaluation relationship by using the same data as the above-described evaluation method.

For each evaluation target, the evaluation suitability for each evaluator candidate selected based on the evaluation method may be calculated. That is, the evaluation suitability refers to the evaluation suitability of the candidate evaluator for the evaluation target according to the evaluation method. In the end, it is objectively identified whether and to what extent it is suitable as an evaluator of an individual evaluator candidate for the evaluation target. It can be defined as data provided by calculating it to be possible.

In this case, the selection of an evaluator candidate based on the evaluation method may mean selection of each evaluator's leader in the case of top-down, and selection of subordinates of each subject in the case of bottom-up.

On the other hand, evaluation suitability may be selected by considering the length of service between the subject and the evaluator candidate, joint work performance, quality and quantity of interviews, consistency of behavior patterns, evaluation tendencies, and the like. This , the evaluation tendency may represent, for example, remuneration, generosity, and the like.

After calculating the degree of suitability for each evaluation candidate, the evaluation burden on the evaluator can be minimized by considering the evaluation that is already in progress or at the same time.

That is, the computing device 200 may recommend and provide an evaluation relationship in which the fitness of each evaluator-evaluated subject (ie, subject) pair is greater than or equal to a threshold or maximum while minimizing the evaluation burden of each evaluator.

The computing device 200 may automatically grant appropriate evaluation authority according to the relationship between each evaluator-subject pair and preset evaluation guidelines. The relationship between each rater-subject pair may be expressed as a subordinate relationship, a peer relationship, or the like.

The member evaluation and management service system (or device) according to the present disclosure may present an evaluation item according to the evaluation method proposed for the evaluation item.

The member evaluation and management service system (or device) categorizes the evaluation histories of users using the service platform and the evaluation items they input into a database, and among them, based on the presented evaluation method and the preset organization value and evaluation guidelines The fitness of each evaluation item can be calculated.

Accordingly, according to the present disclosure, it is possible to present an evaluation item set capable of calculating a degree of suitability greater than or equal to a threshold while avoiding evaluation of unnecessary overlapping items through evaluation items. In this case, the fitness above the threshold may indicate the maximum fitness, but is not limited thereto.

6 is a diagram illustrating a detailed configuration block diagram of the processing unit 240 according to an embodiment of the present disclosure.

Referring to FIG. 6 , the evaluation method calculation unit 310 of FIG. 3 will be described as follows.

The evaluation method calculation unit 310 may include a data classification unit 610 , a data extraction unit 620 , a data conversion unit 630 , a learning unit 640 , and an operation unit 650 . However, the present disclosure is not limited thereto.

The data classification unit 610 may classify the external data collected through the collection unit 230 of FIG. 2 according to a predetermined criterion.

The data extraction unit 620 may extract information on the behavior of using the first terminal 100 indicating the characteristics of the user.

The data conversion unit 630 may convert the data extracted through the data extraction unit 620 into a learnable form.

The learning unit 640 may learn by combining the personal information input by the user through the first terminal 100 , the result of the data conversion unit 630 , and the result of the first data classification unit 610 .

The operation unit 650 may derive appropriate evaluation methods based on the result of the learning unit 640 and transmit them to the communication unit 220 .

Accordingly, the evaluation method calculation unit 310 may provide an appropriate type of evaluation method in consideration of the characteristics of the organization and the characteristics of the subject to be evaluated.

As described above, the characteristics of the organization may be classified by collecting external data such as duration, industry/industry group, and location.

In addition, the characteristics of the subject are the use behavior information accumulated through the service platform using the first terminal 100 and pre-input personal information (eg, position, position, task, past evaluation results, etc.) can be specified based on

Meanwhile, the usage behavior information may indicate a behavior sequence vector for sequencing the time difference between various behavior logs and behaviors on the service platform, tasks performed on the service platform, and various behaviors. This may be referred to, for example, in determining the degree of sincerity or participation in the evaluation of the evaluation candidate.

Accordingly, the evaluation method calculation unit 310 may recommend and present an optimal evaluation method based on the specified organization and information on the target.

That is, the evaluation method may mean the direction of evaluation, the order of evaluation, the cycle and timing of evaluation, evaluation items (items), and the like.

In the above, the cycle and timing of the evaluation may be expressed as, for example, regular or irregular. A predefined periodic evaluation cycle may be adjusted according to an evaluation result (eg, average) immediately before or during a predetermined round. For example, if the evaluation result is less than the threshold (or judged to be bad), it is adjusted to be shorter than the preset regular evaluation cycle for the member, and in the opposite case, that is, the evaluation result is above the threshold (or judged to be good) ), it can be adjusted to be longer than the regular evaluation cycle set for the member concerned. In the above, the degree of adjustment of the periodic evaluation cycle may be differentially adjusted and controlled according to how far the evaluation result is from the threshold. Alternatively, after the current evaluation is made, feedback data for the corresponding evaluation may be generated within a predetermined time so that it can be reflected in a preset evaluation period and time point adjustment. Similarly, by calculating the reliability of the evaluation made immediately before, when the corresponding reliability value is less than the threshold according to the calculated reliability value, the evaluation of the corresponding member through another evaluator may be made at an arbitrary evaluation time. In this case, the evaluation of the member by the other evaluator may be performed according to the same evaluation method as the previous evaluator or a different evaluation method. Another evaluation method may be document evaluation based on evaluation data, such as a conversation between a previous evaluator and a corresponding member in the evaluation process, as well as the above-mentioned evaluation direction, evaluation order, or evaluation items.

In the evaluation cycle, irregular evaluation may be arbitrarily performed according to occurrence of a predetermined event. For the above event, for example, if the evaluation result made in the regular evaluation cycle falls below the threshold compared to the previous one, the evaluation cycle can be set irregularly, and the member's work performance, qualification changes such as promotion or other qualifications, changes in position or job If there is such a thing or is scheduled, irregular evaluation may be performed immediately or at any time regardless of a preset regular evaluation cycle.

A weight is assigned to each of the above-described evaluation period and time and related contents, and an evaluation method may be influenced based on this.

It is possible to present an evaluation type suitable for the user by continuously learning and improving by using the satisfaction and adoption of the evaluation method recommended for the same type or the same organization and subject in the past.

The evaluation relationship calculating unit 320 of FIG. 3 will be described with reference to FIG. 6 .

Referring to FIG. 4 , the evaluation relationship calculation unit 320 includes a data classification unit 610 , a data extraction unit 620 , a data conversion unit 630 , a learning unit 640 , and an operation unit 650 . can be However, the present disclosure is not limited thereto.

The data classification unit 610 may classify the external data collected through the collection unit 230 of FIG. 2 according to a predefined criterion.

