Support device for personnel evaluation

Abstract

To support objective and fair performance evaluations that do not require comparison with other members. [Solution] The personnel evaluation support device 1 according to the present invention comprises: an information acquisition unit 111 that acquires workplace information and subject information relating to the person subject to personnel evaluation; a provisional target provision unit 112 that provides provisional targets relating to the person's personnel evaluation to the person's supervisor, which are generated by the process of inputting the workplace information and person information into a first machine learning model; a final target acquisition unit 113 that acquires final targets relating to the personnel evaluation provided by the supervisor; a provisional evaluation provision unit 114 that provides provisional evaluations relating to the person's personnel evaluation to the supervisor, which are generated by the process of inputting the achievements relating to the final targets and the final targets into a second machine learning model; a final evaluation acquisition unit 115 that acquires final evaluations relating to the personnel evaluation provided by the supervisor; and a machine learning unit 116 that performs machine learning relating to the first machine learning model and the second machine learning model.

Description

Technical Field

[0001] The present invention relates to a device for assisting personnel evaluation.

Background Art

[0002] Regarding the operation management of an organization exemplified by a company or the like, a personnel evaluation for evaluating the performance of members exemplified by employees or the like is carried out. Appropriate personnel evaluation conveys to members what they should do as a member of the organization and enhances the motivation of the members.

[0003] However, there is a concern that a manual personnel evaluation exemplified by a personnel evaluation in which a supervisor sets a goal and evaluates the degree of achievement thereof may not fully convince the member to be evaluated. For example, members may not be convinced by the evaluation because the goal setting or the evaluation of the degree of achievement related to the personnel evaluation depends on the subjectivity of the evaluator and is biased, or there is a variation in the goals determined by human hands. Therefore, a technology for supporting objective personnel evaluation is demanded.

[0004] Regarding technologies for supporting objective personnel evaluation in conventional services and prior arts, etc., Patent Document 1 discloses a sales data storage means for storing sales data indicating the content sold by a salesperson, and when a request for information for evaluating the performance of the sales activities performed by the salesperson is acquired, the sales data stored in the sales data storage means is aggregated for each salesperson, and an information generation means for generating the information, and an information supply means for supplying the information generated by the information generation means to the information terminal of the requester, and a server and the information terminal having the same are disclosed. The information generated by the information generation means of the performance evaluation system is, for example, sales data corrected using sales of selected products, regional economic indices, customer satisfaction, etc.

[0005] The technology described in Patent Document 1 can support the proper evaluation of sales representatives' performance by enabling sales representatives to determine the relative superiority or inferiority of their own performance compared to other sales representatives based on the amount of corrected sales data. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2003-022361 [Overview of the project] [Problems that the invention aims to solve]

[0007] However, evaluations that use comparisons with other members raise concerns that they may cause dissatisfaction among members who are evaluated low simply because their performance is inferior to others, even if all members have achieved excellent results. Furthermore, while such relative evaluations using comparisons may function as a fair evaluation method in organizations composed only of members performing similar tasks, as covered in Patent Document 1, they are difficult to apply to organizations where performance evaluation using the same indicators is difficult, such as organizations that combine a wide variety of tasks to achieve their goals.

[0008] Therefore, the technology described in Patent Document 1 has room for further improvement in that it enables objective and appropriate personnel evaluations that do not require comparison with other members.

[0009] The objective of this invention is to support objective and appropriate personnel evaluations that do not require comparison with other members. [Means for solving the problem]

[0010] As a result of diligent research to solve the above problems, the inventors of this invention found that the above objectives could be achieved not only by utilizing machine learning models not only for evaluating performance but also for generating the performance targets themselves that serve as indicators for such evaluation. Thus, the inventors of this invention completed the present invention.

[0011] One aspect of the present invention provides a personnel evaluation support device comprising: an information acquisition unit that acquires workplace information and subject information relating to persons subject to personnel evaluation; a provisional target provision unit that provides provisional targets relating to the personnel evaluation of the persons, generated by a process of inputting the workplace information and the subject information into a first machine learning model, to the subject's supervisor; a final target acquisition unit that acquires final targets relating to the personnel evaluation provided by the supervisor; a provisional evaluation provision unit that provides provisional evaluations relating to the personnel evaluation of the persons, generated by a process of inputting the performance related to the final targets and the final targets into a second machine learning model, to the supervisor; a final evaluation acquisition unit that acquires final evaluations relating to the personnel evaluation provided by the supervisor; and a machine learning unit that performs machine learning relating to the first machine learning model and the second machine learning model, wherein the machine learning unit performs machine learning relating to the first machine learning model using learning data that includes the workplace information and the subject information as explanatory variables and the final targets as the objective variable, and machine learning relating to the second machine learning model using learning data that includes the performance and the targets as explanatory variables and the final evaluation as the objective variable.

[0012] In this aspect of the present invention, a provisional target related to personnel evaluation is generated by a first machine learning model that performs the machine learning described below relating to the present objective. Since this provisional target is generated by a machine learning model, it can be said to be an objective target.

[0013] However, due to factors such as insufficient pre-training of the machine learning model, there are concerns that the provisional goals generated in this way may not adequately reflect the requirements for personnel evaluation of organizations, etc. Therefore, in this embodiment of the present invention, the provisional goals are provided to the supervisor, and the final goals are obtained after being modified by the supervisor as necessary. Furthermore, in this embodiment of the present invention, machine learning is performed on the first machine learning model using training data including the final goals, so that even more appropriate provisional goals are generated.

[0014] In addition, in this embodiment of the present invention, a provisional evaluation related to personnel evaluation is generated by a second machine learning model that performs the machine learning described later in relation to this evaluation. Since this provisional evaluation is generated by a machine learning model, it can be said to be an objective evaluation.

[0015] However, due to factors such as insufficient pre-training of the machine learning model, there are concerns that the preliminary evaluation generated in this manner may not be a proper evaluation. Therefore, in this embodiment of the present invention, the preliminary evaluation is provided to a supervisor, and a final evaluation is obtained that has been revised by the supervisor as necessary. Furthermore, in this embodiment of the present invention, machine learning is performed on a second machine learning model using training data including the final evaluation, so that an even more appropriate preliminary evaluation is generated.

[0016] Thus, this aspect of the present invention utilizes machine learning models not only for evaluating performance but also for generating performance targets that serve as indicators for such evaluation. Furthermore, by feeding back targets or evaluations modified by supervisors as needed to each machine learning model, it is possible to support objective and appropriate personnel evaluations that do not require comparison with other members. [Effects of the Invention]

[0017] Based on the above, the present invention can support objective and appropriate personnel evaluations that do not require comparison with other members. [Brief explanation of the drawing]

[0018] [Figure 1]FIG. 1 is a block diagram showing an example of the hardware configuration and software configuration of the system S of the present embodiment. [Figure 2] FIG. 2 is an example of the workplace information database 131. [Figure 3] FIG. 3 is a main flowchart showing an example of a preferable flow of the support process executed by the support device 1 of the present embodiment. [Figure 4] FIG. 4 is a flowchart continuing from the previous figure. [Figure 5] FIG. 5 is a flowchart continuing from the previous figure. [Figure 6] FIG. 6 is a flowchart continuing from the previous figure. [Figure 7] FIG. 7 is a flowchart continuing from the previous figure. [Figure 8] FIG. 8 is a flowchart continuing from the previous figure. [Figure 9] FIG. 9 is a flowchart continuing from the previous figure. [Figure 10] FIG. 10 is an example of display on the terminal T.

