Information processing device, information processing method, and program

Abstract

To provide a technology that can easily record the actions of people involved in an organization. [Solution] An information processing device 10 according to one embodiment has a receiving means 12 for receiving input of a log of the behavior of an observed person, who is either a related party in an organization, a robot, or an AI, including the specification of one of a plurality of defined types; a writing means 13 for writing the log to a log database; a determining means 51 for determining whether a target log among the logs recorded in the log database satisfies the conditions for writing up a document; and a creating means 17 for creating a document in a predetermined format from the contents of the target log among the logs recorded in the log database if it is determined that the target log satisfies the conditions for writing up a document.

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Traditionally, recording problematic behaviors of relevant parties in an organization and creating documents based on those records has been done manually. For example, Patent Document 1 discloses an information processing device that efficiently extracts desired information from non-standard documents. Patent Document 2 discloses an inheritance business support system for financial institutions. Patent Document 3 discloses a labor-related document creation system that displays aggregated information on other companies. Patent Document 4 discloses a trade support processing system that acquires service plans based on order information and executes output processing for the service plans. Patent Document 5 discloses a system that records nuisance behaviors such as harassment entered by employees in a database, analyzes the reports using a data analysis system, and provides the analysis results as a dashboard. Patent Document 6 discloses a system that uses a machine learning model to calculate the probability of winning or losing a lawsuit regarding harassment based on user responses regarding the extent of the victimization, such as harassment. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-161196 [Patent Document 2] Patent No. 6571597 [Patent Document 3] Patent No. 6671601 [Patent Document 4] Japanese Patent Application Laid-Open No. 2009-294796 [Patent Document 5] US Patent Application Publication No. 2019 / 0180245 [Patent Document 6] Japanese Patent Publication No. 2023-004333 Summary of the Invention [Problem to be solved by the invention]

[0004] None of the techniques described in Patent Documents 1 to 6 are intended to simply record the actions of people involved in an organization and create documents in a prescribed format.

[0005] In response to this, the present invention provides a technique that makes it easy to record the actions of people involved in an organization. [Means for solving the problem]

[0006] One aspect of the present disclosure provides an information processing device having a receiving means for receiving input of a log of the actions of related parties in an organization, the log including the designation of one of a plurality of defined types, a writing means for writing the log to a log database, and a creation means for creating a document in a predetermined format from the contents of the log recorded in the log database.

[0007] This information processing device may have a judgment means for judging whether a log registered in the log database satisfies the grouping conditions, a registration means for registering the group related to the log in the group database if it is judged that the log satisfies the grouping conditions, and an output means for outputting information related to the group to the terminal of a registered advisor when a request is received from the terminal.

[0008] The creation means may create the document from the contents of a log related to the group when it is determined that a case registered in the group database satisfies the conditions for document creation.

[0009] The system may further comprise an output means for, upon receiving a request from a terminal of a registered observer, outputting information relating to the logs recorded in the log database to the terminal.

[0010] The information processing device may further include a suggestion unit that suggests an input template for the log that includes items corresponding to the type of the action.

[0011] The item may include a justification for the party.

[0012] The items may include the organization's actions.

[0013] The item may include a note from the organization to the interested party.

[0014] The system may include an access means for accessing the organization's work rules, an extraction means for extracting clauses from the work rules that may conflict with the behavior, and a presentation means for presenting the extracted clauses.

[0015] The extraction means may extract, from the log and the work regulations, persons involved who are likely to potentially cause problems.

[0016] The accepting means may accept the input of the log by voice of the recorder.

[0017] The system may include a text conversion means for converting the speech into text, and an annotation means for adding annotations to the text based on the speech.

[0018] The organization may have a plurality of groups, and may include a counting unit that counts problematic behaviors in the log for each of the groups, and a presentation unit that presents information indicating the count results of the problematic behaviors.

[0019] The system may further include a measuring unit that measures the time elapsed since the problematic behavior in the log, and a presentation unit that presents information indicating the measured time elapsed.

[0020] The system may include a receiving means for receiving input of a log of the actions of related parties in an organization, the log including the designation of one of a plurality of defined types; a writing means for writing the log to a database; a registering means for registering a group related to the log in the database if the log recorded in the database satisfies a grouping condition; and an output means for outputting information about the group registered in the database to the terminal of a registered advisor upon request from the advisor.