The data extractor 620 may extract relationship information indicating characteristics between a user and another user. In this case, the user and the other user may represent, for example, an evaluator and an evaluator.

The data conversion unit 630 may convert the data extracted through the data extraction unit 620 into a network analysis index such as proximity and mediation degree and a learnable form.

The learning unit 640 collects personal information input by the user through the first terminal 100, the result of the data conversion unit 630, the result of the data classification unit 610 that collects and classifies the organizational characteristic information, and Based on the relationship diagram and results of past evaluations, it is possible to learn to present the evaluation suitability for potential evaluators (eg, the optimal evaluator for the presented subject).

The calculation unit 650 may calculate and present evaluation suitability of potential evaluators based on the learned result.

Such evaluation suitability may be used as a weight for each evaluator. That is, having a high evaluation suitability may mean having a high weight.

The evaluation relationship calculation can be viewed as a process for selecting the optimal evaluator from among evaluator candidates.

In judging the evaluator candidates for calculating the evaluation relationship, if each evaluator candidate has a history of participating in an evaluation before, the contents of the ex post evaluation of the evaluation in the previous evaluation may be referred to. For example, weights may be given to the candidate of the evaluator according to whether or not the candidate made a mistake as a result of the post-evaluation for the corresponding evaluation in the previous evaluation. Whether or not there is a mistake as a result of the ex post evaluation may include, for example, a general mistake and a mistake according to the status of the subject of the corresponding evaluation after the corresponding evaluation. First, a simple mistake, a repeated mistake, an intentional mistake, etc. can be distinguished as common mistakes, and different weights can be assigned to each case. In addition, as a result, by comparing the evaluation of the subject to be evaluated by the candidate of the evaluator at the time of the evaluation and the current state of the subject to be evaluated, the degree of error may be digitized and different weights may be assigned. For example, when a candidate evaluator has given a negative evaluation opinion to the evaluator, but the current evaluator's evaluation is the best, negative weight may be given to the qualification or ability as an evaluator for the evaluator candidate. On the other hand, in the above case, it is possible to refer to the post-mortem examination of such evaluation errors, for example, the evaluation of the candidate at the time was not a mistake, and whether there were any motives or motives of the subject to be evaluated after that. In this case, it is desirable to give lower weight to the evaluation results made too long ago compared to the recent evaluation in this ex post evaluation. In addition, since the value of evaluation may change depending on the head or goal of an organization or group, it is desirable to consider such a situation and reflect it in determining the weight.

In judging an evaluator candidate for calculating the evaluation relationship, relationship information between each evaluator candidate and the evaluator may be considered. At this time, the relational information is determined according to not only the above-mentioned position or position, but also whether the same or similar duties are performed, whether the same department is in the same department, gender, the number of years worked together in the same department, whether the current or recent work in the same department, and whether or not the project has been carried out together. Different weights may be assigned.

In determining the evaluator candidates for calculating the evaluation relationship, information about each evaluator candidate or group to which the evaluator belongs may be further considered. For example, different weights may be assigned to the case where the atmosphere of the evaluator candidate or the group to which the target evaluator belongs is not good and when the atmosphere is good. In addition, when the candidate for the evaluator belongs to a group different from the target evaluator, the relationship with the leader of the group to which the subject evaluator belongs may also be considered. In the above case, the weight may be determined by further considering the relationship between the group to which the evaluator candidate belongs and the group to which the target evaluator belongs.

In addition, in judging an evaluator candidate for calculating the evaluation relationship, the internal reputation of the evaluator candidate, gender, the number of evaluators participation, and whether or not the target evaluator has been previously evaluated may be assigned as individual weights.

Referring to FIG. 6 , it is a block diagram of the evaluation item classification unit 330 of FIG. 3 .

In FIG. 6 , the evaluation item classification unit 330 may include a data classification unit 610 , a learning unit 620 , and an operation unit 630 . However, the present disclosure is not limited thereto.

The data classifying unit 610 may classify the accumulated item data collected through the collecting unit 230 of FIG. 2 according to a predetermined criterion.

Based on the item data classified by the data classifying unit 610 , the learning unit 620 may learn a classification model for automatically classifying, for example, when a new item is input.

The calculator 630 may automatically classify a new item when a new item is input by using the learned model.

In addition, the calculator 630 may present the optimal set of items by calculating the suitability of each item and the suitability of the set of items based on the set evaluation method and relationship.

The evaluation item classification unit 330 may provide information on the order of evaluation items. This is to logically derive the evaluation results.

The evaluation item classification unit 330 may provide different evaluation items according to whether the evaluation method determined according to the present disclosure is a non-face-to-face evaluation or a face-to-face evaluation. For example, if the evaluation method is a non-face-to-face evaluation method and the evaluation order is a 3rd order, the evaluation item classification unit 330 may classify and provide evaluation items according to each order.

The evaluation item classification unit 330 may generate and provide the evaluation item itself by dividing it into a common question and a specific question to be evaluated. The subject specific question may be derived from a common question.

The evaluation item classification unit 330 may generate and store a set for each evaluation item and correct answer in the form of a lookup table, compare the actual subject's response with the lookup table, and provide additional questions. That is, the evaluation item classification unit 330 may include information on an additional question or a related question according to a predictable response rather than a single correct answer in the lookup table. The additional question or related question may be generated in the form of an individual lookup table and link information may be generated so that each lookup table of a hierarchical structure is related to each other, so that additional evaluation item data can be extracted and provided immediately. If a response that is not defined in the lookup table is received, the evaluation item classification unit 330 may induce a question about an evaluation item other than the evaluation item defined in the previous lookup table set.

The evaluation item classification unit 330 may assign weights to the evaluation items based on the evaluation items adopted and used in the previous evaluation and the post evaluation, that is, feedback. In this case, the evaluation item classification unit 330 may adjust the weight with reference to a recent trend. In the above, the trend may include evaluation questions-responses between other evaluators and evaluators, and information that has recently become an issue in in-house information or job/job search sites.

The evaluation item classification unit 330 may reclassify a predefined item set with reference to the above-described information. For example, the evaluation item classification unit 330 selects and provides some individual evaluation items or evaluation item sets to any member at any time before evaluation, performs a test, and based on the result, each evaluation item or evaluation It is possible to set whether or not to accept the item set, weight, and the like. In this case, some of the individual evaluation items or set of evaluation items of the test may include those related to common evaluation or specific to the job. The target of the test may include, in addition to the members, the general public, job preparers, evaluation experts such as evaluation coaching experts or psychiatrists. Depending on the test object, different weights may be given to the result of the corresponding test.

The evaluation item classification unit 330 generates and tests a plurality of evaluation items or item sets in consideration of nuances or gender for the same question or category, and selects a specific item or set according to the result to be used for evaluation. .