MODE FOR CARRYING OUT THE INVENTION

[0019] Firstly, although the following disclosures, figures, and / or claims are described either individually or in combination with one or more other aspects, the subject matter of the immediate disclosure is not intended to be limited in that way. That is, the immediate disclosures, figures, and claims are intended to encompass the various aspects described herein, either individually or in one or more combinations with each other. For example, even if the immediate disclosure describes and illustrates the first, second, and third embodiments in such a way that the first embodiment is described and illustrated particularly in relation to the second embodiment, or the second embodiment is described and illustrated only in relation to the third embodiment, the immediate disclosures and illustrations are not limited in that way and may include only the first embodiment, only the second embodiment, only the third embodiment, or one or more combinations of the first, second, and / or third embodiments, such as the first and second embodiments, the first and third embodiments, the second and third embodiments, or the first, second, and third embodiments.

[0020] In this text, the phrase "or" is used to mean a "non-exclusive" arrangement unless explicitly specified otherwise. For example, when we say "item x is A or B," it means either (1) item x is either A or B, or (2) item x is both A and B. In other words, the word "or" is not used to define an "exclusive" arrangement.

[0021] Furthermore, when the phrases "contain at least one" or "contain at least one of the following" are used in the text, they mean that the system or element contains one or more of the elements listed after the phrase. For example, if there are three types of elements, from element 1 to element 3, the phrases "contain at least one" or "contain at least one of the following" are interpreted as any of the following structural arrangements: a device containing element 1, a device containing element 2, a device containing element 3, a device containing element 1 and element 2, a device containing element 1 and element 3, a device containing element 2 and element 3, or a device containing element 1, element 2, and element 3.

[0022] The same interpretation is intended when the phrase "used in at least one of the following" is used in the text. Furthermore, "and / or" as used in the text is used as a linguistic conjunction to indicate that one or more of the listed elements or conditions are included or occur. For example, a device containing the first element, the second element, and / or the third element is interpreted as any of the following structural arrangements: a device containing the first element, a device containing the second element, a device containing the third element, a device containing the first and second elements, a device containing the first and third elements, a device containing the second and third elements, or a device containing the first, second, and third elements.

[0023] Furthermore, the use of the phrase "and / or" in this text signifies a "non-exclusive" arrangement, as stipulated in the Japanese Industrial Standard (JIS) "Format and Preparation Method of Standards Documents JIS Z 8301".

[0024] The following describes in detail an example of an embodiment of the present invention with reference to the drawings.

[0025] <System S> Figure 1 is a block diagram showing an example of the hardware and software configuration of system S in this embodiment. The following is a description of a preferred example of the hardware and software configuration of system S in this embodiment, using Figure 1.

[0026] System S comprises a personnel evaluation support device 1 and a terminal T configured to communicate with the support device 1 via a network N.

[0027] [Support device 1] The support device 1 comprises a control unit 11, a storage unit 13, and a communication unit 14. The type of support device 1 is not particularly limited and may be, for example, a server device, a cloud server, etc.

[0028] [Control Unit 11] The control unit 11 includes a Central Processing Unit (CPU), Random Access Memory (RAM), and Read Only Memory (ROM), among other things.

[0029] The control unit 11 cooperates with at least one of the storage unit 13 and the communication unit 14 as needed. The control unit 11 then implements the software components of the program of this embodiment executed by the support device 1, such as the information acquisition unit 111, provisional target provision unit 112, actual target acquisition unit 113, provisional evaluation provision unit 114, actual evaluation acquisition unit 115, machine learning unit 116, skill map acquisition unit 117, curriculum provision unit 118, workplace specifications acquisition unit 119, personnel system draft provision unit 120, and subsidy support information provision unit 121.

[0030] [Storage section 13] The storage unit 13 is a device on which data and / or files are stored, and has a storage unit that stores data non-temporarily using a hard disk, semiconductor memory, recording medium, and memory card, etc. The storage unit 13 stores programs executed by a microcomputer, a workplace information database 131, machine learning models, etc.

[0031] (Workplace Information Database 131) The workplace information database 131 stores workplace information and individual information related to the person being evaluated for performance. This allows the support device 1 to generate and provide provisional targets for the individual's performance evaluation by inputting the workplace information and individual information into the first machine learning model.

[0032] In this example, workplace information and subject information are stored in the same workplace information database 131, but workplace information and subject information in this embodiment may be stored in separate databases. In order to efficiently perform data processing such as storage, retrieval, and updating, it is preferable that workplace information and subject information be stored in association with information that identifies the information, such as an information ID.

[0033] Workplace information includes information for various levels within the workplace to which the subject belongs. Examples of such level-specific information include company-wide policies and specifications, departmental policies within the company, and team policies within the department. Policies may also include policies associated with periods such as the current month or current period. Company-wide specifications include, for example, the company's industry, sales volume, or number of employees. By including workplace specifications in the workplace information, the support device 1 can generate and provide support information such as a draft of the workplace's personnel system or assistance in applying for subsidies by inputting the workplace specifications into the fourth machine learning model.

[0034] The subject information includes, for example, the subject's performance, their rank (no title, team leader, section chief, department head, etc.), their skill map, or information related to their previous evaluations. Performance or previous evaluations may be associated with periods such as last week, last month, or last quarter. Performance includes, for example, performance indicated by numerical values ​​or numerical symbols (e.g., symbols indicating ranks such as "A, B, C"), and performance indicated by text. The skill map is, for example, data listing the skills that the subject has acquired or mastered from among the skills exemplified by workplace qualifications, etc. By including the skill map in the subject information, the support device 1 can generate and provide a human resource development curriculum for the subject by processing the skill map into a third machine learning model.

[0035] Figure 2 shows an example of a workplace information database 131. This example database stores workplace information relating to the entire company, identified by information ID "D0001", workplace information relating to the accounting department, identified by information ID "D0011", subject information relating to subject "Tanaka △△ (Accounting Department)", identified by information ID "D0111", subject information relating to subject "Suzuki △△ (Accounting Department)", etc.

[0036] The workplace information relating to the entire company, identified by information ID "D0001," includes information such as industry "manufacturing," sales volume "△△ billion yen," employee size "△△ people," and company-wide policies such as "promoting digitalization of operations" and "promoting paperless operations." This allows support device 1 to generate and provide provisional targets for personnel evaluation that reflect the company's industry, size, or overall policies. Furthermore, this enables support device 1 to identify subsidies appropriate to the company's industry or size, and to generate and provide support information to assist in applying for those subsidies.

[0037] The workplace information related to the department (Accounting Department) identified by information ID "D0011" includes information such as the department's policies, including "Renewal of the billing management system," "Renewal of the document management system," and "Establishment of an invoice workflow." This allows support device 1 to generate and provide provisional targets for personnel evaluation that reflect the department's policies.

