Attendance management device, attendance management method, and attendance management program

Abstract

To enhance employee engagement with the organization based on their work performance. [Solution] The input unit 21 inputs attendance data 31, which shows the work status of the subject, into a pre-trained question generation model 41. The output unit 22 outputs model questions, which are questions that confirm the subject's status, generated by the question generation model 41 in response to the attendance data 31 input by the input unit 21. The response receiving unit 23 receives responses to the model questions.

Description

[[Technical Field]]

[0001] The present disclosure relates to technology for increasing worker engagement with an organization. [[Background Art]]

[0002] In organizations such as companies, it has come to be considered important to increase worker engagement with the organization. A state where worker engagement with the organization is high refers to a state where a trusting relationship is established between the organization and its workers. In other words, a state where worker engagement with the organization is high refers to a state where workers have affection, pride, or similar feelings toward the organization. To increase worker engagement with the organization, it is necessary to create a comfortable working environment. To create a comfortable working environment, it is important to reduce worker stress.

[0003] Patent Document 1 describes measuring the career development status of employees based on their responses to a questionnaire, and determining a career interview implementation policy in accordance with the career development status. Accordingly, Patent Document 1 aims to promote employee career development and improve engagement. [[Prior Art Documents]] [[Patent Documents]]

[0004] [[Patent Document 1]] Japanese Unexamined Patent Application Publication No. 2023-99907 [[Summary of the Invention]] [[Problem to be Solved by the Invention]]

[0005] The technology described in Patent Document 1 aims to improve employee engagement by conducting appropriate career interviews. Understanding employees' careers is also effective in improving engagement. In addition to careers, working conditions such as working hours are related to employee stress, so it is effective in improving engagement for managers to understand employees' working conditions and to try to bring employees' working conditions closer to their wishes. This disclosure aims to enhance employee engagement with the organization based on their work performance. [Means for solving the problem]

[0006] The attendance management device related to this disclosure is An input unit that inputs attendance data showing the work status of the target person into a pre-trained question generation model, An output unit outputs a model question, which is a question to confirm the status of the subject, generated by the question generation model in response to the attendance data input by the input unit. It is equipped with. [Effects of the Invention]

[0007] This disclosure outputs model questions that correspond to attendance data and confirm the status of the individual. By asking the individual these model questions, communication between managers and employees regarding their work status may be facilitated. As a result, employees may feel that their managers are trying to understand their work status, which can easily lead to improved engagement. [Brief explanation of the drawing]

[0008] [Figure 1] Configuration diagram of the attendance management device 10 according to Embodiment 1. [Figure 2] Flowchart of the user-side processing according to Embodiment 1. [Figure 3] A diagram illustrating a specific example of input to the question generation model 41 according to Embodiment 1. [Figure 4]An explanatory diagram of a specific example of a model question according to the first embodiment. [Figure 5] A configuration diagram of an attendance management apparatus 10 according to the second embodiment. [Figure 6] A flowchart of template generation processing according to the second embodiment. [Figure 7] An explanatory diagram of a specific example of an input to a template generation model 42 according to the second embodiment. [Figure 8] An explanatory diagram of a specific example of a template question 33 according to the second embodiment. [Figure 9] A flowchart of administrator-side processing according to the second embodiment. [Figure 10] A flowchart of target person-side processing according to the second embodiment. [Figure 11] A configuration diagram of an attendance management apparatus 10 according to the third embodiment. [Figure 12] A flowchart of administrator-side processing according to the third embodiment. [Figure 13] An explanatory diagram of a specific example of an input to an advice generation model 43 according to the third embodiment. [Figure 14] An explanatory diagram of advice according to the third embodiment. [Figure 15] A configuration diagram of an attendance management apparatus 10 according to the fourth embodiment. [Figure 16] A flowchart of evaluation processing according to the fourth embodiment. [Figure 17] An explanatory diagram of a specific example of an input to an evaluation model 44 according to the fourth embodiment. [Figure 18] A diagram showing an output example of evaluation values according to the fourth embodiment. [Figure 19] A diagram showing an example of evaluation values for each pair of an answer and a question according to Modification 3. [Figure 20] A configuration diagram of an attendance management apparatus 10 according to the fifth embodiment. [Figure 21] A flowchart of evaluation advice generation processing according to the fifth embodiment. [Figure 22] An explanatory diagram of adjustment processing according to the fifth embodiment. [Figure 23] An explanatory diagram of a specific example of an input to an advice generation model 43 according to the fifth embodiment. [Figure 24] An explanatory diagram of advice according to Embodiment 5. [Figure 25] An explanatory diagram of a specific example of input to the template generation model 42 according to Embodiment 6. [Figure 26] An explanatory diagram of a specific example of the template question 33 according to Embodiment 6. DETAILED DESCRIPTION OF THE INVENTION

[0009] Embodiment 1. ***Description of Configuration*** With reference to FIG. 1, the configuration of the attendance management device 10 according to Embodiment 1 will be described. The attendance management device 10 is a computer. The attendance management device 10 includes hardware including a processor 11, a memory 12, a storage 13, and a communication interface 14. The processor 11 is connected to other pieces of hardware via signal lines and controls these pieces of hardware.

[0010] The processor 11 is an IC that performs processing. IC is an abbreviation for Integrated Circuit. Specific examples of the processor 11 include a CPU, a DSP, and a GPU. CPU is an abbreviation for Central Processing Unit. DSP is an abbreviation for Digital Signal Processor. GPU is an abbreviation for Graphics Processing Unit.