[0021] Another aspect of the present disclosure provides an information processing device having a receiving means for receiving input of a log of the actions of related parties in an organization, the log including the designation of one of a plurality of defined types; a writing means for writing the log to a database; a registering means for registering a case related to the log in the database if the log recorded in the database satisfies the conditions for making it a case; and an output means for outputting information related to the case registered in the database to the terminal of a registered advisor upon request from the advisor.

[0022] Yet another aspect of the present disclosure provides an information processing method including the steps of accepting input of a log of the actions of relevant parties in an organization, the log including the designation of one of a plurality of defined types, writing the log to a log database, and creating a document in a predetermined format from the contents of the log recorded in the log database.

[0023] Yet another aspect of the present disclosure provides a program for causing a computer to execute the steps of accepting input of a log of the actions of relevant parties in an organization, the log including the designation of one of a plurality of defined types, writing the log to a log database, and creating a document in a predetermined format from the contents of the log recorded in the log database.

[0024] Yet another aspect of the present disclosure provides an information processing device having: a receiving means for receiving input of a log of the behavior of an observed person, who is either a related party in an organization, a robot, or an AI, the log including the designation of one of a plurality of defined types; a writing means for writing the log to a log database; a first determination means for determining whether a target log among the logs recorded in the log database satisfies the conditions for writing in writing; and a creation means for creating a document in a predetermined format from the contents of the target log among the logs recorded in the log database if it is determined that the target log satisfies the conditions for writing in writing.

[0025] The information processing device may further comprise a second determination means for determining whether the logs registered in the log database satisfy a grouping condition, and the target logs may be logs that satisfy the grouping condition.

[0026] This information processing device may have an extraction means that uses an AI separate from the observed person to extract clauses from the rules concerning the organization that may be in conflict with the behavior, and the creation means may create the document including the extracted clauses.

[0027] The AI ​​may include a machine learning model trained using training data including the rule or a rule similar to the rule, the behavior of the observed person, and identification information of clauses of the rule that the behavior may conflict with.

[0028] The information processing device may include an estimation unit that estimates the cost of resolving the problem related to the target log.

[0029] The estimation means may estimate the cost using an AI separate from the object log and the observed person.

[0030] The AI ​​may include a machine learning model trained using training data including information indicating the behavior of the observed person and the cost of solving problems related to that behavior.

[0031] The creating means may create the document including information indicating the cost.

[0032] The information processing device may include a presentation unit that presents information indicating the cost.

[0033] The costs may include at least one of labor costs of the parties involved in resolving the problem, lost productivity due to the problem, and fees paid to outside experts.

[0034] The costs may include future additional losses incurred in solving the problem.

[0035] The creating means may create the document including an action plan for resolving the problem related to the target log.

[0036] The creating means may create the document including an analysis of the psychological tendencies of the parties involved.

[0037] The creating means may create the document including a development plan for the person being observed.

[0038] The creating means may create the document including an evidence package to be used in legal proceedings regarding problematic behavior related to the target log.

[0039] Yet another aspect of the present disclosure provides an information processing method including the steps of: accepting input of a log of the behavior of an observed person, who is either a related party in an organization, a robot, or an AI, the log including a designation of one of a plurality of defined types; writing the log to a log database; determining whether a target log from among the logs recorded in the log database satisfies the conditions for writing up; and, if it is determined that the target log satisfies the conditions for writing up, creating a document in a predetermined format from the contents of the target log from among the logs recorded in the log database.

[0040] Yet another aspect of the present disclosure provides a program for causing a computer to execute the steps of accepting input of a log of the behavior of an observed person, who is either a related person in an organization, a robot, or an AI, the log including the designation of one of a plurality of defined types; writing the log to a log database; determining whether a target log from among the logs recorded in the log database satisfies the conditions for writing up in writing; and, if it is determined that the target log satisfies the conditions for writing up in writing, creating a document in a predetermined format from the contents of the target log from among the logs recorded in the log database. [Effects of the Invention]

[0041] According to the present invention, problematic behaviors of people involved in an organization can be recorded and a written document in a predetermined format can be prepared. [Brief explanation of the drawings]