[0038] The information pertaining to the individual identified by information ID "D0111," "Tanaka △△ (Accounting Department)," includes information such as: achievements "Requests for renewal of the billing management system have been compiled" and "△△ cases of invoice processing handled (previous month)," rank "Team Leader," skill map "△△ qualification (Level 2)," and previous evaluation "Leadership in △△ work...." This allows support device 1 to generate and provide provisional targets for personnel evaluation that reflect the individual's achievements, rank, skill map, or previous evaluation. Furthermore, this enables support device 1 to generate and provide a human resource development curriculum that reflects the skill map, etc.

[0039] The information about the person identified by information ID "D0112," namely "Suzuki △△ (Accounting Department)," includes details such as: achievements ("Requests for renewal of the billing management system have been compiled," "△△ invoice processing cases handled (previous month)"), rank ("No managerial position"), skill map ("△△ qualification (Level 4)"), and previous evaluation ("Further efforts needed regarding △△ work..."). This allows support device 1 to generate and provide provisional targets for personnel evaluation that reflect the person's achievements, rank, skill map, or previous evaluation. Furthermore, this enables support device 1 to generate and provide a human resource development curriculum that reflects the skill map, etc.

[0040] In this way, the achievements, grades, skill maps, and previous evaluations that differ for each individual are stored in the workplace information database 131 and utilized by the first machine learning model. As a result, the support device 1 can, for example, generate a provisional goal that is even higher than that of "Suzuki △△" for an individual "Tanaka △△" whose achievements, grades, skill maps, and previous evaluations are superior to those of other individuals such as "Suzuki △△," and provide it to the terminal T used by the individual's supervisor. Furthermore, the support device 1 can generate and provide more appropriate provisional goals that reflect the overall workplace policy, departmental policy, etc.

[0041] The storage unit 13 stores at least various data necessary for processing using the first machine learning model and processing using the second machine learning model. Preferably, this data includes the machine learning model itself, but it may also include various data necessary for using the machine learning model stored and executed on an external device (for example, data related to the Application Programming Interface (API)). Preferably, the storage unit 13 further stores various data necessary for processing using the third machine learning model, the fourth machine learning model, or the fifth machine learning model.

[0042] (First machine learning model) The first machine learning model is a machine learning model that generates provisional targets for the performance evaluation of an individual by inputting workplace information and individual information related to the individual being evaluated. The first machine learning model may be a model that has common parts with the second, third, fourth, or fifth machine learning models described later.

[0043] In order to generate provisional goals that appropriately reflect workplace information and target information, it is preferable that the first machine learning model is pre-trained using training data that includes workplace information and target information as explanatory variables and goals based on this information as the dependent variable. This allows the first machine learning model to deduce and generate appropriate provisional goals based on pre-training for inputs with complex correlations, such as workplace goals and the target's performance, grade, and skill map. For example, such correlations include the fact that performance and skills unrelated to workplace goals have a small influence on goal setting, while performance and skills that are deeply related to workplace goals have a large influence on goal setting.

[0044] To enable the use of workplace information and target information written in natural language as input, the first machine learning model preferably includes a language model. To enable the use of workplace information and target information written in a wide variety of natural language expressions as input, the language model preferably is a large-scale language model, such as ChatGPT (registered trademark). The large-scale language model understands input such as workplace information and target information written in a wide variety of natural language expressions through emergent capabilities related to natural language acquired from, for example, a large-scale neural network with a large number of parameters of 100 billion or more, or a massive amount of training data of, for example, 100 billion tokens or more, and contributes to the generation of personnel evaluation goals etc. that reflect this understanding.

[0045] (Second machine learning model) The second machine learning model is a machine learning model that generates a preliminary performance evaluation for an individual based on input of their achievements and goals related to those goals. The second machine learning model may be a model that has common parts with the first machine learning model, the third machine learning model, the fourth machine learning model, or the fifth machine learning model described later.

[0046] In order to generate a provisional evaluation that appropriately reflects the performance related to the objective and the objective, it is preferable that the second machine learning model is pre-trained using training data that includes performance related to the objective and the objective as explanatory variables, and the evaluation based on said performance and objective as the dependent variable. This allows the second machine learning model to deduce and generate an appropriate provisional evaluation based on pre-training for inputs with complex correlations between objectives, performance, etc. For example, such correlations include the fact that performance unrelated to the objective has little influence on the evaluation, while performance closely related to the objective has a large influence on setting the evaluation.

[0047] To enable the use of achievements and goals described in natural language as input, the second machine learning model preferably includes a language model. To enable the use of achievements and goals described in a wide variety of natural language expressions as input, the language model is preferably a large-scale language model, such as ChatGPT®.

[0048] (Third machine learning model) The third machine learning model is a machine learning model that generates a human resource development curriculum for a target person by inputting the target person's skill map. The third machine learning model may be a model that has common parts with the first machine learning model, the second machine learning model, the fourth machine learning model described later, or the fifth machine learning model. In order to reflect other circumstances related to the target person's skill map, it is preferable that the third machine learning model generates a human resource development curriculum that reflects the input of the target person's achievements, rank, or previous evaluations in addition to the skill map.

[0049] In order to generate a human resource development curriculum that appropriately reflects inputs such as skill maps, it is preferable that the third machine learning model is pre-trained using training data that includes input data such as skill maps as explanatory variables and a human resource development curriculum based on said input as the dependent variable. This allows the third machine learning model to deduce and generate an appropriate human resource development curriculum based on pre-training for inputs with complex correlations, such as skill maps containing a large number of skills. For example, one such correlation is that the priority for developing skills unrelated to currently possessed skills is low, while the priority for developing skills that are closely related to currently possessed skills is high.

[0050] To enable the use of inputs such as skill maps written in natural language, the third machine learning model preferably includes a language model. To enable the use of skill maps written in a wide variety of natural language expressions as input, the language model is preferably a large-scale language model, such as those exemplified by ChatGPT®.

[0051] (Fourth Machine Learning Model) The fourth machine learning model is a machine learning model that generates a draft of a workplace personnel system by inputting workplace specifications related to personnel evaluation. The fourth machine learning model may be a model that has common parts with the first machine learning model, the second machine learning model, the third machine learning model, or the fifth machine learning model described later.

[0052] To generate a draft of a workplace personnel system, it is preferable that the fourth machine learning model is pre-trained using training data that includes workplace parameters related to personnel evaluation as explanatory variables and a draft of a workplace personnel system related to those parameters as the dependent variable. This allows the fourth machine learning model to deduce and generate an appropriate draft based on pre-training for inputs with complex correlations, such as parameters containing a large number of elements. For example, such correlations include increasing the number of grades in industries where the company size is small but the chain of command in operations is important, and decreasing the number of grades in industries where the company size is small but the autonomy of individuals in operations is important.

[0053] The draft personnel system referred to here includes, for example, drafts of a grading system, evaluation sheets, and wage tables. This allows the fourth machine learning model to generate a draft personnel system that integrates a grading system, evaluation sheets used for evaluation at each grade, and wage tables corresponding to each grade.

[0054] To enable the use of specifications described in natural language as input, the fourth machine learning model preferably includes a language model. To enable the use of specifications described in a wide variety of natural language expressions as input, the language model is preferably a large-scale language model, such as ChatGPT®.

[0055] (Fifth Machine Learning Model) The fifth machine learning model is a machine learning model that generates support information to assist workplaces in applying for subsidies by inputting workplace specifications. The fifth machine learning model may be a model that has common parts with the first, second, third, or fourth machine learning models.