[0011] The memory 12 is a storage device that temporarily stores data. Specific examples of the memory 12 include SRAM and DRAM. SRAM is an abbreviation for Static Random Access Memory. DRAM is an abbreviation for Dynamic Random Access Memory.

[0012] Storage 13 is a storage device for storing data. A concrete example of storage 13 is an SSD. SSD stands for Solid State Drive. Alternatively, storage 13 may be a portable recording medium such as an SD® memory card, CompactFlash®, NAND flash, flexible disk, optical disk, compact disk, Blu-ray® disc, or DVD. SD stands for Secure Digital. DVD stands for Digital Versatile Disk.

[0013] Communication interface 14 is an interface for communicating with external devices. Specific examples of communication interface 14 include Ethernet®, USB, and HDMI® ports. USB stands for Universal Serial Bus. HDMI stands for High-Definition Multimedia Interface.

[0014] The attendance management device 10 comprises an input unit 21, an output unit 22, and a response receiving unit 23 as functional components. The functions of each functional component of the attendance management device 10 are implemented by software. Storage 13 stores programs that implement the functions of each functional component of the attendance management device 10. These programs are loaded into memory 12 by the processor 11 and executed by the processor 11. This enables the implementation of the functions of each functional component of the attendance management device 10.

[0015] The storage 13 stores attendance data 31 and question answer data 32.

[0016] The attendance management device 10 is connected to a pre-trained question generation model 41 via a communication interface 14. A pre-trained model is a type of generative AI. AI stands for Artificial Intelligence. A pre-trained model may be constructed using algorithms such as BERT or GPT. BERT stands for Bidirectional Encoder Representations from Transformers. GPT stands for Generative Pretrained Transformer. A pre-trained model may be constructed using a combination of multiple algorithms, including these.

[0017] The attendance management device 10 is connected to a user terminal 51 and an administrator terminal 52 via a communication interface 14. The user terminal 51 is a PC or similar terminal used by employees or other people working within the organization. The administrator terminal 52 is a PC or similar terminal used by an administrator who manages employees. PC stands for Personal Computer.

[0018] In Figure 1, the question generation model 41 was located outside the attendance management device 10. However, the question generation model 41 may also be included as a functional component of the attendance management device 10. Hereafter, not limited to the question generation model 41, trained models may be located outside the attendance management device 10, or the attendance management device 10 may include them as functional components.

[0019] In Figure 1, only one processor 11 was shown. However, there may be multiple processors 11, and multiple processors 11 may work together to execute programs that implement each function.

[0020] ***Explanation of operation*** Referring to Figures 2 to 4, the operation of the attendance management device 10 according to Embodiment 1 will be explained. The operation procedure of the attendance management device 10 according to Embodiment 1 corresponds to the attendance management method according to Embodiment 1. Furthermore, the program that implements the operation of the attendance management device 10 according to Embodiment 1 corresponds to the attendance management program according to Embodiment 1. The operation of the attendance management device 10 according to Embodiment 1 includes processing on the part of the person concerned.

[0021] Referring to Figure 2, the user-side processing according to Embodiment 1 will be explained. The processing on the subject's side is executed after the subject, who is an employee of the organization, enters their attendance data for the previous day. Here, we assume that the processing in Figure 2 is executed when the subject enters their attendance data for the previous day. Here, "the previous day" refers to the most recent working day for each subject.

[0022] (Step S11: Input Processing) The input unit 21 inputs attendance data 31, which shows the work status of the subject, into the question generation model 41, which is a trained model. The attendance data 31 shows the work status, including the start time, end time, leave / absence, working hours, etc., for each day of the past reference period. In this process, the input unit 21 takes attendance data 31 into consideration and inputs instructions to the question generation model 41 to generate questions for the subject in order to understand the subject's work status. The input unit 21 also inputs constraints to the question generation model 41, specifying which questions should not be generated. Examples of questions that should not be generated include questions that may lower the subject's motivation and questions that may constitute harassment. The input unit 21 also inputs the output format of the questions to the question generation model 41.

[0023] Referring to Figure 3, a specific example of input to the question generation model 41 according to Embodiment 1 will be explained. In Figure 3, the #Instruction sheet contains instructions to generate questions for the subject to understand their status, taking into account the attendance data 31. The #Constraints list questions that should not be generated. The #Attendance data includes the start time, end time, leave / absence, and working hours for each day of the past reference period. In Figure 3, separate from the #Attendance data, the #Previous day's attendance input is set up, and the #Previous day's attendance input includes the start time, end time, leave / absence, and working hours for the previous day. Therefore, the #Attendance data contains information for each day from two days prior to the past reference period. The #Output format specifies the output format for the questions. In Figure 3, it is specified that there is one question, and the answer should have a total of five options: four choices plus "other".

[0024] The reason why we have a separate entry for the previous day's attendance data is to generate questions that prioritize the information from the previous day within the historical reference period. Alternatively, instead of setting up separate attendance input for the previous day, or in addition to setting up attendance data separately, the input unit 21 may be instructed to generate questions that prioritize the previous day's information. The input unit 21 may also be instructed to generate questions that prioritize recent information over older information.