[0042] [Figure 1] 1 is a diagram showing an overview of an information processing system 1 according to an embodiment. [Figure 2] FIG. 1 is a block diagram showing the functional configuration of an information processing system 1. [Figure 3] FIG. 1 is a diagram illustrating an example of a hardware configuration of an information processing device 10. [Figure 4] FIG. 2 is a diagram illustrating an example of the hardware configuration of a user terminal 20. [Figure 5] FIG. 3 is a diagram illustrating an example of the recorded contents of a user database 113. [Figure 6] 3 is a sequence chart showing an outline of the operation of the information processing system 1. [Figure 7A] FIG. 10 is a diagram illustrating an example of an introduction screen. [Figure 7B] FIG. 10 is a diagram illustrating an example of a log input screen. [Figure 8] FIG. 2 is a diagram illustrating an example of the recorded contents of a log database 111. [Figure 9] 10 is a sequence chart showing an outline of a process related to viewing an action log. [Figure 10] FIG. 10 is a diagram illustrating a list display screen. [Figure 11] FIG. 10 is a diagram illustrating a detailed display screen. [Figure 12] 10 is a flowchart showing an outline of a process related to the generation of an instance. [Figure 13] FIG. 10 is a diagram illustrating an example of a case creation screen. [Figure 14] FIG. 2 is a diagram illustrating an example of a case database 112. [Figure 15] FIG. 2 is a block diagram showing the functional configuration of the information processing system 2. DETAILED DESCRIPTION OF THE INVENTION

[0043] 1. Configuration FIG. 1 is a diagram showing an overview of an information processing system 1 according to a first embodiment of the present invention. The information processing system 1 is a system for recording the actions of people involved in an organization as a log (hereinafter referred to as an "action log") and providing a service for creating documents based on the action log. Hereinafter, this service will be referred to as a "recording service." ... This system assists in recording the actions of those involved and allows easy reference to those records when trouble occurs, thereby contributing to problem solving. Information processing system 1 particularly considers problems in relation to the rules that must be observed within the organization. Behavior that violates these rules is considered problematic behavior and has a negative impact on the organization's environment. The ultimate goal of information processing system 1 is to reduce such problematic behavior and improve the organization's environment.

[0044] In one example, the rules targeted by the information processing system 1 include at least one of the following: rules established by the organization, such as work regulations; policies for dealing with specific external parties (e.g., policies for dealing with unreasonable requests from customers or guardians); terms of use or privacy policies for web services; laws, such as the Power Harassment Prevention Act or the Civil Code; or unwritten rules, such as common sense. Problematic behavior refers to behavior that may violate these rules. Examples of problematic behavior include being late, absent from work without permission, leaving the workplace without permission, power harassment, and sexual harassment. In another example, the rules targeted by the information processing system 1 include the work content, roles, scope of responsibilities, required skills, or experience for each job (position) defined in a job description. In this way, including job descriptions in the rules targeted by the information processing system 1 can address the difficulties of job performance evaluation in job-based employment, making the information processing system 1 a tool that can be used consistently from recruitment to retirement.

[0045] The parties involved in the information processing system 1 in a broad sense, i.e., users, are divided into the observed person Ut, the observer Us, and the advisor Ua. The observed person Ut is a person who has performed an action recorded in the action log, i.e., the subject of the action log. The observed person Ut is not limited to a human being, and may be, for example, a robot. It may be a device that operates (or appears to operate) autonomously, such as a bot or an AI agent (or simply an AI). A device that operates (or appears to operate) autonomously refers to a device that operates adaptively according to the situation without human instruction input, and some of its operations may be set by a program or the like. The observer Us is a person who records the behavior log of the observed, that is, the behavior log. The advisor Ua refers to the person who records or inputs the action log and other information. In one example, the organization is a company, the people involved are the employees and officers of that company, and the behavior being recorded is the problematic behavior of the employees. In this case, the observed person Ut The observer Us is the employee's supervisor or colleague, or the person in charge of providing the service. The adviser Ua is a lawyer, a social insurance labor consultant, or a are professionals such as industrial counselors.

[0046] The information processing system 1 includes an information processing device 10 and a user terminal 20. The information processing device 10 is a server in a recording service. The user terminal 20 is a client in the recording service. The information processing device 10 and the user terminal 20 are communicatively connected via a network 30. The network 30 is, for example, a computer network such as the Internet, a LAN, or a VPN.

[0047] 2 is a block diagram showing the functional configuration of the information processing system 1. The information processing system 1 includes a storage unit 11, a receiving unit 12, a writing unit 13, a determining unit 14, a registering unit 15, an output unit 16, a creating unit 17, and a control unit 19. In this example, these elements are implemented in an information processing device 10.