[0056] In order to generate support information that assists in applying for appropriate subsidies based on workplace characteristics, it is preferable that the fifth machine learning model is pre-trained using training data that includes workplace characteristics as explanatory variables and support information that assists in applying for appropriate subsidies based on those characteristics as the dependent variable. This allows the fifth machine learning model to deduce and generate appropriate support information based on pre-training for inputs with complex correlations, such as characteristics containing many elements. An example of such a correlation is the correlation between the size of a company and its industry, and support information related to applying for subsidies for small and medium-sized enterprises in that industry.

[0057] To help apply for appropriate subsidies, it is preferable that the fifth machine learning model, upon input of workplace specifications, generates support information that includes information indicating appropriate subsidies for workplaces with those specifications. In this case, it is preferable that the fifth machine learning model is pre-trained using training data that includes workplace specifications as explanatory variables and support information that includes information indicating appropriate subsidies for those specifications as the dependent variable.

[0058] To enable the use of specifications described in natural language as input, the fifth machine learning model preferably includes a language model. To enable the use of specifications described in a wide variety of natural language expressions as input, the language model is preferably a large-scale language model, such as ChatGPT®.

[0059] [Communications Section 14] The communication unit 14 is not particularly limited as long as it connects the support device 1 to the network N and enables communication. Examples of the communication unit 14 include a network card compatible with the Ethernet standard and a communication device compatible with wireless LAN.

[0060] [Network N] The type of network N is not particularly limited as long as it enables communication with the support device 1, etc. Examples of network N include the internet, a mobile phone network, a wireless LAN, etc.

[0061] [Terminal T] Terminal T is used by supervisors or other personnel within the organization conducting performance evaluations. Terminal T performs processes such as instructing support device 1 to provide provisional goals or provisional evaluations, retrieving and displaying the generated provisional goals or provisional evaluations from support device 1, providing the inputted final goals or final evaluations to support device 1 based on the displayed provisional goals or provisional evaluations, and instructing support device 1 to provide support information for talent development curricula, personnel system drafts, or grant application support information. The types of terminal T include, for example, mobile terminals such as smartphones, tablet terminals, and laptop computers, and stationary terminals such as desktop computers.

[0062] [Main flowchart for support processing] Figure 3 is a main flowchart showing an example of a preferred flow of support processing performed by the support device 1 of this embodiment. Figures 4, 5, 6, 7, 8, and 9 are flowcharts that continue from the previous figures. The following is a description of an example of a preferred flow of support processing using Figures 3 to 9.

[0063] The support process generates provisional goals related to the performance evaluation of the target individual, provides the generated provisional goals, obtains the final goals provided by the supervisor to whom the provisional goals were provided, and executes a series of processes using the final goals in the machine learning of the first machine learning model. Steps S1 to S8 are an example of this process.

[0064] [Step S1: Determine whether to generate a provisional target] The control unit 11, in cooperation with the storage unit 13 and the communication unit 14, performs a process to determine whether to generate provisional targets related to the performance evaluation of the subject (provisional target generation determination step). If it is determined that targets should be generated, the control unit 11 moves the process to step S2. If it is determined that targets should not be generated, the control unit 11 moves the process to step S5.

[0065] In the provisional target generation determination step, the control unit 11 determines to generate a provisional target if, for example, a command is issued from terminal T instructing it to provide a provisional target exemplified by a command related to AI target setting.

[0066] [Step S2: Obtain workplace information and target information] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the information acquisition unit 111. The control unit 11 then performs the process of acquiring workplace information and individual information related to the person being evaluated by the information acquisition unit 111 (information acquisition step). The control unit 11 then moves the process to step S3.

[0067] In the information acquisition step, the information acquisition unit 111 acquires workplace information and target person information, for example, as exemplified in the section on workplace information database 131, from the workplace information database 131. In order to add data to or update data stored in the workplace information database 131, the information acquisition unit 111 may acquire some or all of the workplace information or target person information from terminal T.

[0068] [Step S3: Generate a provisional target] The control unit 11 works in cooperation with the memory unit 13 and the communication unit 14 to execute the provisional target provision unit 112. The control unit 11 then uses the provisional target provision unit 112 to input the workplace information and target information acquired in the information acquisition step into the first machine learning model, thereby executing a process to generate provisional targets related to the personnel evaluation of the target person (provisional target generation step). The control unit 11 then moves the process to step S4.

[0069] If the first machine learning model includes a language model, the provisional target provision unit 112 may generate the provisional targets by inputting prompts to the language model, which include workplace information, target information, and instructions for generating targets.

[0070] If the number of provisional targets to be generated is specified, the provisional target provision unit 112 may generate that number of provisional targets. In order to provide provisional targets for each policy separately determined by the entire workplace, department, group, etc., the provisional target provision unit 112 may generate provisional targets for each policy indicated by the workplace information or target person information.

[0071] [Step S4: Provide a provisional goal] The control unit 11 executes a process to provide the provisional target generated in the provisional target generation step by the provisional target provision unit 112 (provisional target provision step). The control unit 11 moves the process to step S5. The provisional target provision unit 112 provides the provisional target to terminal T, for example.

[0072] Through the series of processes from step S1 to step S4, the supervisor using terminal T can obtain a provisional goal related to the performance evaluation of the employee. The supervisor can then use the provisional goal as the final goal, or modify the provisional goal and use it as the final goal.

[0073] [Step S5: Determine whether to achieve this objective] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the main target acquisition unit 113. The control unit 11 then performs a process to determine whether to acquire the main target related to the provisional target mentioned above using the main target acquisition unit 113 (main target acquisition determination step). If it is determined that the target should be acquired, the control unit 11 moves the process to step S6. If it is determined that the target should not be acquired, the control unit 11 moves the process to step S9.

[0074] In this target acquisition determination step, the control unit 11 determines, for example, that it will acquire the target when it receives data of the target transmitted from terminal T.

[0075] [Step S6: Achieve this objective] The control unit 11 executes the process of acquiring the target described above using the target acquisition unit 113 (target acquisition step). The control unit 11 then moves the process to step S7.

[0076] [Step S7: Store this objective] The control unit 11 uses the target acquisition unit 113 to perform the process of storing the aforementioned target in the storage unit 13 (target storage step). The control unit 11 then moves the process to step S8.

[0077] [Step S8: Run machine learning on the first machine learning model] The control unit 11 works in cooperation with the memory unit 13 and the communication unit 14 to execute the machine learning unit 116. The control unit 11 then performs a process to run machine learning on the first machine learning model using training data that includes the aforementioned workplace information and target information as explanatory variables and the aforementioned objective as the target variable (first machine learning step). The control unit 11 then moves the process to step S9.

[0078] In the first machine learning step, the machine learning unit 116 performs the machine learning described above, and as necessary, the supervisor modifies the provisional goal to reflect the final goal in the first machine learning model. As a result, the support device 1 generates a more appropriate provisional goal that reflects the final goal as modified as necessary.

[0079] After the objective is provided, the support process of this embodiment generates a provisional evaluation based on the results and the objective, obtains the final evaluation provided by the supervisor who provided the provisional evaluation, and executes a series of processes to use the final evaluation in the machine learning of the second machine learning model. Steps S9 to S16 are an example of this process.