[0025] When attendance data 31 is entered in step S11, the question generation model 41, following instructions, generates model questions to understand the subject's status based on the attendance data 31. Questions to understand the subject's status based on the attendance data 31 might include, for example, if the subject works a lot of overtime, a question such as, "You seem to be working a lot of overtime; are you feeling tired?" or, if the subject has worked on a holiday, a question such as, "How is your work progressing?" might be asked to check on their work progress.

[0026] (Step S12: Output processing) The output unit 22 obtains model questions, which are questions to confirm the status of the subject, generated by the question generation model 41 in response to the attendance data 31 entered in step S11. The output unit 22 outputs the model questions to the user terminal 51 used by the subject.

[0027] Referring to Figure 4, a specific example of the model question related to Embodiment 1 will be explained. Figure 4 shows an example of a model question generated by the question generation model 41 when the information shown in Figure 3 is input to the question generation model 41. In Figure 3, it was specified that there should be one question and a total of five answers: four choices plus "other". Therefore, in Figure 4, one question with a total of five choices is generated.

[0028] (Step S13: Response reception processing) The response receiving unit 23 receives input from the subject for the model question output in step S12. The response receiving unit 23 writes the pair of model question and answer, along with the subject's identification information and date and time, to the storage 13 as question and answer data 32. Administrators, such as the supervisor of the subject, can understand the subject's status by referring to the model question and answer pairs written to the question and answer data 32 from the administrator terminal 52.

[0029] ***Effects of Embodiment 1*** As described above, the attendance management device 10 according to Embodiment 1 outputs model questions, which are questions to confirm the status of the person in question, in response to attendance data. By referring to the model question and answer pairs for the person in question, it is possible to facilitate smoother communication between the manager and the person in question regarding their work status. As a result, the person in question will feel that the manager is trying to understand their work status, and their engagement is likely to improve.

[0030] ***Other configurations*** <Example 1> If the process shown in Figure 2 is executed when the previous day's attendance data is entered, the process shown in Figure 2 will be executed every time the employee works. As a result, a model question will be generated each time the employee works, and the employee will have to answer the model question. If there is little change in the employee's attendance data 31, there is a possibility that similar model questions will be generated every day. Therefore, in step S11, the input unit 21 may be instructed to take past questions as input and generate different questions. This prevents similar model questions from being generated every day.

[0031] Alternatively, in step S11, the input unit 21 may be instructed to input past questions and generate questions of different types. This will result in the generation of model questions of different types each day. The input unit 21 may also be instructed to generate questions of randomly selected types without inputting past questions. This will also generally result in the generation of model questions of different types each day. One possible approach to the assessment questions is to establish multiple evaluation categories related to engagement with the organization to which an employee belongs. These evaluation categories could include, for example, tolerance for overtime, proactiveness in taking leave, work-life balance, skill development, and positivity. Tolerance for overtime is an indicator of whether an employee can complete their work without working overtime. Proactiveness in taking leave is an indicator of whether an employee actively takes leave. Work-life balance is an indicator of whether an employee has achieved harmony between work and life. Skill development is an indicator of whether an employee is working to improve their skills or abilities. Positivity is an indicator of whether an employee maintains a positive attitude.

[0032] Embodiment 2. Embodiment 2 differs from Embodiment 1 in that it accepts input of designated questions 34, which are questions to confirm the status of the subject, from the subject's administrator. Embodiment 2 explains this difference, and omits explanations of the same points.

[0033] ***Explanation of the structure*** Referring to Figure 5, the configuration of the attendance management device 10 according to Embodiment 2 will be described. The attendance management device 10 differs from the attendance management device 10 shown in Figure 1 in that it includes a question receiving unit 24 as a functional component. Furthermore, the attendance management device 10 differs from the attendance management device 10 shown in Figure 1 in that template questions 33 and specified questions 34 are stored in the storage 13. Finally, the attendance management device 10 differs from the attendance management device 10 shown in Figure 1 in that it is connected to a template generation model 42, which is a trained model, via a communication interface 14.

[0034] ***Explanation of operation*** Referring to Figures 6 to 10, the operation of the attendance management device 10 according to Embodiment 2 will be explained. The operation of the attendance management device 10 according to Embodiment 2 includes template generation processing, administrator-side processing, and user-side processing.

[0035] Referring to Figure 6, the template generation process according to Embodiment 2 will be described. The template generation process is performed during the initial setup of the attendance management device 10. The template generation process may also be performed when updating the template questions 33 that were set during the initial setup.

[0036] (Step S21: Input Processing) The input unit 21 inputs multiple question types into the template generation model 42, which is a pre-trained model. As explained in Modification 2, the question types can be multiple evaluation types related to engagement with the organization to which the person belongs. In this process, the input unit 21 inputs instructions to the template generation model 42 to generate questions to confirm the status of the subject for each category. The input unit 21 also inputs instructions to the template generation model 42 to generate advice for the administrator based on the purpose of each question. Furthermore, the input unit 21 inputs the output format of the questions to the question generation model 41.

[0037] Referring to Figure 7, a specific example of input to the template generation model 42 according to Embodiment 2 will be explained. Figure 7 shows that the #Instruction sheet contains instructions for generating questions to check the status of each target person and instructions for generating advice for managers. The #Template contains the types of questions. In Figure 7, the types of questions are converted and displayed in the desired format. In Figure 7, the types of questions include items such as wanting to know the status of skill development, wanting to know if work-life balance is being maintained, and wanting to encourage taking vacations. The #Output format contains the output format for the questions.