7 is a flowchart illustrating a member evaluation and management service providing method according to an embodiment of the present disclosure. 8 to 10 are diagrams illustrating a user interface for providing member evaluation and management services according to an embodiment of the present disclosure.

Referring to FIG. 7 , in operation 11 , the computing device 200 may receive member information.

In operation 12, the computing device 200 may generate evaluation method data by extracting evaluation target and organization information from the received member information.

In operation 13 , the computing device 200 may generate evaluation relation data including an evaluator candidate and evaluation suitability for the extracted evaluation target based on the generated evaluation method data.

In operation 14 , the computing device 200 may generate evaluation item data for an evaluator candidate selected from the evaluation relation data based on the generated evaluation method data.

In operation 15 , the computing device 200 may provide the generated evaluation method data, evaluation relation data, and evaluation item data.

In (a) of FIG. 8, member information, which is an evaluator, is provided through a user interface provided by the service platform. Through the user interface of Fig. 8 (a), it is possible to know the evaluation target and its information.

In (b) of FIG. 8, an evaluator candidate and its suitability are provided through a user interface provided by the service platform.

For example, through the user interface shown in FIG. 8A , the evaluator may obtain information about a member who is an evaluation target and information on a recommended evaluation method for the selected member.

For example, through the user interface shown in (b) of FIG. 8 , the evaluator may know information and suitability of an evaluator candidate, and may obtain information on a recommended evaluator candidate most suitable for him/her.

In the user interface 900 shown in FIG. 9 , evaluation method and evaluation relationship information 910 , a set of items and their suitability 920 , an individual question and its suitability 930 , and a recommended item set and individual item information 940 . is being provided. When an item set is selected in the above, information on individual items belonging to or constituting the item set and a degree of suitability thereof may be provided. Also, the recommended item set and individual item may be a completely new item set and individual item selected from among the item set information 920 and the individual item information 930 or extracted and processed therefrom.

In the user interface 1000 shown in FIG. 10 , information on the subject to be evaluated 1010 , information on the recommended evaluator suitable for the subject to be evaluated 1020 , information about the recommended evaluation method 1030 , the evaluation relationship information 1040 , and the recommended item set and individual Item information 1050 may be provided.

4 is a block diagram of a processing unit 240 according to another embodiment of the present disclosure.

Referring to FIG. 4 , the processing unit 240 may include an evaluation propensity diagnosis unit 410 , an evaluation content determination unit 420 , an evaluation content correction unit 430 , an evaluation guide processing unit 440 , and the like.

According to an embodiment of the present disclosure, by diagnosing the propensity of the evaluator based on the evaluation data, evaluating the evaluator's evaluation content, recommending an appropriate evaluation guide from the database (DB), or calculating the word use characteristic information of the company Based on this, improvement of evaluation sentences and feedback can be output.

To this end, in the present disclosure, evaluation of members may be performed in advance, recorded, and stored.

In relation to this, the present disclosure discloses a diagnosis of an evaluator's propensity based on evaluation data and providing feedback on evaluation content, securing accuracy based on an existing evaluation propensity when writing a quantitative evaluation of an evaluator, and an automatic sentence conversion system based on evaluation feedback.

The evaluation of the evaluation contents may consist of fidelity and suitability. Integrity indicates whether the evaluator has prepared the evaluation with sincerity and sincerity, and suitability may indicate appropriateness and appropriateness of the evaluation contents.

The degree of sincerity indicating whether the computing device 200 has faithfully participated in the evaluation may be largely calculated based on the evaluation preparation and the prepared evaluation similarity.

The computing device 200 may determine the fidelity based on the action log indicating the time difference between the action and the action performed by the evaluator until the completion of the assessment preparation in relation to the assessment preparation. The computing device 200 may sequence the behavior log to create a sequence vector representing the rater behavior characteristic. The computing device 200 may calculate the types of pre-tasks performed to write the evaluation and the time devoted thereto based on the action sequence vector.

The evaluation similarity may indicate a degree of similarity between evaluation contents created in the past and evaluation contents created in the past. The computing device 200 may calculate the fidelity by synthesizing the time devoted to the evaluation preparation, the type of action performed prior to the evaluation, and the evaluation similarity.

The suitability/evaluation type can be calculated, for example, it can be divided into general suitability and suitability for the evaluation field and purpose.

Through the morpheme analysis of the evaluation content through the general suitability, it is possible to identify profanity, negative/critical tone and expression, and whether or not it refers to a specific individual. The computing device 200 estimates the subject through machine learning techniques such as morphological analysis and topic modeling of the evaluation content with respect to the suitability for the evaluation field and purpose, and calculates the degree of similarity between the evaluation subject and purpose and the calculated subject. can

In addition, the computing device 200 may calculate the suitability of the corresponding evaluation by synthesizing the general suitability and the suitability for the evaluation field/purpose.

The evaluator propensity diagnosis can be diagnosed through comparison with internal/external evaluators by synthesizing information about the evaluator, the relationship information between the evaluator and the evaluator, and the organization to which the evaluator/evaluator belongs.

The computing device 200 may classify types of each of the evaluator information, the relationship information, and the organization information. That is, the evaluator type, relationship type, and organization type can be calculated. The computing device 200 may extract an internal evaluation history matching the calculated type and an external (other customer) evaluation history existing on the service platform. The computing device 200 may diagnose the evaluation tendency of the evaluator based on the extracted internal/external evaluation history. The computing device 200 may determine whether to evaluate the evaluator's tendency generously or conservatively by comparing the distribution of the evaluation results.

In relation to the evaluation sentence correction, the computing device 200 learns the corporate sentence pattern recorded through 1:1, review, and target product of the service platform, and calculates and calculates corporate word usage characteristic information and representative tone. It is possible to modify the sentence of evaluation through the backlash of the sentence using the specified characteristics. Representative tone is not only text data of evaluation data within the members of the company, but also used by the company itself in internal data stored or sent/received in-house, in-house announcements, in-house notices, and external promotional materials, even if it is not evaluation-related data. It can be calculated from sentences and words. The calculation condition at this time may be calculated by weighting it based on the frequency of the word and/or sentence, or weighted based on the importance of the document in which the word and/or sentence appears. , weights may be given and calculated according to the rank of the in-house member using the corresponding word and/or sentence. In addition, the computing device 200 may collect information on representative tones of not only the corresponding company but also the corresponding company and similar companies belonging to a similar technology field from an external source, and may convert the evaluation sentence based on the collected information. In this case, the external source may be a social network service (SNS) of a similar company, related columns, related articles, company introductions, promotional materials such as IR (Investor Relations) materials.