[0080] [Step S9: Determine whether to generate a preliminary evaluation] The control unit 11, in cooperation with the storage unit 13 and the communication unit 14, performs a process to determine whether to generate a provisional evaluation related to the performance evaluation of the subject (provisional evaluation generation determination step). If it is determined that an evaluation should be generated, the control unit 11 moves the process to step S10. If it is determined that an evaluation should not be generated, the control unit 11 moves the process to step S13.

[0081] In the preliminary evaluation generation determination step, the control unit 11 determines to generate a preliminary evaluation if, for example, a command is issued from terminal T instructing it to provide a preliminary evaluation exemplified by a command related to AI evaluation.

[0082] [Step S10: Achieve results and achieve this goal] The control unit 11 executes a process to acquire the performance and objectives of the person subject to the personnel evaluation using the information acquisition unit 111 (performance acquisition step). The control unit 11 then moves the process to step S11.

[0083] In the performance acquisition step, the information acquisition unit 111 acquires, for example, the performance exemplified in the section on workplace information database 131 from the workplace information database 131. In order to add data to or update data stored in the workplace information database 131, the information acquisition unit 111 may acquire some or all of the performance from terminal T.

[0084] [Step S11: Generate preliminary evaluation] The control unit 11 works in cooperation with the memory unit 13 and the communication unit 14 to execute the provisional evaluation provision unit 114. Then, the control unit 11 performs a process to generate a provisional evaluation related to the personnel evaluation of the subject by inputting the above-mentioned performance and target into the second machine learning model (provisional evaluation generation step). The control unit 11 moves the process to step S12.

[0085] If the second machine learning model includes a language model, the provisional evaluation provision unit 114 may generate the above-mentioned provisional evaluation by inputting prompts to the language model that include instructions for generating results, main goals, and evaluations.

[0086] [Step S12: Provide a preliminary assessment] The control unit 11 executes a process to provide the provisional evaluation generated in the provisional evaluation generation step by the provisional evaluation provision unit 114 (provisional evaluation provision step). The control unit 11 moves the process to step S13. The provisional evaluation provision unit 114 provides the provisional evaluation to terminal T, for example.

[0087] Through the series of processes from step S9 to step S12, the supervisor using terminal T can obtain a preliminary evaluation related to the performance evaluation of the subject. The supervisor can then use the preliminary evaluation as is for the final evaluation, or modify the preliminary evaluation and use it as the final evaluation.

[0088] [Step S13: Determine whether to obtain this evaluation] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to execute the main evaluation acquisition unit 115. The control unit 11 then performs a process to determine whether to acquire the main evaluation related to the provisional evaluation described above using the main evaluation acquisition unit 115 (main evaluation acquisition determination step). If it is determined that the evaluation should be acquired, the control unit 11 moves the process to step S14. If it is determined that the evaluation should not be acquired, the control unit 11 moves the process to step S17.

[0089] In this evaluation acquisition determination step, the control unit 11 determines, for example, to acquire this evaluation when it receives evaluation data transmitted from terminal T.

[0090] [Step S14: Obtain this evaluation] The control unit 11 executes the process of acquiring the above-mentioned evaluation using the evaluation acquisition unit 115 (evaluation acquisition step). The control unit 11 then moves the process to step S15.

[0091] [Step S15: Store this evaluation] The control unit 11 uses the evaluation acquisition unit 115 to perform the process of storing the above-mentioned evaluation in the storage unit 13 (evaluation storage step). The control unit 11 then moves the process to step S16.

[0092] [View all ratings] To facilitate the confirmation of evaluations for multiple members by supervisors, the support process preferably includes a procedure (evaluation display step) for displaying a list of the aforementioned provisional or final evaluations for multiple members. To make the evaluation stages easy to see at a glance, it is preferable that the procedure displays the evaluations in different colors. To make it easy to understand the correspondence between grades and evaluations, it is preferable that the procedure displays the grades alongside the evaluations. To make the grade stages easy to see at a glance, it is preferable that the procedure displays the grades in different colors. To make the display easy for supervisors to read, it is preferable that the criteria for these color codings are customizable.

[0093] [Step S16: Run machine learning on the second machine learning model] The control unit 11 works in cooperation with the memory unit 13 and the communication unit 14 to execute the machine learning unit 116. The control unit 11 then executes a process to run machine learning on a second machine learning model using training data that includes the above-mentioned performance and this objective as explanatory variables and the above-mentioned this evaluation as the objective variable (second machine learning step). The control unit 11 then moves the process to step S17.

[0094] In the second machine learning step, the machine learning unit 116 performs the machine learning described above, and reflects the final evaluation, which has been revised by the supervisor as needed, into the second machine learning model. As a result, the support device 1 generates a more appropriate preliminary evaluation that reflects the final evaluation that has been revised as needed.

[0095] The support process preferably includes a series of processes for generating and providing a human resource development curriculum for the target individual based on the target individual's skill map, etc. Steps S17 to S20 are an example of such a process.

[0096] [Step S17: Determine whether to generate a talent development curriculum] The control unit 11, in cooperation with the memory unit 13 and the communication unit 14, performs a process to determine whether to generate a personnel training curriculum for the target person (curriculum generation determination step). If it is determined that a curriculum should be generated, the control unit 11 moves the process to step S18. If it is not determined that a curriculum should be generated, the control unit 11 moves the process to step S21.

[0097] In the curriculum generation determination step, the control unit 11 determines to generate a human resource development curriculum when a command is issued from terminal T instructing it to provide a human resource development curriculum exemplified by, for example, an AI curriculum generation command.

[0098] [Step S18: Obtain Skill Map] The control unit 11, in cooperation with the storage unit 13 and the communication unit 14, executes the skill map acquisition unit 117. The control unit 11 then executes the process of acquiring the skill map relating to the above-mentioned target person using the skill map acquisition unit 117 (skill map acquisition step). The control unit 11 then moves the process to step S19.

[0099] In the skill map acquisition step, the skill map acquisition unit 117 acquires, for example, the skill map exemplified in the section on the workplace information database 131 from the workplace information database 131. In order to add data to or update data stored in the workplace information database 131, the skill map acquisition unit 117 may acquire part or all of the skill map from terminal T.

[0100] [Regarding the display of the skill map] The support process preferably includes a step for displaying the skill map acquired in the skill map acquisition step (skill map display step). In this step, in order to facilitate comparison between multiple members, it is preferable that the step displays the skill maps of multiple members in a list-like manner. For example, such a display may be a table display with members and skills to be displayed in the skill map on different axes.

[0101] In the list display, it is preferable to include the rank in the procedure so that the correspondence between rank and skill can be easily understood. To make the skill and rank levels easy to see at a glance, it is preferable to color-code the skills or ranks in the procedure. To make the display easy for supervisors to understand, it is preferable that the criteria for these color codings be customizable.

[0102] [Step S19: Generate the Human Resources Development Curriculum] The control unit 11 works in cooperation with the memory unit 13 and the communication unit 14 to execute the curriculum provision unit 118. The control unit 11 then performs the process of generating the aforementioned human resource development curriculum for the target person by inputting the input data such as the skill map into the third machine learning model using the curriculum provision unit 118 (human resource development curriculum generation step). The control unit 11 then moves the process to step S20.