[0038] When the question type and other information are entered in step S21, the template generation model 42 generates questions as template questions 33 for each type, following the instructions.

[0039] (Step S22: Output processing) The output unit 22 obtains template questions 33, which are questions for each of the multiple question types generated by the template generation model 42 in response to the multiple question types input in step S21. The output unit 22 outputs template questions 33 for each category to the administrator terminal 52 used by the administrator. As a result, the template questions 33 are displayed on the display device of the administrator terminal 52, allowing the administrator to review the template questions 33.

[0040] Referring to Figure 8, a specific example of the template question 33 according to Embodiment 2 will be explained. Figure 8 shows an example of a template question 33 generated by the template generation model 42 when the information shown in Figure 7 is input to the template generation model 42. One or more template questions 33 are generated for each type of question. In Figure 8, template questions 33 are generated for each item, such as wanting to know about skill development, wanting to know if work-life balance is being maintained, and wanting to encourage vacation taking. In addition, advice for managers is generated for each template question 33.

[0041] (Step S23: Memory processing) The output unit 22 writes the template questions 33 for each type obtained in step S22 to the storage 13.

[0042] Furthermore, when the template question 33 is output to the administrator terminal 52, the output unit 22 may accept modifications to the template question 33 from the administrator. The output unit 22 may then write the modified template question 33 to the storage 13.

[0043] Referring to Figure 9, the administrator-side processing according to Embodiment 2 will be explained. Administrator-side processing is executed at a time specified by the administrator, targeting each person managed by the administrator. Here, it is assumed that administrator-side processing is executed when approving the attendance data of the target person for the previous day.

[0044] (Step S31: Answer confirmation process) The response receiving unit 23 retrieves the questions asked by the subject when they entered their attendance data for the previous day, along with their answers, from the storage unit 13. The response receiving unit 23 then outputs the questions and answers to the administrator terminal 52. This allows the administrator to verify the questions asked by the subject the previous day and their answers.

[0045] (Step S32: Question determination process) The question reception unit 24 accepts the administrator's choice of whether or not to enter a designated question 34. The designated question 34 is a question entered by the administrator to confirm the status of the person concerned. For example, if there is something the administrator wants to confirm with the person concerned based on the previous day's questions and answers, the designated question 34 will be entered. If the question receiving unit 24 selects "Enter a designated question 34", it proceeds to step S33. On the other hand, if the question receiving unit 24 selects "Do not enter a designated question 34", it terminates the process.

[0046] (Step S33: Question reception processing) The question receiving unit 24 outputs the template questions 33 for each type generated in the template generation process to the administrator terminal 52. At this time, the question receiving unit 24 may also output advice to the administrator to the administrator terminal 52. The question receiving unit 24 then accepts designated questions 34 that have been edited and generated by the administrator from one of the template questions 33. The question receiving unit 24 may also accept designated questions 34 created by the administrator, not based on the template questions 33.

[0047] (Step S34: Question memory processing) The question receiving unit 24 writes the designated question 34 received in step S33 to the storage 13 along with the subject's identification information, date and time, and a flag indicating whether it is unprocessed.

[0048] Referring to Figure 10, the user-side processing according to Embodiment 2 will be explained. The user-side processing corresponds to the processing shown in Figure 2. Similar to the processing shown in Figure 2, the user-side processing is executed after the user, who is an employee of the organization, has entered their attendance data for the previous day. Here, it is assumed that the user-side processing is executed at the time the user enters their attendance data for the previous day.

[0049] The processes in steps S42 and S43 are the same as the processes in steps S11 and S12 in Figure 2.

[0050] (Step S41: Specification determination process) The input unit 21 determines whether or not a specified question 34 about the subject, for which the flag is set to "unprocessed," is stored in the storage unit 13. If the specified question 34 is not stored in the input unit 21, the process proceeds to step S42. On the other hand, if the specified question 34 is stored in the input unit 21, the process proceeds to step S44.

[0051] (Step S44: Processing the output of the specified question) The output unit 22 reads the designated questions 34 for the subject that have the flag set to "unprocessed" from the storage unit 13. Then, the output unit 22 outputs the designated questions 34 to the user terminal 51 used by the subject. At this time, the output unit 22 changes the flag for the read designated questions 34 to "processed".

[0052] (Step S45: Processing received response) The response receiving unit 23 receives input from the subject to the model question output in step S43 or the designated question 34 output in step S44. The response receiving unit 23 writes the pair of the model question or designated question 34 and the answer, along with the subject's identification information and date and time, to the storage 13 as question and answer data 32.

[0053] Through the above process, the questions and answers are output to the administrator terminal 52 on the next business day after the subject enters attendance data and answers the questions, and the administrator determines whether or not the specified question 34 needs to be entered. If the specified question 34 is entered, the next time the subject enters attendance data, the specified question 34 is output to the user terminal 51.

[0054] ***Effects of Embodiment 2*** As described above, the attendance management device 10 according to Embodiment 2 receives a designated question 34 from the administrator and outputs it to the user terminal 51. This makes it possible for the administrator to ask any question they wish to confirm with the person concerned.

[0055] ***Other configurations*** <Modification 2> In Embodiment 2, it was assumed that only one question would be output per day. Therefore, only one of either the model question or the designated question 34 was output. However, if it is acceptable to output multiple questions per day, then if the designated question 34 is input, both the model question and the designated question 34 may be output.