The computing device 200 uses information about the company (work history, industry group, media exposure, organization chart, etc.) and information about each individual's characteristics when it is impossible to calculate the word usage characteristics and tone due to the lack of necessary data to provide a similar type It is possible to calculate the tone by extracting the sentence data of the company or individual of

The computing device 200 may collect all sentences written by members of the company, extract context information through keyword extraction and topic extraction, and refine the meaning of a word used based on the extracted context information. In addition to specifying general dictionary meanings and synonyms, it is possible to specify the meaning of specific words in the company. In this way, it is possible to group companies that have a similar tendency to use sentences as well as information about the company.

In addition, the computing device 200 decomposes the sentence into token units, builds data that the natural language processing model can learn through stopword processing, etc., and through this, a specific company or the corresponding company in the existing natural language processing model such as BERT It is possible to build a model that reflects the characteristics by learning data that reflects the characteristics of similar types of companies.

The computing device 200 may perform a single language translation using the built model to convert a sentence through the same language translation/paraphrase, thereby ensuring anonymity. That is, the computing device 200 may convert the evaluation written by all individuals into a representative tone of a corresponding company or a representative tone of companies similar to the corresponding company. At this time, representative tones of companies similar to the corresponding company may be collected from an external source, and evaluation sentences may be converted based on this. In this case, the external source may be a social network service (SNS) of a similar company, related columns, related articles, company introductions, promotional materials such as IR (Investor Relations) materials.

After the conversion, the computing device 200 may be substituted as needed according to the specific word use case in the company specified above.

The computing device 200 may identify and remove information and personal sensitive information that can identify the evaluation target and the evaluation author, such as “~~ sir, Mr., team leader” included in the evaluation contents. As illustrated in FIG. 12 , the computing device 200 may compare and present before/after the evaluation sentence correction.

The evaluation propensity diagnosis unit 410 may diagnose an evaluation propensity for a target evaluator.

The evaluation content determination unit 420 may determine the evaluation content of the target evaluator.

The evaluation content correction unit 430 may correct the evaluation contents for the evaluation contents determined by the evaluation contents determination unit 420 .

The evaluation guide providing unit 440 may provide an evaluation guide according to the evaluation content determined by the evaluation content determining unit 420 .

The computing device 200 may recommend an item of a related evaluation guide based on the evaluation result of the evaluator and the evaluation content. For example, the computing device 200 presents content that conveys the pros and cons of the generosity tendency to evaluators who show a strong tendency toward generosity, and when the suitability is low, the same evaluation as the content indicating the importance of evaluation in accordance with the topic. Evaluation contents of other organizations in the field can be presented as an example.

First, an embodiment in which the computing device 200 according to an embodiment of the present disclosure diagnoses an evaluator's disposition based on evaluation data and provides feedback on evaluation contents based on the diagnosis will be described.

The evaluation of the rater may be made, for example, by identifying the disposition of the rater. The computing device 200 may compare, for example, with at least one or more other evaluators in order to identify the propensity for the target evaluator. That is, the computing device 200 may define the tendency of the target evaluator based on the evaluation of other evaluators. In this case, the at least one other evaluator may include, for example, a member of the same rank or a member of the same organization. Meanwhile, in the above description, the tendency may be defined as directness, generosity, or the like.

Computing device 200, for example, by comparing the evaluation scores of at least one or more other raters for the same member, it is possible to determine what kind of tendency the evaluation score of the target evaluator. At this time, if the evaluation scores of a plurality of other evaluators are used, the reference data for judging the propensity of the evaluation score of the target evaluator is the average evaluation score of the evaluation scores of all other evaluators and the evaluation scores of all other evaluators. Calculate the average evaluation score of evaluation scores excluding the highest evaluation score and the lowest evaluation score, and the distribution of evaluation scores of all other raters, and define the propensity separately based on the calculated distribution, for example, and define the distribution of the target evaluation score. judged to belong to, and if the distribution is greater than or equal to the threshold (for example, very wide), the other appraisers are classified again based on common items (for example, the same rank, etc.) among the plurality of other appraisers It is possible to re-evaluate by calculating the distribution of one evaluation score.

If the above description relates to a kind of quantitative evaluation, the computing device 200 may also evaluate the evaluation contents of the evaluator to perform qualitative evaluation. That is, the computing device 200 may calculate and provide a qualitative evaluation result for the evaluation content, such as whether the evaluation of the evaluator is of good quality, for example, the evaluation suitability of the evaluator, evaluation reliability, and sincerity of evaluation.

The computing device 200 may evaluate the evaluation content of the evaluator based on, for example, a keyword.

The computing device 200 may calculate the suitability of the evaluation content by evaluating whether the content is suitable for the evaluation field, for example, based on a keyword of the evaluation content of the evaluator. In this case, the evaluation field may include, for example, competency, achievement, interpersonal, peer evaluation, and the like.

The computing device 200 may calculate the suitability of the evaluation content by learning the main keywords mainly used for each evaluation field using artificial intelligence technology.

For example, when the evaluation content, that is, the evaluation sentence is input into the AI model, if the evaluation sentence contains more than a threshold value of keywords mainly used in the evaluation field, the suitability of the evaluation content of the evaluation sentence can be determined as positive or appropriate. .

On the other hand, the computing device 200 extracts keywords (words) irrelevant to the evaluation field from the corresponding evaluation sentence, and when the number of extracted keywords is greater than or equal to a threshold value, the appropriateness of the evaluation content of the evaluator is negative or It can be judged as non-conformity.

The computing device 200 may define the main keywords in advance, but may individually randomly assign a preset weight to each keyword, and use the appropriateness of the evaluation content to determine the suitability of the evaluation content based on the keyword. In this case, the main keywords may be defined separately for each category, such as positive and negative, for example. On the other hand, the computing device 200 may give additional weight according to how repeatedly the defined main keyword is used in the corresponding evaluation sentence.

In defining the main keyword, the computing device 200 is repeatedly used or used as a main keyword in the evaluation sentences of the evaluator(s) of the same or opposite evaluation tendency when at least one or more other evaluators or the propensity of the evaluators are judged. A keyword, etc. with a number of exposures greater than or equal to a threshold may be referred to in a predefined or evaluation sentences of a plurality of other evaluators.

The computing device 200 may provide data on reliability, sincerity, etc. of the evaluation of the target evaluator through analysis of the evaluation sentence writing pattern of the target evaluator.

In this case, the evaluation sentence writing pattern may include, for example, a time required for writing an evaluation sentence, a time of writing the evaluation sentence, and the like.

Computing device 200 compares the time taken to write the evaluation statement of the target evaluator with the writing time required by other evaluators for the same member, stores the writing time data, and calculates the evaluation sentence writing reliability score of the corresponding evaluator based thereon can be calculated. The score calculated in this way may be provided along with results such as trust, doubt, and distrust.