[0103] In the human resource development curriculum generation step, it is preferable that the curriculum provision unit 118 uses the subject's achievements, rank, or previous evaluation as additional input data to the third machine learning model, in addition to the skill map. This allows the curriculum provision unit 118 to generate a human resource development curriculum that reflects the additional input data. If the third machine learning model includes a language model, the curriculum provision unit 118 may generate the above-mentioned human resource development curriculum by inputting prompts including instructions for generating the skill map and the human resource development curriculum into the language model.

[0104] [Step S20: Provide a human resource development curriculum] The control unit 11 executes a process to provide the human resource development curriculum generated in the human resource development curriculum generation step by the curriculum provision unit 118 (curriculum provision step). The control unit 11 moves the process to step S21. The curriculum provision unit 118 provides the human resource development curriculum to terminal T, for example.

[0105] [Regarding supervisory revisions and machine learning in talent development curricula] In order for supervisors to correct any shortcomings in the automatically generated curriculum, the process of generating and providing the curriculum preferably includes, similar to the process of obtaining the final objectives based on the provided provisional objectives, a procedure for obtaining the revised curriculum from a terminal T used by the supervisor, after the curriculum generated by the third machine learning model has been provided.

[0106] In this case, it is preferable that the series of processes further include a step of performing machine learning on a third machine learning model using training data that includes a skill map as an explanatory variable and a human resource development curriculum obtained from terminal T as an objective variable. This allows the third machine learning model to generate a more appropriate human resource development curriculum.

[0107] The support process preferably includes a series of processes that generate and provide a draft of the personnel system for the workplace based on the workplace specifications, etc. Steps S21 to S24 are an example of such a process.

[0108] [Step S21: Determine whether to generate a personnel system] The control unit 11, in cooperation with the memory unit 13 and the communication unit 14, performs a process to determine whether to generate a draft of the personnel system related to the workplace (personnel system draft generation determination step). If it is determined that a draft should be generated, the control unit 11 moves the process to step S22. If it is not determined that a draft should be generated, the control unit 11 moves the process to step S25.

[0109] In the personnel system draft generation determination step, the control unit 11 determines to generate a draft if, for example, a command is issued from terminal T instructing it to provide a draft of a personnel system exemplified by a command related to AI personnel system generation.

[0110] [Step S22: Obtain workplace specifications] The control unit 11, in cooperation with the storage unit 13 and the communication unit 14, executes the workplace data acquisition unit 119. Then, the control unit 11 executes the process of acquiring the above-mentioned workplace data using the workplace data acquisition unit 119 (first workplace data acquisition step). The control unit 11 then moves the process to step S23.

[0111] In the first workplace specification acquisition step, the workplace specification acquisition unit 119 acquires workplace specifications from the workplace information database 131, for example, as exemplified in the section on workplace information database 131. In order to add data to or update data stored in the workplace information database 131, the workplace specification acquisition unit 119 may acquire some or all of the workplace specifications from terminal T.

[0112] [Step S23: Generate the personnel system] The control unit 11 works in cooperation with the memory unit 13 and the communication unit 14 to execute the personnel system draft provision unit 120. The control unit 11 then uses the personnel system draft provision unit 120 to input the input data, such as the workplace specifications, into the fourth machine learning model, thereby generating a draft of the workplace personnel system (personnel system draft generation step). The control unit 11 then moves the process to step S24.

[0113] In the personnel system draft generation step, the personnel system draft provision unit 120 generates a personnel system draft that includes, for example, a grading system, evaluation sheets, and wage tables. As a result, the personnel system draft provision unit 120 can generate a personnel system draft that integrates a grading system, evaluation sheets used for evaluation for each grade, and wage tables corresponding to the grades.

[0114] If the fourth machine learning model includes a language model, the personnel system draft provision unit 120 may generate the above-mentioned personnel system draft by inputting prompts to the language model that include instructions for generating workplace specifications and a personnel system draft.

[0115] [Step S24: Provide a personnel system] The control unit 11 executes a process to provide the personnel system draft generated in the personnel system draft creation step by the personnel system draft provision unit 120 (personnel system draft provision step). The control unit 11 moves the process to step S25. The personnel system draft provision unit 120 provides the personnel system draft to terminal T, for example.

[0116] [Regarding user modifications and machine learning in the HR system] To allow users to correct any shortcomings in the automatically generated system, the process of generating and providing a draft personnel system preferably includes, similar to the process of obtaining the final objectives based on the provided provisional objectives, a procedure for obtaining the personnel system, modified by the user as needed, from terminal T after the draft personnel system generated by the fourth machine learning model is provided.

[0117] In this case, it is preferable that the series of processes further include a procedure for performing machine learning on a fourth machine learning model using training data that includes workplace specifications as explanatory variables and the personnel system obtained from terminal T as the dependent variable. This allows the fourth machine learning model to generate a more appropriate draft of the personnel system.

[0118] [Regarding payroll management] The support process preferably includes a procedure (salary increase assessment step) for automatically assessing the periodic salary increases of the target individuals based on the personnel system and the target individuals' performance related to the generated draft. This procedure may include, for example, a procedure for automatically assessing the periodic salary increases of the target individuals based on the correspondence between performance and periodic salary increases included in the personnel system, and the processing based on the target individuals' performance.

[0119] To achieve objective assessments across a wide range of performance metrics, it is preferable that the process utilizes a machine learning model. To ensure appropriate assessments, it is preferable that the machine learning model is pre-trained using training data that includes performance metrics and corresponding relationships as explanatory variables, and the assessed periodic salary increases as the dependent variable.

[0120] Furthermore, the support process preferably includes a procedure (bonus assessment step) for automatically assessing the bonus of the target person based on the personnel system and the target person's performance related to the generated draft. This procedure may include, for example, a procedure for automatically assessing the bonus of the target person based on the correspondence between performance and bonus included in the personnel system, and the target person's performance.

[0121] In order to achieve objective assessments across a wide variety of performance data, it is preferable that the process utilizes a machine learning model. To ensure appropriate assessments, it is preferable that the machine learning model is pre-trained using training data that includes performance data and corresponding relationships as explanatory variables, and the assessed bonus as the dependent variable.

[0122] In order to enable integration with payroll management software, it is preferable that the support process includes a procedure for outputting the aforementioned assessment results, etc., in a manner that can be used by the payroll management software.

[0123] The support process preferably includes a series of processes that generate and provide support information to assist a workplace in applying for subsidies based on the workplace's specifications and other factors. Steps S25 to S29 are an example of such a process.

[0124] [Step S25: Determine whether to support the grant application] The control unit 11 works in cooperation with the storage unit 13 and the communication unit 14 to perform a process to determine whether to support the grant application (support information generation determination step). If it determines that support is to be provided, the control unit 11 moves the process to step S26. If it does not determine that support is to be provided, the control unit 11 returns the process to step S1 and repeats the process from step S1 to step S29.

[0125] In the support information generation determination step, the control unit 11 determines to support the grant application when a command is issued from terminal T to provide support information for grant applications, for example, a command related to AI grant applications.

[0126] [Step S26: Obtain workplace specifications] The control unit 11, in cooperation with the storage unit 13 and the communication unit 14, executes the workplace data acquisition unit 119. Then, the control unit 11 executes the process of acquiring the above-mentioned workplace data using the workplace data acquisition unit 119 (second workplace data acquisition step). The control unit 11 moves the process to step S27. The second workplace data acquisition step may be the same as the first workplace data acquisition step.