[0056] Embodiment 3. Embodiment 3 differs from Embodiments 1 and 2 in that it generates advice for administrators. Embodiment 3 explains this difference, while omitting explanations of the same points. Embodiment 3 describes a case in which a modification has been made to Embodiment 2. However, it is also possible to modify Embodiment 1.

[0057] ***Explanation of the structure*** Referring to Figure 11, the configuration of the attendance management device 10 according to Embodiment 3 will be described. The attendance management device 10 differs from the attendance management device 10 shown in Figure 5 in that it is connected to the advice generation model 43, which is a trained model, via a communication interface 14.

[0058] ***Explanation of operation*** Referring to Figures 12 to 14, the operation of the attendance management device 10 according to Embodiment 3 will be described. The administrator-side processing differs from that of Embodiment 2. The template generation process and the user-side processing are the same as in Embodiment 2.

[0059] Referring to Figure 12, the administrator-side processing according to Embodiment 3 will be explained. Similar to Embodiment 2, the administrator-side processing is assumed to be executed at the time of approving the employee's attendance data for the previous day.

[0060] The process in step S51 is the same as the process in step S31 in Figure 9. Also, the processes from steps S54 to S56 are the same as the processes from steps S32 to S34 in Figure 9.

[0061] (Step S52: Input Processing) The input unit 21 inputs pairs of answers and model questions or designated questions 34 into the trained advice generation model 43. In this process, the input unit 21 inputs instructions to the advice generation model 43 to generate advice for the administrator to use when dealing with the target person. The input unit 21 also inputs the output format of the advice to the advice generation model 43. Here, the output format is, for example, a character limit.

[0062] Referring to Figure 13, a specific example of the input to the advice generation model 43 according to Embodiment 3 will be explained. In Figure 13, the #Instruction document contains the instructions for generating advice. The #Question contains the model question or specified question 34. The #Answer contains the answer from the subject. The #Output format contains the output format for the advice. Alternatively, attendance data 31 for the subject from the past reference period, including the previous day, may also be entered, and advice may be generated considering the attendance data 31.

[0063] When the answer and question pair is entered in step S52, the advice generation model 43 generates advice according to the instructions.

[0064] (Step S53: Output processing) The output unit 22 retrieves the advice generated by the advice generation model 43 in response to the answer and question pair entered in step S52. The output unit 22 outputs the advice to the administrator terminal 52 used by the administrator. For example, advice like that shown in Figure 14 is output. As a result, the advice is displayed on the display device of the administrator terminal 52, allowing the administrator to review it. The administrator can then refer to the advice and decide whether or not to enter the specified question 34.

[0065] ***Effects of Embodiment 3*** As described above, the attendance management device 10 according to Embodiment 3 outputs advice for the manager to use when responding to the person in question, based on the answer and question pair. By referring to the advice, the manager can understand how to respond to the person in question, and communication between the manager and the person in question becomes smoother. As a result, the person in question feels that the manager is trying to understand their work situation, and engagement is more likely to improve. In addition, it becomes possible to decide whether or not to input the designated question 34 after referring to the advice, and appropriate input of the designated question 34 becomes possible.

[0066] Embodiment 4. Embodiment 4 differs from Embodiments 1 to 3 in that it calculates an engagement evaluation value for the target person from the question and answer pair. Embodiment 4 explains this difference, while omitting explanations of the same points.

[0067] ***Explanation of the structure*** Referring to Figure 15, the configuration of the attendance management device 10 according to Embodiment 4 will be described. Figure 15 shows an example of a modified version of the attendance management device 10 shown in Figure 1. The attendance management device 10 differs from the attendance management device 10 shown in Figure 1, etc., in that evaluation data 35 is stored in storage 13. Furthermore, the attendance management device 10 differs from the attendance management device 10 shown in Figure 1, etc., in that it is connected to the evaluation model 44, which is a trained model, via a communication interface 14.

[0068] ***Explanation of operation*** Referring to Figures 16 to 18, the operation of the attendance management device 10 according to Embodiment 4 will be explained. The operation of the attendance management device 10 according to Embodiment 4 includes an evaluation process.

[0069] The evaluation process according to Embodiment 4 will be described with reference to Figure 16. The evaluation process is executed at times specified by the administrator or other designated personnel. For example, the evaluation process is executed when the administrator wants to check the engagement level of the target individual.

[0070] (Step S61: Input Processing) The input unit 21 inputs pairs of answers for each day of the evaluation period and questions that are model questions or designated questions 34 into the trained evaluation model 44. The input unit 21 may also input attendance data for each day of the evaluation period into the evaluation model 44. The evaluation period may be, for example, a month or a year, or the period during which a certain project was carried out. In this process, the input unit 21 inputs an instruction to the evaluation model 44 to calculate a numerical value representing the evaluation of the subject's engagement with the organization to which the subject belongs. The input unit 21 also inputs a target type, which is one or more evaluation types from among several evaluation types related to engagement with the organization to which the subject belongs, into the evaluation model 44.

[0071] Referring to Figure 17, a specific example of the input to the evaluation model 44 according to Embodiment 4 will be explained. In Figure 17, the #Instruction sheet contains instructions for calculating evaluation values ​​for each target type. The #Attendance data contains attendance data for each day of the evaluation period. For each day of the evaluation period, a #Question and #Answer pair is set up, with the #Question containing a model question or designated question 34, and the #Answer containing the response from the target person. The #Evaluation type contains the target type.