According to another embodiment, the computing device 200 does not determine the reliability of writing an evaluation sentence based only on the time required for writing the evaluation sentence, but compares the average evaluation time required by other evaluators, the distribution of evaluation time required by other evaluators, etc. can be considered. In addition, the computing device 200 may assign a weight in consideration of the determined propensity of the evaluator so that the evaluator's propensity may be reflected in the time required for writing the evaluation statement.

With respect to the writing time, the computing device 200 may compare the writing time with respect to other evaluators for the evaluation of the same member or another member of the corresponding evaluator. For example, the computing device 200 may recognize the evaluation data of a short required time as a factor of reliability deterioration when the evaluation data of the short required time is concentrated and distributed just before the evaluation deadline as a result of the evaluation required time and evaluation time.

The computing device 200 may evaluate the evaluation sentence of the target evaluator through a combination of both, not just the time required or the time of writing the evaluation sentence by the target evaluator.

The computing device 200 may extract and recommend an appropriate evaluation guide from a database (eg, an in-house DB) based on the target evaluator evaluation and evaluation of the evaluation content.

The computing device 200 may provide a guide or recommended article for insight and quality feedback on the evaluation tendency by comparing and analyzing the evaluation tendency based on the accumulated evaluation preparation data of the target evaluation author with other evaluation authors. .

The computing device 200 may, for example, extract some of the evaluations of members distributed above a preset evaluation score among members of the same rank and present them as a guide.

The computing device 200 may extract data containing content corresponding to a factor in which an evaluation score is reflected low among in-house columns or guides written in-house and provide it to a target evaluator. For example, the computing device 200 may recommend a column titled “How to perform an evaluation suitable for the task of the target member?” among internal columns.

The computing device 200 may collect articles from an external source and provide a guide in a recommended manner. That is, the computing device 200 may analyze a factor with a low evaluation score from an external source, and may bring an article containing content capable of improving the factor and provide it to the target evaluator. The computing device 200 may recommend a 'radical candor' article, for example, if the determined propensity of the author of the target evaluation is tolerant.

The computing device 200 may provide a quantitative evaluation correction system to ensure accuracy based on an existing evaluation tendency when writing a quantitative evaluation of a target evaluator.

To this end, the computing device 200 may first identify the tendency of the target evaluator, for example, and correct the quantitative evaluation result of the evaluation content according to the identified evaluation tendency of the target evaluator.

Computing device 200, for example, when the evaluator of the generous evaluation propensity gives a generous evaluation score compared to other evaluators in the same way as in the existing evaluation, components corresponding to the generosity propensity (eg, evaluation items, generous Quantitative scores for propensity, etc.) can be removed and evaluation scores for the remaining evaluation items can be recommended. As an example, the computing device 200 sets the score for each evaluation item excluding the generosity component when the initial evaluation score of the target evaluator is '5-4-5-5-4'. As shown in 4-3`, it is possible to provide a result window that suggests applying corrections.

The computing device 200 may provide an automatic conversion system for at least one sentence included in an evaluation sentence based on evaluation feedback in relation to the qualitative evaluation correction following the quantitative evaluation correction described above.

To this end, the computing device 200 may calculate word usage characteristic information within an organization (company) including the target evaluation author. In this case, the word usage characteristic information may be extracted by targeting not only the words used in the evaluation within the corresponding organization but also the words used in various documents and texts of the corresponding organization. This may be to ensure anonymity by lowering the risk of exposure of the target evaluator by referring to objectivity.

The computing device 200 may extract frequently used words and sentence structures for each industry of the company and use them for qualitative evaluation. On the other hand, the data to be extracted may be converted into a database. The data to be extracted may include internal reports, external reports, notices, bulletin boards, and the like.

The computing device 200 collects, for example, language or sentence data mainly used for the extraction target data as described above within the company, and is a synonym for a word frequently used in the organization among the words included in the evaluation sentence of the target evaluator. In this case, it can be automatically replaced or a recommendation to be provided can be provided.

The computing device 200 may replace a specific word included in the inscription correction and the written evaluation sentence with a representative word reflecting the characteristics of the company or may provide a replacement recommendation.

The computing device 200 may provide a service so that it is difficult to easily infer who the target evaluator is by substituting sentence structures, sentence endings, words, and the like, with sentence structures, sentence endings, words, etc. frequently used in the company.

The computing device 200 may learn a word or sentence structure mainly used for evaluation based on the evaluation database DB and use it for sentence substitution.

The computing device 200, for example, when a word that is not frequently used in the evaluation field is used, if there is a word mainly used in the evaluation field among synonyms, it is replaced and qualitatively corrected with evaluation data that is more uniform and consistent can be performed or recommended.

In the above, for example, the language mainly used in the evaluation field is not necessarily a word used inside the target company, but also data crawled from an external source such as an external evaluation site or an evaluation column may be used or referenced.

The computing device 200 may provide overall sentence improvement and feedback data in consideration of the qualitative feedback tendency.

Computing device 200, for example, to ensure the anonymity of the target evaluation author, based on the evaluation database (DB) may be a word or sentence structure learning and evaluation sentence substitution mainly used for evaluation, at this time Overall sentence improvement and feedback data may be provided in consideration of sexual feedback tendencies (eg, tone, dialect, specific words, etc.).

The computing device 200 may output, for example, a result feedback that may lead to correction and flexible evaluation by improving the end of a sentence to a flexible tone that is not rigid in the evaluation sentence of the target evaluator determined to be a strict evaluation tendency. . In this case, as the result feedback, for example, a phrase such as “a flexible and soft tone of voice can help to improve a relationship with the person being evaluated!” may be provided.

For example, the computing device 200 extracts a word having a relatively low work relevance from the evaluation sentence of the target evaluator whose reliability or fidelity is evaluated as low as a result of the evaluation propensity determination, and provides a service to improve it by notifying it can provide

When a target evaluator frequently exposes a word having relatively low relevance to evaluation as described above, the computing device 200 may provide a service with feedback data based on the keyword and evaluation relevance. At this time, by recognizing keywords such as “health, alcohol, and tobacco” as the feedback data, guide data such as 'Why not focus on feedback related to work?' can be provided to target evaluators. have.

Such disposition identification, qualitative correction, and quantitative correction for the target evaluator may be performed, for example, on the evaluation text submitted by the target evaluator after the target evaluator is evaluated, or may be provided in advance before the evaluation sentence is submitted. In addition, the correction may be provided so that pre-learning can be performed by making a simulation or a previous evaluation sentence (for the same or different subjects) before the evaluation of the subject.

The above contents may be applied or recommended for evaluation by the target evaluator individually or in a form in which at least two or more contents are combined.

11 is a flowchart illustrating a member evaluation and management service providing method according to another embodiment of the present disclosure. 12 is a diagram illustrating a user interface 1200 for providing member evaluation and management services according to another embodiment of the present disclosure.