[0048] The storage means 11 stores various data and programs. In this example, the storage means 11 stores a log database 111, a case database 112, a user database 113, and a rule database 114 (databases are written as DB in the figure). The log database 111 is a database that records action logs. The case database 112 is a database that records cases. The user database 113 is a database that records user attributes. The rules database 114 is a database that records rules that an organization must comply with. For example, a job description is stored in the rules database 114 in a data structure that includes items such as a job ID that uniquely identifies the job, multiple task items included in the job, skills required for each task item, and key performance indicators (KPIs) that measure the degree of task achievement. Details of these databases will be described later.

[0049] The receiving means 12 receives input of a behavior log sent from the user terminal 20. The behavior log includes behavior logs of related parties. In this example, the behavior is particularly problematic behavior. In one example, the behavior log includes the type of problematic behavior, the date and time of the problematic behavior, the content of the problematic behavior, the names of related parties, etc. The writing means 13 writes the behavior log received by the receiving means 12 into the log database 111.

[0050] The determination means 14 determines whether the action log satisfies the conditions for event creation. When the behavior of a related party exceeds a certain level, the organization may need to take some kind of action against that related party. In this paper, this state in which the related party's behavior exceeds a certain level is referred to as "event creation," and the event related to that action is referred to as "event." In one example, event creation refers to grouping one or more action logs and managing the grouped action logs separately from other action logs. Although "grouping" is used here, examples of grouping also include assigning labels, tags, or flags for separate management to one or more action logs. In other words, whether there is one or multiple action logs, both can be grouped. Separate management refers to managing action logs separately from normal action logs, such as granting access rights to users different from those who use action logs, recording training history, or recording personnel and labor responses. Note that action logs and information other than action logs may be managed together in a single group. For example, the information processing system 1 may record action logs and training history separately and manage them as groups. The registration means 15 registers an event that is determined by the determination means 14 to satisfy the conditions for becoming a case as a case in the case database 112. The case database 112 is an example of a group database that records information about grouped action logs. In this example, a separate case database 112 is prepared for each case, and case information is recorded therein.

[0051] The output means 16 outputs (or provides) information recorded in the log database 111 or the case database 112 to the user terminal 20 in response to access from a related person using the user terminal 20. It can be said that the output means 16 controls access to the log database 111 or the case database 112.

[0052] The creation means 17 creates a document in a predetermined format from the contents of the behavior log recorded in the log database 111. This document may be, for example, an observation history, a training history, a report, a notice, or a warning. When the creation means 17 creates a document, the output means 16 may output the created document to another device such as a user terminal 20. The control means 19 executes various processes.

[0053] 3 is a diagram illustrating an example of a hardware configuration of the information processing device 10. The information processing device 10 includes a CPU (Central Processing Unit) 101, a memory 102, a storage device 103, and The computer device is equipped with a communication device 104. These components are connected to each other via a bus 107.

[0054] The CPU 101 is an arithmetic processing unit that controls the overall operation of the information processing device 10 by reading and executing programs stored in the storage device 103. The memory 102 is, for example, a volatile memory such as a RAM (Random Access Memory) and temporarily stores programs and data executed by the CPU 101. The storage device 103 is, for example, a hard disk The communication device 104 is a non-volatile memory such as a hard disk drive (HDD) or a solid state drive (SSD), and permanently stores programs, an operating system (OS), and various data. The communication device 104 is, for example, a network interface card (NIC), and communicates with other devices via the network 30.

[0055] In this example, the programs stored in storage device 103 include a program (hereinafter referred to as a "server program") for causing a computer to function as a server in information processing system 1. When CPU 101 is executing the server program, CPU 101 is an example of receiving means 12, writing means 13, determining means 14, registering means 15, outputting means 16, creating means 17, and control means 19, and at least one of memory 102 and storage device 103 is an example of storage means 11.

[0056] 4 is a diagram illustrating an example of the hardware configuration of the user terminal 20. The user terminal 20 can be configured using a general-purpose computer. The user terminal 20 is a computer, such as a personal computer, smartphone, or tablet terminal, that includes a CPU 201, a memory 202, a storage device 203, a communication device 204, an input device 205, and an output device 206. These components are connected to each other via a bus 207.