[0127] If the fifth machine learning model is configured to generate support information that includes information indicating appropriate subsidies for a workplace having the above-described parameters, the above-described series of processes preferably includes a series of steps to determine whether there are appropriate subsidies for a workplace having the above-described parameters, and if so, to generate and provide support information. Steps S27 to S29 are an example of such steps. This allows the support device 1 to determine appropriate subsidies for the workplace and generate support information to assist in applying for such subsidies.

[0128] [Step S27: Determine if there are appropriate grants available] The control unit 11, in cooperation with the storage unit 13 and the communication unit 14, executes the subsidy support information provision unit 121. The control unit 11 then inputs the above-mentioned parameters into the fifth machine learning model via the subsidy support information provision unit 121, generates information indicating appropriate subsidies that the workplace related to those parameters should receive, and performs a process to determine whether there are appropriate subsidies based on that information (subsidy availability determination step). If it is determined that there are subsidies, the control unit 11 moves the process to step S28. If it is not determined that there are subsidies, the control unit 11 returns the process to step S1 and repeats the process from step S1 to step S29.

[0129] [Step S28: Generate support information] The control unit 11, using the grant support information provision unit 121, executes a process to generate support information to assist the workplace in applying for grants by inputting the aforementioned workplace specifications and other input data into the fifth machine learning model (support information generation step). The control unit 11 then moves the process to step S29.

[0130] If information indicating an appropriate grant is generated in the grant availability determination step, it is preferable that the grant support information provision unit 121 generates support information to assist in applying for the appropriate grant by further inputting input data including said information into the fifth machine learning model. This allows the support device 1 to generate support information to assist in applying for a grant that has been determined to be appropriate for the workplace.

[0131] [Step S29: Provide support information] The control unit 11 executes a process to provide the support information generated in the support information generation step by the grant support information provision unit 121 (grant support information provision step). The control unit 11 returns the process to step S1 and repeats the process from step S1 to step S29. The grant support information provision unit 121 provides support information to terminal T, for example.

[0132] [Regarding user modifications and machine learning in support information related to grants] To allow users to correct any shortcomings in the automatically generated data, the series of processes for generating and providing support information preferably includes, similar to the series of processes for acquiring the final goal based on the provided provisional goal, a procedure for acquiring the support information corrected by the user as needed from terminal T after the support information generated by the fifth machine learning model is provided.

[0133] In this case, it is preferable that the series of processes further include a procedure for performing machine learning on a fifth machine learning model using training data that includes workplace specifications as explanatory variables and support information obtained from terminal T as the target variable. This allows the fifth machine learning model to generate even more appropriate support information.

[0134] [Managing interviews] To support the recording and use of records related to meetings with subordinates, the support process preferably includes a procedure (meeting management step) for recording and making available the meetings between superiors and subordinates. This procedure preferably includes a procedure for recording the audio or video call related to the meeting. To alleviate the subordinate's anxiety regarding the meeting, this procedure preferably includes a procedure for sharing notes written by the superior about the meeting with the subordinate.

[0135] [Effects of support processing] By performing the support processes described above, the support device 1 of this embodiment generates provisional targets related to personnel evaluation using a first machine learning model that performs machine learning related to the target (steps S1 to S3). Since these provisional targets are generated by a machine learning model, they can be considered objective targets.

[0136] However, due to factors such as insufficient pre-training of the machine learning model, there are concerns that the provisional goals generated in this way may not adequately reflect the requirements of the organization regarding personnel evaluation. On the other hand, the support process described above provides the provisional goals to the supervisor, and the supervisor obtains the final goals that have been modified as needed (steps S4 to S7). Furthermore, the support process described above performs machine learning on the first machine learning model using the training data including the final goals, so that even more appropriate provisional goals are generated (step S8).

[0137] In addition, the support process described above generates a provisional evaluation related to the personnel evaluation using a second machine learning model, which performs the machine learning described later in relation to this evaluation. Since this provisional evaluation is generated by a machine learning model, it can be said to be an objective evaluation (steps S9 to S11).

[0138] However, due to factors such as insufficient pre-training of the machine learning model, there are concerns that the preliminary evaluation generated in this way may not be a proper evaluation. Therefore, in the support process described above, the preliminary evaluation is provided to the supervisor, and the supervisor corrects it as needed to obtain the final evaluation (steps S12 to S15). Furthermore, in the support process described above, a second machine learning model is trained using the training data including the final evaluation to generate an even more appropriate preliminary evaluation (step S16).

[0139] Thus, the support device 1 that performs the above-mentioned support processing utilizes machine learning models not only for performance evaluation but also for generating the performance targets themselves that serve as indicators for said evaluation. Furthermore, by feeding back the targets and evaluations modified by the supervisor as needed to each machine learning model, it can support objective and appropriate personnel evaluations that do not require comparison with other members.

[0140] Furthermore, the support processing described above can take the form of generating a human resource development curriculum based on the skill map of the target person (steps S17 to S20). In this embodiment, the support device 1 further comprises a skill map acquisition unit 117 that acquires the skill map relating to the target person, and a curriculum provision unit 118 that provides the human resource development curriculum for the target person, which is generated by the process of inputting the skill map into a third machine learning model. The third machine learning model is pre-trained using learning data that includes the skill map as an explanatory variable and the human resource development curriculum for the person relating to the skill map as an objective variable. As a result, the support device 1 of this embodiment in this embodiment can provide a human resource development curriculum linked to personnel evaluation.

[0141] In addition, the above-described support process can take the form of generating a draft personnel system based on workplace specifications (steps S21 to S24). In this embodiment, the support device 1 further comprises a workplace specification acquisition unit 119 that acquires workplace specifications related to personnel evaluation, and a personnel system draft provision unit 120 that provides a draft of the workplace personnel system generated by a process of inputting these specifications into a fourth machine learning model. The fourth machine learning model is pre-trained using training data that includes workplace specifications as explanatory variables and the personnel system related to the workplace as the dependent variable. As a result, the support device 1 of this embodiment in this embodiment can support the construction of a personnel system linked to personnel evaluation.

[0142] Furthermore, the support processing described above can take the form of supporting the application for subsidies based on workplace specifications (steps S25 to S29). In this form, the support device 1 further comprises a workplace specification acquisition unit 119 that acquires workplace specifications related to personnel evaluation, and a subsidy support information provision unit 121 that provides support information to support the application for subsidies that the workplace can receive, which is generated by the process of inputting the workplace specifications into a fifth machine learning model. The fifth machine learning model is pre-trained using training data that includes workplace specifications as explanatory variables and information to support the application for subsidies that the workplace can receive as an objective variable. As a result, the support device 1 of this embodiment in this form can support the work of the workplace where personnel evaluation is conducted.

[0143] Therefore, the support device 1 that performs the above-mentioned support processing can help ensure that objective and appropriate personnel evaluations are conducted without requiring comparison with other members.

[0144] <Example of use of support device 1> The following is an example of how to use the support device 1 of this embodiment.

[0145] [Registration of workplace information and target person information] Supervisors and the individuals being evaluated register workplace information and individual information related to performance evaluations via terminal T, etc., in the support device 1. The support device 1 stores the registered workplace information and individual information in the workplace information database 131, etc.