[0072] Once the answer and question pairs are entered in step S61, the evaluation model 44 calculates evaluation values ​​for each target category according to the instructions.

[0073] (Step S62: Output processing) The output unit 22 obtains evaluation values ​​for each target type calculated by the evaluation model 44 in correspondence with the answer and question pairs entered in step S61. The output unit 22 outputs the evaluation values ​​to the administrator terminal 52 used by the administrator. As a result, the evaluation values ​​are displayed on the display device of the administrator terminal 52, allowing the administrator to check the evaluation values. For example, as shown in Figure 18, the output unit 22 may display the evaluation values ​​for each target type as a radar chart or the like.

[0074] (Step S63: Memory processing) The output unit 22 writes the evaluation values ​​for each target type obtained in step S62 to the storage 13 as evaluation data 35.

[0075] Here, steps S61 and S62 are executed sequentially for each target type. Once processing for all target types is complete, the evaluation value for each target type is displayed as a radar chart or similar.

[0076] ***Effects of Embodiment 4*** As described above, the attendance management device 10 according to Embodiment 4 calculates an engagement evaluation value for the subject from the question and answer pair. This makes it easy for the administrator to check the engagement status of the subject.

[0077] ***Other configurations*** <Variation 3> In Embodiment 4, the evaluation value is calculated for each target type. Alternatively, the evaluation value may be calculated for each set of answer and question, and the evaluation value for each target type may be calculated from the evaluation value for each set. In this case, in step S61, the input unit 21 should input an instruction to the evaluation model 44 to calculate the evaluation value for each set of answer and question, and then calculate the evaluation value for each target type from the evaluation value for each set. As a result, as shown in Figure 19, evaluation values ​​are obtained not only for each target type, but also for each pair of answers and questions. Administrators can refer to the evaluation values ​​for each pair of answers and questions to check the engagement status in more detail. In this embodiment, the term "attendance management device" may include other devices that use attendance management data and questions and answers from the subject. For example, it may be a data analysis device that receives attendance management data and questions and answers from the subject from the attendance management device and evaluates engagement.

[0078] Embodiment 5. Embodiment 5 differs from Embodiment 4 in that it generates advice based on evaluation values. Embodiment 5 explains this difference, while omitting explanations of the same points.

[0079] ***Explanation of the structure*** Referring to Figure 20, the configuration of the attendance management device 10 according to Embodiment 5 will be described. The attendance management device 10 differs from the attendance management device 10 shown in Figure 15 in that it includes an adjustment reception unit 25 as a functional component. Furthermore, the attendance management device 10 differs from the attendance management device 10 shown in Figure 15 in that it is connected to the advice generation model 43 via a communication interface 14.

[0080] ***Explanation of operation*** Referring to Figures 21 to 24, the operation of the attendance management device 10 according to Embodiment 5 will be explained. The operation of the attendance management device 10 according to Embodiment 5 includes an evaluation advice generation process.

[0081] Referring to Figure 21, the evaluation advice generation process according to Embodiment 5 will be described. The evaluation advice generation process is executed after the evaluation process described in Embodiment 4. For example, the evaluation advice generation process is executed at a timing specified by the subject after the evaluation process has been completed.

[0082] (Step S71: Adjustment process) The adjustment reception unit 25 outputs the evaluation values ​​for each target category obtained from the evaluation process to the user terminal 51 used by the target. At this time, the adjustment reception unit 25 also outputs the average value of the evaluation values ​​for each target category to the user terminal 51, along with the evaluation values ​​for each target category for the target. As a result, as shown in Figure 22, the evaluation value and the average value are displayed on the display device of the user terminal 51 for each target category. The average value is the average value for people within the range being compared with the target. For example, the average value is the average value in the department to which the target belongs. The adjustment reception unit 25 receives input from the subject for each subject type, which is the adjusted evaluation value. If the subject's evaluation value differs from their self-assessment, they adjust the evaluation value by referring to the average value, etc.

[0083] (Step S72: Evaluation adjustment value storage processing) The adjustment reception unit 25 adds the evaluation adjustment value received in step S71 to the evaluation data 35 for the subject.

[0084] (Step S73: Input Processing) The input unit 21 inputs the adjusted evaluation values ​​of the subject for each subject type into the trained advice generation model 43. The input unit 21 also inputs the evaluation values ​​of others into the advice generation model 43. Here, "others" refers to the people whose evaluation values ​​were used when calculating the average value mentioned above. Note that "others" may not refer to all the people whose evaluation values ​​were used when calculating the average value, but rather to a select group of people close to the subject. For example, if the average value is the average value within the subject's department, "others" may refer to members of the project team to which the subject belongs within that department. In this case, the input unit 21 inputs instructions to the advice generation model 43 to generate advice for subjects whose evaluation value will be increased based on the subject type of the subject.

[0085] Referring to Figure 23, a specific example of the input to the advice generation model 43 according to Embodiment 5 will be explained. Figure 23 shows that the #order document contains instructions for generating advice for individuals whose evaluation scores need to be improved. Figure 23 also shows the target categories as: overtime tolerance, proactiveness in taking leave, work-life balance, skill development, and positivity. #Evaluation scores are set for both the target individual, Person A, and others, with #evaluation scores specific to each target category. Additionally, an evaluation adjustment value is set for the target individual's evaluation score.