[0057] The CPU 201 is an arithmetic processing unit that controls the overall operation of the user terminal 20 by reading and executing programs stored in the storage device 203. The memory 202 is, for example, a volatile memory such as RAM, and temporarily stores programs and data executed by the CPU 201. The storage device 203 is, for example, a non-volatile memory such as an HDD or SSD, and permanently stores programs, an OS, and various data. The communication device 204 is, for example, an NIC or a wireless LAN module, and communicates with the information processing device 10 via the network 30. The input device 205 is a device that accepts instructions or information input from a user, and includes, for example, at least one of a keyboard, a mouse, a touch sensor, and a microphone. The output device 206 is a device that outputs information to a user, and includes, for example, at least one of a display, a speaker, and a printer.

[0058] In this example, the programs stored in the storage device 203 include a program for causing a computer to function as a client in the information processing system 1 (hereinafter referred to as a "client program").

[0059] 2.Operation The following describes the operation of the information processing system 1. The operation of the information processing system 1 can be roughly divided into recording an action log, viewing an action log, and managing a case. In this example, the organization is a company, and the people involved include employees of the company as the observed person Ut, and the problematic behavior of the employees is All employees and executives are assigned at least one role of observer Us and observed Ut. In cases where the company has a hierarchical structure, one employee may be assigned both the roles of observer Us and observed Ut. For example, a section manager is an observer Us for the section staff in charge, i.e., his subordinates, and is an observer Us for the employees to whom that section belongs. The person in charge is the person being observed Ut in relation to the employee who is the manager of the department, i.e., the superior. The user is registered in advance in the system 1, and the attributes are recorded in the user database 113. The correspondence between the observer Us and the observed Ut is also defined in the user database 113. The advisor Ua is also registered as a user in the user database 113. It will be recorded.

[0060] FIG. 5 is a diagram showing an example of the recorded contents of the user database 113. The user database 113 has a plurality of records. Each record corresponds to the attributes of one user, i.e., a related person. Only one record is shown in FIG. 5. Each record contains a plurality of attribute items. In this example, the information includes user ID, role, name, affiliation, job title, superior, date of birth, and gender. The user ID is identification information that uniquely identifies the person involved. The role indicates the role of the person involved in the recording service. In this example, there are three roles: observer, observed, and advisor. One person may hold multiple roles. Name, job title, superior, date of birth, and gender are examples of general user attributes. This information is provided by the organization when using the information processing system 1.

[0061] Each related person accesses the information processing device 10 from his / her own user terminal 20. Access rights are defined for each related person according to his / her role. The access rights are recorded, for example, in the log database 111 or the case database 112. Each related person can access information recorded in the information processing system 1 within the scope according to his / her role. There are multiple categories of access rights, for example, "view only" and "editable".

[0062] 2-1. Recording of behavior logs 6 is a sequence chart showing an outline of the operation related to recording of an action log in the information processing system 1. In the following, hardware such as the information processing device 10 will be described as the subject of processing. This means that a hardware element such as the CPU 101 executes processing in cooperation with other hardware elements such as the memory 102 in accordance with software such as a server program.

[0063] In step S101, the user terminal 20 accesses the information processing device 10 and requests a log input screen from the information processing device 10. The information processing device 10 transmits data for displaying the log input screen to the user terminal 20 that is the sender of this request (step S102).

[0064] 7A is a diagram illustrating an example of an introduction screen that is displayed when the information processing device 10 is accessed for the first time. This introduction screen has an area 601, a button 602, and a button 603. The area 601 is an area that displays a message that prompts the user to enter a log. The button 602 is a UI object for displaying a screen for entering a new log. The button 603 is a UI object for displaying a screen for checking or editing a log that was previously entered.

[0065] FIG. 7B is a diagram illustrating an example of a log input screen. This input screen has input fields 611, 612, 613, 614, 615, an area 616, a check box 617, and a button 618. The input field 611 is a UI object for inputting the designation of the type of problematic behavior to be recorded. The types of problematic behavior are divided into five types, for example, lateness / absence, harassment, compliance violation, lack of ability / poor performance, and others. The input field 612 is a UI object for inputting the identification information of the observer (name in this example). The input field 613 is a UI object for inputting the identification information of the perpetrator, i.e., the person who committed the problematic behavior (observed person Ut). This is a UI object for inputting identification information. Input fields 611 to 613 are used to input information in the form of a drop-down list, for example. Input field 614 is a UI object for inputting the date and time when the problematic behavior occurred. Input field 615 is a UI object for inputting the details of the problematic behavior. The details of the problematic behavior are freely entered. Area 616 is a UI object for uploading files. Here, files such as images or audio related to the problematic behavior can be uploaded. Check box 617 is a UI object for inputting whether or not the log should be made public to other related parties, in this example, human resources personnel. When check box 617 is turned on (i.e., checked), this log will be made public to human resources personnel. Button 618 is a UI object for inputting an instruction to record this log.