Referring to FIG. 11 , in operation 21 , the computing device 200 may receive an evaluation content of an evaluator for a member.

In operations 22 and 23 , the computing device 200 may determine an evaluator's propensity and suitability to the evaluation content from the received evaluator's evaluation content, and generate evaluation content feedback including the determined suitability.

In operation 24, the computing device 200 may correct the received evaluation contents of the evaluator in stages according to the generated evaluation contents feedback.

In operation 25, the computing device 200 may provide the generated evaluation content feedback and the corrected evaluation content.

In the above, step-by-step correction of the evaluation content may be performed through quantitative correction and qualitative correction.

In the above, the computing device 200 may store the evaluation contents of the evaluator for the previously received member.

The computing device 200 may compare the evaluator's disposition with the evaluation scores of at least one or more other evaluators for the same member to compare the evaluator's evaluation scores.

The computing device 200 may determine the degree of suitability for the evaluation content based on a keyword included in the AI-based evaluation content.

The computing device 200 may determine the suitability of the evaluation content based on the evaluation reliability and fidelity through analysis of the evaluation writing pattern included in the evaluation content.

The computing device 200 may extract evaluation guide data from the database based on the generated evaluation content feedback and provide the extracted evaluation guide data.

The computing device 200 may perform quantitative correction of the evaluation content of the evaluator by modifying the evaluation score for each evaluation item included in the evaluation content based on the propensity of the evaluator.

The computing device 200 may perform qualitative correction on the evaluation content of the evaluator by converting at least one of a word and a sentence included in the evaluation content of the evaluator based on the generated evaluation feedback.

In the user interface 1200 shown in FIG. 12 , evaluator information 1210 , subject information 1220 , evaluation content feedback information 1230 , evaluator propensity information 1240 , evaluation content suitability information 1250 , and original evaluation content (1260), an evaluation content correction version 1270, and other information may be included and provided.

Next, a method of recommending a meeting agenda for a personal interview, etc. based on the evaluation contents or evaluation records of the evaluator of the above-mentioned evaluator will be described.

Referring back to FIG. 1 , in relation to the present disclosure, the first terminal 100 in the member evaluation and management system may represent the terminals of all customers participating in the meeting, and the computing device 200 is the first terminal ( 100), information on the user's personal information, work performed, and past evaluation history can be collected, and the second terminal 150 is an external information about the user of the first terminal 100, that is, the company to which the meeting participant belongs. data can be collected.

5 is a block diagram of a processing unit 240 according to another embodiment of the present disclosure.

Referring to FIG. 5 , the processing unit 240 may include a conference topic calculation unit 510 , a related information calculation unit 520 , and the like.

The meeting topic calculation unit 510 may calculate an appropriate meeting topic and agenda based on the information collected through the collection unit.

The related information calculating unit 520 may present information about the meeting participants with respect to the information used for calculating the meeting topic by the meeting topic calculating unit and the topic/agenda.

The computing device 200 analyzes the member's information to provide a main topic of the interview (eg, first meeting data), such as growth, immersion, collaboration, and performance, and a specific talking point (agenda) (eg, the second meeting data) and related information, it can help efficient and effective interview among members.

The computing device 200 according to the present disclosure may provide information including a major topic, an agenda, an interview time, and related information based on information on each of at least two interview participants, information on a relationship between the participants, and organizational data. have.

In the above, the interview participant and organization data can be further divided into external data and internal data. In this case, as described above, the internal data means information input by a user to the service platform and data generated and stored according to the use of the service platform, and the external data may mean data that cannot be obtained through the service platform. .

In the above, the internal data of the participant information may be divided into user input data, user behavior information, and user context information. The user input data means data accumulated by using the service of the service platform, such as job, position, position, year of employment, department, performance, etc. entered by the user to use the service platform, and past evaluation/meeting records. can do. The user behavior information may refer to information extracted based on log data indicating behaviors (page inquiry, evaluation writing, interview application, etc.) occurring as a user uses a service platform. This may also include records of past meetings/interviews and evaluations. The user context information means a recent user's situation, and this situation may mean a major event affecting performance, such as whether or not to be recently evaluated and whether or not to be promoted. In the above, the behavior log data is composed of the type of behavior, the time difference between each behavior, and the target of the behavior, and the processor collects each log and extracts the user's type and additional information through machine learning. For example, in preparing for an interview (1:1) between members who are a part of the service of the service platform, access the service platform in advance to register the discussion agenda (agenda), whether to add an opinion on the other party's agenda, and when to add it (e.g. For example, it is possible to classify the user type by adding just before the interview, adding one hour before the interview, etc.).

External data refers to the user's SNS data other than the service platform, and through this, information such as total career and SNS activity can be extracted and applied to user type classification.

Meanwhile, the relationship information may refer to information indicating the difference or homogeneity between the affiliation, rank, age, and evaluation experience (subject and evaluator) of each interview participant. That is, it may mean a relationship between information about each participant specified above. For example, in the case of an interview between the manager and the deputy heads A and B, the manager/deputy manager is the participant information about A and B, and the fact that A is higher than B is the relationship information.

In addition, with respect to the organization information, the internal data may refer to a configuration (number of organizations, relationships, etc.) of an organization input by a user, a size, and the like. In addition, it may also mean information on evaluation/review that is in progress or has been in progress. External data may mean the existence of an organization, industry, industry group information, media exposure, and the like. The computing device 200 may synthesize various internal/external data to estimate an organization's tendency and philosophy, and may present a suitable major topic or agenda accordingly.

In addition, in relation to the related information, the computing device 200 may provide information corresponding to the presented major topic and agenda. In this case, the information sharing range with respect to the counterpart may be limited in consideration of the relationship information. For example, if an agenda for recent evaluation results is presented for the major topic of evaluation, information on recent and past evaluation results and trends is provided to the other party's direct supervisor (where evaluation results are disclosed), but otherwise, evaluation results Or, in some cases, the fact that you have been recently assessed may not be provided.

The computing device 200 may calculate a degree of similarity/relevance between actions and histories of recent meeting participants with respect to the presented topic and agenda, and may provide related information showing a high degree of similarity to all participants.

Referring to FIG. 6 , the configuration of the meeting topic calculating unit 510 of FIG. 5 will be described as follows.

The conference topic calculation unit 510 may include a data classification unit 610 , a data extraction unit 620 , a data conversion unit 630 , a learning unit 640 , and an operation unit 650 .

The data classification unit 610 may classify the external data collected through the collection unit 230 of FIG. 2 according to a predetermined criterion.

The data extraction unit 620 may extract information on the behavior of using the first terminal 100 indicating the characteristics of the user. Also, the data extractor 620 may extract relationship information indicating a relationship between conference participants (supervisor, motive, etc.).