[0086] (Step S74: Output processing) The output unit 22 obtains advice for subjects whose target type is being improved, based on the evaluation values ​​for the target type entered in step S73, and which is generated by the advice generation model 43. The output unit 22 outputs the advice to the user terminal 51 used by the subject. This makes the advice available to the subject. As shown in Figure 24, advice is generated for each target type. The output unit 22 may also output the advice to the administrator terminal 52 used by the administrator.

[0087] ***Effects of Embodiment 5*** The attendance management device 10 according to Embodiment 5 generates advice to improve evaluation scores. This makes it possible to understand what the person needs to do to improve their condition.

[0088] Embodiment 6. Embodiment 6 differs from Embodiment 2 in that it inputs current events information when generating a template. Embodiment 6 explains this difference, while omitting explanations of the same points.

[0089] ***Explanation of operation*** The operation of the attendance management device 10 according to Embodiment 6 will be described with reference to Figures 6, 25, and 26. The template generation process differs from that of Embodiment 2. The administrator-side processing and the user-side processing are the same as in Embodiment 2.

[0090] Referring to Figure 6, the template generation process according to Embodiment 6 will be described. The template generation process according to Embodiment 6 is executed when updating the template question 33 that was set during the initial setup. Normally, the template generation process according to Embodiment 2 is executed, and the template generation process according to Embodiment 6 may be executed periodically. For example, the template generation process according to Embodiment 6 may be executed around the middle of the month or at the change of seasons.

[0091] The processing in steps S22 and S23 is the same as in Embodiment 2.

[0092] (Step S21: Input Processing) The input unit 21 receives current events information in addition to the types of multiple questions. Current events information includes news articles and information about events that occurred around the time the template generation process was executed. Current events information may be manually set from news such as organizational events, or the input unit 21 may retrieve news stored on the company's network using web cloning or an API and set it. API stands for Application Programming Interface.

[0093] Referring to Figure 25, a specific example of input to the template generation model 42 according to Embodiment 6 will be described. Figure 25 differs from Embodiment 2 in that it includes current events information under #Current Events. Figure 25 includes two pieces of current events information: "Please ask a positive question regarding the possibility that the price of gasoline may rise to around 185 yen per liter by January next year" (Current Events Information 1) and "A bowling tournament will be held as a company event on December 5th. Please ask a question about your enthusiasm for the bowling tournament" (Current Events Information 2).

[0094] The template generation model 42 then generates template questions 33 that take current events into consideration. For example, as shown in Figure 26, when the current events information shown in Figure 25 is input, a question that takes current events information 1 into consideration and a question that takes current events information 2 into consideration are generated. In this case, the input unit 21 may instruct the template generation model 42 to acquire current events information and then generate questions and answers based on that information.

[0095] ***Effects of Embodiment 6*** As described above, the attendance management device 10 according to Embodiment 6 inputs current information when generating templates. This prevents the template questions 33 from always being the same and becoming outdated.

[0096] <Modification 4> In Embodiment 1, each functional component was implemented in software. However, in Modification 4, each functional component may be implemented in hardware. The differences between this Modification 4 and Embodiment 1 will be explained below.

[0097] When each functional component is implemented in hardware, the attendance management device 10 includes an electronic circuit 15 instead of a processor 11, memory 12, and storage 13. The electronic circuit 15 is a dedicated circuit that implements the functions of each functional component, as well as the functions of the memory 12 and storage 13.

[0098] Electronic circuits 15 can include single circuits, complex circuits, programmed processors, parallel programmed processors, logic ICs, GAs, ASICs, and FPGAs. GA stands for Gate Array. ASIC stands for Application Specific Integrated Circuit. FPGA stands for Field-Programmable Gate Array. Each functional component may be implemented in a single electronic circuit 15, or each functional component may be implemented by distributing them across multiple electronic circuits 15.

[0099] <Modification 5> As a fifth variation, some of the functional components may be implemented in hardware, while others may be implemented in software.

[0100] The processor 11, memory 12, storage 13, and electronic circuit 15 are collectively referred to as the processing circuit. In other words, the function of each functional component is realized by the processing circuit.

[0101] Furthermore, the term "part" in the above explanation may be replaced with "circuit," "process," "procedure," "processing," or "processing circuit."