[0066] In another example, when the observer Us inputs an action log on the log input screen, The user may specify which task item in the job description of the observed person Ut the behavior is related to using a UI object such as a drop-down list. Based on this specification, the information processing device 10 associates the behavior log with the specific task item in the job description and records it in the log database 111.

[0067] Referring again to FIG. 6, the observer Us inputs the behavior log of the observed person Ut on the input screen of FIG. 7B. When the observer Us presses the button 618, the user terminal 20 displays the input behavior log. The observer Us sends the action log to the information processing device 10 (step S103). The information processing device 10 writes the action log sent from the user terminal 20 into the log database 111 (step S104). The observer Us can input the action log at any timing. In this system, the log database 111 accumulates the behavior logs of a plurality of observed persons Ut.

[0068] FIG. 8 is a diagram showing an example of the recorded contents of the log database 111. The log database 111 contains multiple records. Each record corresponds to one behavior log input at a certain timing. Each record contains information on multiple items, specifically, a log ID, the date and time of occurrence, the subject, the type, the content, the person who updated, and the status. The log ID is identification information for uniquely identifying one behavior log. The date and time of occurrence indicates the date and time when the problem behavior occurred. The subject is identification information of the observed person Ut. The type indicates the type of problem behavior. The content indicates the problem. The updater is the identification information of the person who last updated the action log. The status indicates whether the action log is public or not.

[0069] 2-2. Viewing the action log FIG. 9 is a sequence chart showing an outline of processing related to viewing an action log. When a related person instructs viewing of an action log, the user terminal 20 transmits a request to view the action log to the information processing device 10 (step S201). This request includes the user ID of the related person. The information processing device 10 receives the request from the user terminal 20 (step S202). Upon receiving the request, the information processing device 10 extracts, from the log database 111, the action log corresponding to the user ID included in the request and entered by the related person (step S203). The information processing device 10 transmits data for displaying a screen displaying a list of the extracted action logs (hereinafter referred to as a "list display screen") to the user terminal 20, which is the sender of the request for viewing (step S204). The user terminal 20 displays the list display screen (step S205).

[0070] FIG. 10 is a diagram illustrating an example of a list display screen. This list display screen has areas 91 and 92. Area 91 is an area that displays a UI object that specifies the information to be displayed. In this example, the displayed screens include three types: "My Records," "View Records," and "Case Management." "My Records" is a screen that displays a list of action logs entered by the person involved. "View Records" is a screen that displays a list of action logs to which the person involved has access rights. Action logs entered by people other than the person involved are also included in the list, as long as the person has access rights. "Case Management" is a screen that displays a list of cases. Since various information can be viewed by specifying the information to be viewed in area 91, this list display screen can be said to be a type of so-called dashboard screen.

[0071] The list display screen may be provided to a related person who is the observed. In this case, "My Records" is a screen that displays a list of action logs in which the related person is recorded as the actor. For a related person who has two roles, that of observer and observed, the list display screen may be provided with a list relating to the role as observer and a list relating to the role as observed on separate screens, or may be provided on a single screen.

[0072] Figure 10 shows a list of "My Records," but the screen layout is the same for "View Records." In other words, on this screen, the person involved can view the action logs that they are involved in. If the person has the role of an observer (Us), for example, they can view the interviews. It is possible to confirm whether there are interested parties, and based on that confirmation, the parties can record richer information about the action log, such as taking further actions on the action log (such as conducting interviews) and even entering information about the results of those actions.

[0073] Whether it is "My Records" or "View Records," a list of action logs is displayed in the area 92. At least a portion of the attribute information of each action log is displayed in the area 92. The person concerned can input an instruction to view an individual action log by, for example, clicking or tapping on the position of one action log in the area 92. Upon receiving this instruction, the information processing device 10 transmits data to the user terminal 20 for displaying a screen displaying details of each action log (hereinafter referred to as a "detail display screen").

[0074] FIG. 11 is a diagram illustrating a details display screen. On the details display screen, details of a specified behavior log are displayed in area 92. The details of the behavior log include, for example, the full text of the content and playback of attached files. The configuration of this screen and the actions that can be accepted vary depending on the scope of access rights granted to the person involved with the behavior log. For example, a screen that allows only viewing of the details is provided to a person involved with only viewing rights. A screen including a UI object that accepts instructions to edit the details is provided to a person involved with editing rights. From this screen, the person involved can check the details of each behavior log and, as necessary, write or edit additional information. Additional information, for example, the contents of an interview with the person involved after the problem behavior occurred, is recorded.