The data conversion unit 630 may convert the data extracted through the data extraction unit 620 into a learnable form.

The learning unit 640 may learn by combining the personal information input by the user through the first terminal 100 , the result of the data conversion unit 630 , and the result of the data classification unit 610 .

The calculation unit 650 presents the optimal topic among the selected meeting participants based on the result of the learning unit 640, and the information and information synthesis methods referenced for this are as follows, and are not limited to the following elements does not

By inquiring the past meeting history of meeting participants, you can check whether the meeting is repeated and the topic/agenda that has not been resolved or is scheduled to be discussed in the previous meeting.

By synthesizing various information such as personal information of meeting participants, other meeting information, and ongoing evaluation/review, you can check the status/situation of current meeting participants, such as waiting for evaluation results and viewing evaluation results.

It is possible to identify and synthesize topics/agendas used by users with similar orientations, organizations, situations, and personal orientations.

An appropriate topic and agenda may be presented by synthesizing the above aggregated information, personal information of each participant, and relationship information between participants.

Agenda can create and present a recommended agenda pool by considering the similarity between the recommended agenda list developed by the existing service platform and the types of agendas used by other users, including viewpoints, agendas, and types of meeting participants.

Referring to FIG. 6 , the configuration of the related information calculating unit 520 of FIG. 5 will be described as follows.

The related information calculating unit 520 may also include a data classifying unit 610 , a data extracting unit 620 , a data converting unit 630 , a learning unit 640 , and a calculating unit 650 .

The data classification unit 610 may classify the external data collected through the collection unit 230 of FIG. 2 according to a predetermined criterion.

The data extraction unit 620 may extract information on participants of a meeting related to the presented topic/agenda.

The data conversion unit 630 may convert the data extracted through the data extraction unit 620 and the degree of similarity and relevance between the selected topic/agenda into an arithmetic form using a method such as natural language processing.

The learning unit and the calculating units 640 and 650 may calculate information having a high degree of learning and relevance based on the converted data.

The computing device 200 controls the information calculated by selecting the degree of disclosure/privacy and masking based on the relationship information between the participants calculated in the above-mentioned subject/agenda calculation process and applying the calculated information to expose the calculated information to the participants. can

The computing device 200 may perform an evaluation by an evaluator on a member, that is, an evaluator in advance, store a corresponding evaluation record, and may help a meeting between the evaluator and the evaluator based on the stored evaluation record. In this case, the evaluator is not necessarily limited to the evaluator who has performed the evaluation of the subject.

When an input for an evaluation target is received, the computing device 200 may extract and provide evaluation information about a corresponding member, that is, an evaluation record.

The computing device 200 may provide, for example, quantify and provide evaluation target evaluation information, compare it with the evaluation average of the same job group/position, and separately classify and display items in which the evaluation score of the evaluated person is lower than the average value.

In this case, the computing device 200 may include evaluation grounds including reference information in the evaluation information of the members. For example, the reference information may include goal achievement progress, promotion records, past experience, and the like.

The computing device 200 may provide a recommendation service for a meeting agenda based on the evaluation information of the subject to be evaluated.

The computing device 200 may recommend an agenda suitable for a meeting, such as a 1:1 interview, based on the evaluation information of the evaluator. In this case, the computing device 200 may recommend the related item as a meeting agenda, for example, when a low evaluation score is confirmed for a specific item based on evaluation information of the subject to be evaluated.

The computing device 200 may automatically recommend a topic for a meeting with a review subject whose evaluation of the corresponding item is low through machine learning.

The computing device 200 sets, for example, “a method of not stressing within the organization” and “maintaining a smooth relationship with co-workers” as an agenda after the meeting for the evaluation subject whose interpersonal relationship-related items such as friendliness felt by colleagues are low. At least one of the following: “Action guidelines for employees”, “Discussing programs/clubs to promote friendship among in-house members”, “Effective countermeasures when dealing with injustice to superiors”, and “How to approach successors in a friendly manner” As a 1:1 meeting agenda, it can be automatically recommended as an agenda or question related to interpersonal relationships within the company.

As described above, the computing device 200 may provide a list of questions (surveys) based on the evaluation information and the meeting agenda.

When the evaluator selects a specific agenda from among the recommended and provided agendas, the computing device 200 may provide a list of questions for collecting related information.

The computing device 200 may be, for example, when the “how to avoid stress in the organization” agenda is selected, “Q: Is there any behavior of co-workers that makes you feel difficult?”, “Q: Persons who are difficult to approach within the department. Is there a meeting agenda?”, “Q: Do you want to change departments?”, “Q: Is it more efficient to work alone than to work with a large number of people?” A list of questions may be provided as a recommendation service.

The computing device 200 may calculate an item requiring additional information collection based on the evaluation information, and may provide a question list for collecting related information.

For example, according to the evaluation information, when the recent progress toward achieving the goal has decreased and is below the average value, the computing device 200 may display “Q: How do you rate your work efficiency recently?”, “Q: Have you ever thought that you need a break? Have you ever?”, “Q: Is there anything that you would like to see improved in your work?”, “Q: If there are any factors that hinder your work, what would it be?”, “Q: Do you have any wishes for your work-related colleagues? ?”, “Q: Are you losing interest in your job or are you interested in other fields?” Based on such evaluation information, it is possible to extract a list of questions suitable for understanding the member's status and provide an automatic recommendation service.

The computing device 200 may provide the evaluation information automatic processing and output service in consideration of the aforementioned meeting agenda.

The computing device 200 may set the purpose of the meeting directly when the evaluator opens the meeting, or may set the meeting agenda after being recommended according to the above-described method.

The computing device 200 may process and output the evaluation information of the members according to the purpose of establishing a meeting set by the selection of the evaluator.

The computing device 200 may not provide a simple output service of evaluation information, but may process and print the evaluation information according to the purpose of establishing a meeting and/or an agenda of the meeting.

For example, when the purpose of the meeting is “salary negotiation”, the computing device 200 provides information on the goal achievement set by agreement between the organization and the individual within the salary contract period, not the simple goal achievement of the member, based on the member's evaluation information. information can be printed.

Based on the evaluation information of the member, the computing device 200 may output information about how much the member contributed to the goal of the organization within the salary contract period, the organization contribution, the company contribution, and the like.

The computing device 200 refers to or outputs information on whether the goal agreed upon in last year's annual salary contract has been achieved or not, with reference to the next year's goal at the time of last year's annual salary contract of the member and the expected achievement OKR (Object Key Result) and KPI, etc. can service

The computing device 200 may output salary information of another member having an evaluation score similar to that of the member or present it as comparison data.