[0102] The various aspects of this disclosure are summarized below as an appendix. (Note 1) An input unit that inputs attendance data showing the work status of the target person into a pre-trained question generation model, An output unit outputs a model question, which is a question to confirm the status of the subject, generated by the question generation model in response to the attendance data input by the input unit. A time and attendance management device equipped with the following features. (Note 2) The aforementioned attendance management device further, Question reception unit that receives input of designated questions from the manager of the aforementioned subject, which are questions to confirm the status of the aforementioned subject. Equipped with, The output unit outputs the specified question received by the question receiving unit. The attendance management device described in Appendix 1. (Note 3) The input unit inputs multiple question types into a template generation model, which is a trained model. The output unit outputs template questions, which are questions that confirm the status of the subject for each of the multiple question types generated by the template generation model in accordance with the types of the multiple questions. The question receiving unit receives input of designated questions generated by the administrator based on the template questions. The attendance management device described in Appendix 2. (Note 4) The aforementioned attendance management device further, Answer receiving unit that receives input of the answer to the model question output by the output unit. Equipped with, The input unit inputs the pair of the answer received by the answer receiving unit and the model question into the advice generation model, which is a trained model. The output unit outputs advice generated by the advice generation model for when the administrator interacts with the subject, corresponding to the pair of the answer and the model question. The attendance management device described in Appendix 1. (Note 5) The aforementioned attendance management device further, Answer receiving unit that receives input of the answer to the model question output by the output unit. Equipped with, The input unit inputs the response received by the response receiving unit and the model question into the evaluation model, which is a trained model. The output unit outputs an evaluation value of the subject's engagement with the organization to which the subject belongs, calculated by the evaluation model in correspondence with the pair of the answer and the model question. The attendance management device described in Appendix 1. (Note 6) The input unit inputs instructions to the evaluation model to calculate an evaluation value for a target type which is at least one of several evaluation types related to engagement with the organization. The output unit outputs evaluation values ​​for the target type calculated by the evaluation model. The attendance management device described in Appendix 5. (Note 7) The input unit inputs the evaluation value for the target type into the advice generation model, which is a trained model. The output unit outputs advice for the subject that increases the evaluation value for the subject type, which is generated by the advice generation model in accordance with the evaluation value for the subject type. The attendance management device described in Appendix 6. (Note 8) The input unit inputs information on current events, in addition to the types of the multiple questions, into the template generation model. The output unit outputs template questions, which are questions for each of the multiple question types generated by the template generation model in correspondence with the multiple question types and the current events information. The attendance management device described in Appendix 3. (Note 9) The computer inputs attendance data showing the work status of the subject into a pre-trained question generation model. A time and attendance management method in which a computer outputs model questions, which are questions that confirm the status of the subject, generated by the question generation model in response to the time and attendance data. (Note 10) The input process involves inputting attendance data, which shows the work status of the target individuals, into a pre-trained question generation model. An output process outputs a model question, which is a question to confirm the status of the subject, generated by the question generation model in response to the attendance data input by the input process. An attendance management program that enables a computer to function as an attendance management device.

[0103] The embodiments and variations of this disclosure have been described above. Some of these embodiments and variations may be implemented in combination. Alternatively, some or all of them may be implemented in part. However, this disclosure is not limited to the embodiments and variations described above, and various modifications are possible as needed. [Explanation of Symbols]

[0104] 10 Attendance management device, 11 Processor, 12 Memory, 13 Storage, 14 Communication interface, 21 Input unit, 22 Output unit, 23 Answer reception unit, 24 Question reception unit, 25 Adjustment reception unit, 31 Attendance data, 32 Question answer data, 33 Template questions, 34 Specified questions, 35 Evaluation data, 41 Question generation model, 42 Template generation model, 43 Advice generation model, 44 Evaluation model, 51 User terminal, 52 Administrator terminal.

Claims

1. An input unit that inputs attendance data showing the work status of the target person into a pre-trained question generation model, An output unit outputs a model question, which is a question to confirm the status of the subject, generated by the question generation model in response to the attendance data input by the input unit. A time and attendance management device equipped with the following features.

2. The aforementioned attendance management device further, Question reception unit that receives input of designated questions from the manager of the aforementioned subject, which are questions to confirm the status of the aforementioned subject. Equipped with, The output unit outputs the specified question received by the question receiving unit. The attendance management device according to claim 1.

3. The input unit inputs multiple question types into a template generation model, which is a trained model. The output unit outputs template questions, which are questions that confirm the status of the subject for each of the multiple question types generated by the template generation model in accordance with the types of the multiple questions. The question receiving unit receives input of designated questions generated by the administrator based on the template questions. The attendance management device according to claim 2.

4. The aforementioned attendance management device further, Answer receiving unit that receives input of the answer to the model question output by the output unit. Equipped with, The input unit inputs the pair of the answer received by the answer receiving unit and the model question into the advice generation model, which is a trained model. The output unit outputs advice generated by the advice generation model for when the administrator interacts with the subject, corresponding to the pair of the answer and the model question. The attendance management device according to claim 1.

5. The aforementioned attendance management device further, Answer receiving unit that receives input of the answer to the model question output by the output unit. Equipped with, The input unit inputs the response received by the response receiving unit and the model question into the evaluation model, which is a trained model. The output unit outputs an evaluation value of the subject's engagement with the organization to which the subject belongs, calculated by the evaluation model in correspondence with the pair of the answer and the model question. The attendance management device according to claim 1.

6. The input unit inputs instructions to the evaluation model to calculate an evaluation value for a target type which is at least one of several evaluation types related to engagement with the organization. The output unit outputs evaluation values ​​for the target type calculated by the evaluation model. The attendance management device according to claim 5.

7. The input unit inputs the evaluation value for the target type into the advice generation model, which is a trained model. The output unit outputs advice for the subject that increases the evaluation value for the subject type, which is generated by the advice generation model in accordance with the evaluation value for the subject type. The attendance management device according to claim 6.

8. The input unit inputs information on current events, in addition to the types of the multiple questions, into the template generation model. The output unit outputs template questions, which are questions for each of the multiple question types generated by the template generation model in correspondence with the multiple question types and the current events information. The attendance management device according to claim 3.

9. The computer inputs attendance data showing the work status of the subject into a pre-trained question generation model. A time and attendance management method in which a computer outputs model questions, which are questions that confirm the status of the subject, generated by the question generation model in response to the time and attendance data.

10. The input process involves inputting attendance data, which shows the work status of the target individuals, into a pre-trained question generation model. An output process outputs a model question, which is a question to confirm the status of the subject, generated by the question generation model in response to the attendance data input by the input process. An attendance management program that enables a computer to function as an attendance management device.

Images