[0075] The detailed display screen displayed on the user terminal 20 of the person involved who is the observed may include a UI object for inputting an explanation for the action log. By allowing the observed person to input an explanation, it becomes possible to grasp the circumstances from multiple angles, rather than just one-sided evaluation of the observed person from the observer.

[0076] 2-3.Project management 12 is a flowchart showing an outline of a process for generating an instance. This flow is repeatedly executed at regular intervals while the information processing device 10 is in operation.

[0077] In step S301, the information processing device 10 determines whether the conditions for creating a case have been met. The conditions for creating a case are defined in advance. In one example, the condition for creating a case is that a related party with the authority to create a case has explicitly instructed a certain action log to be created as a case. If the conditions for creating a case have been met, the information processing device 10 proceeds to step S302. If the conditions for creating a case have not been met, the information processing device 10 terminates the flow of Figure 12. A series of action logs that have been determined to have met the conditions for creating a case in this way become an example of a "target log" in the subsequent document creation process.

[0078] In step S302, the information processing device 10 creates a record of the case. The record of the case includes the log ID of at least one corresponding action log. In one example, the "View Record" screen includes a UI object (e.g., a button) for instructing the case creation. This UI object is associated with a UI object (e.g., a check box) that specifies the corresponding action log. In other words, when a related person instructs the case creation, the related person specifies the corresponding action log on the case creation screen or the list display screen.

[0079] 13 is a diagram illustrating an example of a case creation screen. The case creation screen has an area 93. Reference numeral 93 denotes an area displaying UI objects for inputting case attributes or related information. In this example, case attributes include the last update date and time, disciplinary action content, occurrence date and time, target, secondary target, and content details. The last update date indicates the date and time the case record was last updated. The occurrence date indicates the date and time the case occurred. The target is a related party who is the target of the disciplinary action. The secondary target is a related party related to the case. Multiple related parties may be specified as secondary targets. The content of the disciplinary action indicates the content of the disciplinary action. In one example, the content of the disciplinary action is determined from predetermined options based on the content of the behavior log. The options include, for example, at least one of disciplinary action, warning, caution, deduction in personnel evaluation, transfer, suspension, recommendation for resignation, and dismissal. The information processing device 10 may automatically input at least some of this information as initial values ​​by referring to the behavior log recorded in the log database 111. The content details indicate the details of the disciplinary action content. The content details are input, for example, by the observer Us.

[0080] In this example, the related information for a case includes logs, files, and disclosure ranges. The logs include a list of action logs corresponding to the case. The UI objects related to the logs include objects for adding or deleting action logs corresponding to the case. When an instruction to add an action log is input via a UI object for adding an action log (e.g., an add button), a list of candidate action logs is displayed. The observer Us selects the action log to add from this list.

[0081] The file indicates an attachment related to the case. An attachment is data in a specific file format, such as a document, image, audio, or video. The file-related UI object includes an object for adding or deleting a file related to this case. The observer (Us) specifies the file to add from this UI object.

[0082] The disclosure range indicates the disclosure range of the case. In one example, the disclosure range includes free-entry text. The party creating the case can input the disclosure range of the case in text. The disclosure range may include a link to an email. In this case, the email sender, email destination, and email text are recorded as the disclosure range. The disclosure range may also include identification information of the party with viewing or editing rights. For example, the identification information of the party who is the advisor Ua By including the information, the information of the case can be shared with the advisor Ua.

[0083] The information processing device 10 inputs the data input via the case creation screen into the case database 112.

[0084] FIG. 14 is a diagram illustrating an example of the case database 112. The case database 112 includes multiple records. Each record corresponds to one case. Each record includes multiple items of information, in this example, the case ID, the date of occurrence, the subject, the secondary subject, the disposition content, the content details, a log list, attached files, and the disclosure range. The case ID is identification information that uniquely identifies the case. The rest is as explained on the case input screen.

[0085] The information processing device 10 outputs data for displaying a list display screen of cases to the advisor Ua who has accessed the information processing device 10 using the user terminal 20. The list display screen of cases has a configuration similar to that of the list display screen of action logs (FIG. 10), for example. Furthermore, when the advisor Ua selects one case from the list of cases, details of that case are displayed. The case detail display screen is similar to the case input screen (FIG. 13). The advisor Ua can view or edit cases according to his / her access authority.