13 is a flowchart illustrating a member evaluation and management service providing method according to another embodiment of the present disclosure. 14 is a diagram illustrating a user interface for providing member evaluation and management services according to an embodiment of the present disclosure.

Referring to FIG. 13 , in operation 31 , the computing device 200 may receive member information.

In operation 32, the computing device 200 may extract evaluation information from the received member information.

In operation 33 , the computing device 200 may determine a first meeting participant and a second meeting participant based on the extracted evaluation information.

In operation 34, the computing device 200 extracts the determined meeting participant information about the first meeting participant and the second meeting participant and relationship information between the first meeting participant and the second meeting participant, and based on artificial intelligence, First meeting data and second meeting data related to the first meeting data may be generated based on the extracted meeting participant information and the meeting participant relationship information.

In operation 35 , the computing device 200 may provide the generated first and second conference data.

In the user interface 1400 shown in FIG. 14 , evaluator information 1410 , information to be evaluated 1420 , evaluation content feedback information 1430 , recommended agenda and suitability information 1440 , and meeting agenda-based question list information 1460 . ), the meeting agenda, the processed evaluation content 1470, and other information may be included and provided.

According to various embodiments of the present disclosure described above, it is possible to provide an optimal configuration evaluation and management service to the user through the member evaluation and management service in consideration of various matters regarding the organization and the members.

The steps of a method or algorithm described in relation to an embodiment of the present disclosure may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. A software module may 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 in any type of computer-readable recording medium well known in the art to which the present disclosure pertains.

Above, although embodiments of the present disclosure have been described with reference to the accompanying drawings, those of ordinary skill in the art to which the present disclosure pertains can realize that the present disclosure can be embodied in other specific forms without changing its technical spirit or essential features. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

100: first terminal 150: second terminal
200: computing device 210: processor
220: communication unit 230: collection unit
240: processing unit 250: control unit
260: storage

Claims (10)

In the method of providing an AI-based member evaluation feedback service performed by a computing device,
receiving the evaluation contents of the evaluator for the member;
generating evaluation content feedback including the appraiser's disposition and suitability for the evaluation content from the received evaluator's evaluation content;
step-by-step correcting the received evaluation contents of the evaluator according to the generated evaluation contents feedback; and
Comprising the step of providing the generated evaluation contents feedback and the corrected evaluation contents,
Step-by-step correction of the evaluation contents is made through quantitative and qualitative correction,
The degree of suitability for the evaluation content is obtained by extracting a main keyword included in the evaluation content using artificial intelligence, and is calculated based on whether the extracted main keyword is above a threshold value, and the main keyword is determined by the evaluator's disposition , is defined by referring to keywords that are repeatedly used in the evaluation content of evaluators with the same or opposite evaluation tendency or the number of exposures is more than a threshold,
The step of generating the evaluation content feedback,
extracting evaluation guide data from a database based on the generated evaluation content feedback; and providing the extracted evaluation guide data;
The qualitative correction is
Extracting word usage characteristic information within a company, and converting at least one of words and sentences included in the evaluation contents of the evaluator based on the extracted word usage characteristic information, but when converting the sentence, a technology similar to the company or the company It is made through conversion into the representative tone of similar companies in the field,
A method of providing an AI-based member evaluation feedback service. According to claim 1,
Further comprising the step of storing the evaluation content of the evaluator for the received member,
A method of providing an AI-based member evaluation feedback service. 3. The method of claim 2,
The evaluator's tendencies are:
It is determined by comparing the evaluation score of at least one other evaluator for the same member with the evaluation score of the evaluator,
A method of providing an AI-based member evaluation feedback service. delete 4. The method of claim 3,
The degree of suitability for the above evaluation contents is,
It is calculated based on the evaluation reliability and fidelity through the analysis of the evaluation writing pattern included in the evaluation contents,
A method of providing an AI-based member evaluation feedback service. delete 6. The method of claim 5,
Quantitative correction of the evaluation contents of the evaluator is,
It is made by modifying the evaluation score for each evaluation item included in the evaluation content based on the propensity of the evaluator,
A method of providing an AI-based member evaluation feedback service. delete In the AI-based member evaluation feedback service providing device,
Memory; and
a processor in communication with the memory;
The processor is
When an evaluator's evaluation of a member is received, an evaluation content feedback including the propensity of the evaluator and suitability to the evaluation content is generated from the received evaluator's evaluation content, and according to the generated evaluation content feedback, the received evaluator's Correcting the evaluation contents step by step, controlling to provide the generated evaluation contents feedback and corrected evaluation contents,
The processor extracts a main keyword included in the evaluation content using artificial intelligence, calculates a degree of suitability for the evaluation content based on whether the extracted main keyword is above a threshold value, and the main keyword is the evaluator's propensity If this is determined, it is defined with reference to a keyword that is used repeatedly in the evaluation contents of the evaluators of the same evaluation tendency or the opposite evaluation tendency, or the number of exposures is more than a threshold,
The processor extracts the evaluation guide data from the database based on the generated evaluation content feedback, and provides the extracted evaluation guide data,
The processor extracts word usage characteristic information within the company, and converts at least one of a word and a sentence included in the evaluation content of the evaluator based on the extracted word usage characteristic information, but when converting the sentence, the company or the corresponding company performing the qualitative correction by converting the company and the representative tone of a similar company belonging to a similar technology field;
AI-based member evaluation feedback service providing device. In the AI-based member evaluation feedback service providing system,
at least one terminal for uploading evaluation content information for members; and
Comprising a computing device that processes the evaluation content information for the uploaded member,
The computing device is
When an evaluator's evaluation of a member is received, an evaluation content feedback including the propensity of the evaluator and suitability to the evaluation content is generated from the received evaluator's evaluation content, and according to the generated evaluation content feedback, the received evaluator's Comprising a processor that corrects the evaluation contents step by step, and controls to provide the generated evaluation contents feedback and corrected evaluation contents,
The processor extracts a main keyword included in the evaluation content using artificial intelligence, calculates a degree of suitability for the evaluation content based on whether the extracted main keyword is above a threshold value, and the main keyword is the evaluator's propensity If this is determined, it is defined with reference to a keyword that is repeatedly used in the evaluation content of the evaluators of the same evaluation tendency or the opposite evaluation tendency, or the number of exposures is more than a threshold,
The processor extracts the evaluation guide data from the database based on the generated evaluation content feedback, and provides the extracted evaluation guide data,
The processor extracts word usage characteristic information in the company, and converts at least one of a word and a sentence included in the evaluation content of the evaluator based on the extracted word usage characteristic information, but when converting the sentence, the company or the corresponding company performing the qualitative correction by converting the company and the representative tone of a similar company belonging to a similar technology field;
AI-based member evaluation feedback service provision system.

Images