[0146] [AI Goal Setting Instructions] Figure 10 shows an example of the display on terminal T. In this example, the supervisor clicks the "AI Goal Setting" button displayed in the upper right corner of terminal T and instructs support device 1 to display a pop-up menu related to providing provisional goals for subordinate "Tanaka △△". Support device 1 displays a pop-up menu on terminal T with the heading "Implementing AI Goal Setting" prompting the user to set parameters related to providing provisional goals.

[0147] In this example, the supervisor instructs the employee to check "Work Objectives" in the pop-up menu's objective category and provide provisional objectives related to work. The supervisor also instructs the employee to check "Competency Objectives" in the pop-up menu's objective category and provide objectives related to competence. Furthermore, the supervisor instructs the employee to uncheck "Behavioral Objectives" in the pop-up menu's objective category and not provide provisional objectives related to behavioral objectives.

[0148] Furthermore, the supervisor sets the number of goals for each category to "4" in the pop-up menu. The supervisor then specifies that the "Select relevant information" user interface element in the pop-up menu should refer to company-wide goals, departmental goals, job grade, skill map, and previous evaluation sheet.

[0149] The supervisor then enters any special notes in the "Other Special Notes" field as needed and clicks "Execute" at the bottom of the pop-up menu to instruct the support device 1 to set AI goals. In response to this instruction, the support device 1 generates provisional goals and a personnel development curriculum through processing using the first machine learning model, etc. The support device 1 then provides this generated information to the supervisor by displaying it on terminal T. In this example, the generated provisional goal displayed is "Promote the digitalization of operations and digitize five or more operations," which corresponds to the company-wide goal (workplace-wide policy) theme "Digitalization." In addition, in this example, corresponding provisional goals are also displayed for each of the departmental goals (policies of the department to which the employee belongs) "Troubleshooting," "Goal Setting and Quantification," and "Coordination with Related Departments."

[0150] [Setting this objective] The supervisor sets the final goal for subordinate "Tanaka △△" by referring to the provisional goal provided in the above procedure and making appropriate modifications. Because a provisional goal is provided, the supervisor can set the final goal quickly and easily without having to consider the goal from scratch. Support device 1 acquires the set final goal and stores it in the workplace information database 131, etc. Support device 1 also performs machine learning using the said final goal.

[0151] [Revision of the human resource development curriculum] The supervisor sets up a human resource development curriculum for their subordinate "Tanaka △△" by referring to the human resource development curriculum provided through the above procedure and making appropriate modifications. Since the curriculum drafted by support device 1 is provided, the supervisor can set up the curriculum quickly and easily without having to consider the curriculum from scratch. Support device 1 acquires the set human resource development curriculum and stores it in the workplace information database 131, etc. Support device 1 also performs machine learning using the human resource development curriculum.

[0152] [Instruction for generating a provisional evaluation] After the work period related to the objective set in the above procedure has ended, the supervisor instructs support device 1 to provide a preliminary evaluation of the performance during that period and the objective. Support device 1 generates a preliminary evaluation by processing the performance and the objective using a second machine learning model, and provides the generated preliminary evaluation to the supervisor by displaying it on terminal T.

[0153] [Settings for this evaluation] The supervisor refers to the preliminary evaluation provided through the above procedure and sets the final evaluation for subordinate "Tanaka △△" by making appropriate modifications. Because a preliminary evaluation is provided, the supervisor can set the final evaluation quickly and easily without having to consider the evaluation from scratch. Support device 1 acquires the set final evaluation and stores it in the workplace information database 131, etc. Support device 1 also performs machine learning using the said final evaluation.

[0154] [Drafting of personnel system proposal] The user provides workplace specifications to the support device 1 and instructs the support device 1 to draft a personnel system based on those specifications. The support device 1 generates a draft personnel system by inputting the specifications into a fourth machine learning model and provides the generated draft personnel system to the user by displaying it on terminal T. The user then proceeds with reviewing the personnel system by referring to the draft.

[0155] [Applying for Grants] The user provides workplace specifications to the support device 1 and instructs the support device 1 to provide support information to assist in applying for subsidies based on those specifications. The support device 1 generates support information by inputting the specifications into a fifth machine learning model and provides the generated support information to the user by displaying it on terminal T. The user then proceeds with applying for subsidies by referring to this support information.

[0156] Within the scope of the concept of this invention, those skilled in the art can conceive of various modifications and alterations. Therefore, such modifications and alterations are understood to fall within the scope of this invention. For example, any addition, deletion, or design change of components, or addition, omission, or modification of processes, made by a person skilled in the art to the above-described embodiments, is also included within the scope of this invention, as long as it retains the essence of this invention. [Explanation of Symbols]

[0157] S System 1 Support equipment 11 Control Unit 111 Information Acquisition Department 112 Provisional Target Provision Department 113 Book Objective Acquisition Department 114 Provisional Evaluation Department 115 This Evaluation Acquisition Department 116 Machine Learning Department 117 Skill Map Acquisition Section 118 Curriculum Provision Department 119 Workplace Specification Acquisition Department 120 Personnel system draft provision department 121 Grant Support Information Department 13 Storage section 131 Workplace Information Database 14 Communications Department N Network T terminal

Claims

1. An information acquisition unit that acquires workplace information and information about individuals subject to performance evaluation, A provisional goal provision unit provides provisional goals related to the performance evaluation of the subject, which are generated by the process of inputting the aforementioned workplace information and the aforementioned subject information into a first machine learning model, to the subject's supervisor. The objective acquisition unit acquires the objectives related to the performance evaluation provided by the aforementioned supervisor, A provisional evaluation provision unit provides the supervisor with a provisional evaluation of the personnel evaluation of the subject, which is generated by the process of inputting the results related to the aforementioned objective and the aforementioned objective into a second machine learning model. The evaluation acquisition unit acquires the evaluation related to the personnel evaluation provided by the aforementioned supervisor, A machine learning unit that performs machine learning related to the first machine learning model and the second machine learning model, Equipped with, The machine learning unit performs machine learning related to the first machine learning model using training data that includes the workplace information and the target information as explanatory variables and the objective variable, and machine learning related to the second machine learning model using training data that includes the performance and the objective as explanatory variables and the objective variable. A support device for personnel evaluation.

2. A skill map acquisition unit that acquires a skill map relating to the aforementioned target person, A curriculum provision unit that provides a human resource development curriculum for the target person, generated by the process of inputting the aforementioned skill map into a third machine learning model, Furthermore, The third machine learning model has undergone pre-training using training data that includes a skill map as an explanatory variable and a human resource development curriculum for the personnel related to the skill map as the dependent variable. The support device according to claim 1.

3. A workplace data acquisition unit that acquires workplace data related to the aforementioned personnel evaluation, A personnel system draft provision unit provides a draft of the personnel system for the workplace, which is generated by the process of inputting the aforementioned parameters into a fourth machine learning model, Furthermore, The fourth machine learning model has undergone pre-training using training data that includes workplace specifications as explanatory variables and the personnel system related to the workplace as the dependent variable. The support device according to claim 1.

4. The system further includes a grant support information provision unit that provides support information to assist the workplace in applying for grants, generated by the process of inputting the aforementioned workplace specifications into a fifth machine learning model. The fifth machine learning model has undergone pre-training using training data that includes workplace specifications as explanatory variables and information supporting the application for subsidies received by the workplace as the dependent variable. The support device according to claim 3.

Images