[0086] Referring again to FIG. 12, in step S303, the information processing device 10 It is determined whether the conditions are met. The conditions for document creation are predefined, for example, a condition that an authorized person has given an explicit instruction to create the document. If it is determined that the conditions for document creation are met, the information processing device 10 proceeds to step S304. If it is determined that the conditions for document creation are not met, the information processing device 10 ends the flow of FIG. 12.

[0087] In step S304, the information processing device 10 creates a document. The format of the document to be created for each disposition is defined in advance. In one example, the documents created here include an observation history and a training history. The observation history is a history of observations of the target person's problematic behavior. The training history is a history of training given to the target person. Format templates are defined for the observation history and the training history. The format template defines the areas on the document and the type of information to be entered into those areas. The information processing device 10 creates the observation history and the training history by entering information from the corresponding behavior log into the template. The information processing device 10 stores the created file. The information processing device 10 writes information indicating the location of this file into the case database 112. In the case database 112, this file is displayed, for example, as an attachment.

[0088] A related person who is an advisor Ua accesses the information processing device 10 from his / her user terminal 20. The advisor Ua can be, for example, a lawyer, a certified social insurance labor consultant, or Professionals such as industrial counselors who can view or download the observation and guidance resumes and advise or prepare for disciplinary or legal action in the organization.

[0089] As explained above, the information processing system 1 allows an organization to record actions that may become problematic in the future, and creates an environment in which necessary information can be provided in an emergency. Furthermore, when a situation arises in which specific disciplinary action or legal action is required, information can be easily and quickly shared with the advisor Ua, contributing to the early resolution of the problem. By creating an environment where problems can be solved in this way, the overall environment of the organization can be improved.

[0090] 3. Second embodiment 15 is a block diagram showing the functional configuration of an information processing system 2 according to the second embodiment. Here, common reference numerals are used for elements common to the information processing system 1 according to the first embodiment, and detailed explanations will be omitted. The information processing system 2 has an AI 40 and performs processing utilizing the AI ​​40.

[0091] In this example, the information processing device 10 has a determination means 14, a determination means 51, an access means 52, an extraction means 53, and an estimation means 54. The determination means 51 determines whether a target log among the logs registered in the log database 111 satisfies the conditions for document creation (an example of a first determination means). When a grouping (case) determination is made, the determination means 14 is an example of a second determination means. When it is determined that the target log satisfies the conditions for document creation, the creation means 17 creates a document in a predetermined format from the contents of the target log among the logs recorded in the log database 111.

[0092] The access means 52 accesses the AI ​​40. The AI ​​40 includes at least one of a general-purpose model and a model trained for a specific application. In one example, the AI ​​40 includes a generative AI model. When data in a format commonly handled by humans, such as natural language, images, or voice, is input, the generative AI model outputs data corresponding to the input. The output data is also data in a format commonly handled by humans, such as natural language, images, or voice. In other words, the AI ​​40 may include an LLM. Alternatively, the AI ​​40 may be a specific model readable by a computer. The AI ​​40 includes a machine learning model that, when data of a certain format is input, outputs data corresponding to the input data. In the information processing system 2, the AI ​​40 can be used for various purposes, and individual models may be used for each purpose. In Fig. 15, only a single AI 40 is shown for simplicity.

[0093] The extraction means 53 and the estimation means 54 are both examples of functions that utilize AI 40. The extraction means 53 uses AI 40 to extract clauses from the rules related to the organization that may conflict with the behavior related to the target log. In this case, the creation means 17 creates a document including information about the extracted clauses. The information about the extracted clauses is, for example, information that identifies the extracted clauses (e.g., "Article 12, Paragraph 1 of the Work Rules"). The estimation means 54 uses AI 40 to estimate the cost of resolving the problem related to the target log. Data including this information is output from the output means 16 to another device such as the user terminal 20. In other words, the output means 16 is an example of a presentation means that presents this information to the user.

[0094] In this example, when the CPU 101 is executing a server program, the CPU 101 is an example of the determination means 51 and the access means 52. The AI ​​40 may be implemented in any hardware, and may be implemented in the information processing device 10 itself, or may be implemented in a server separate from the information processing device 10.

[0095] Below are some examples of processes that utilize the AI ​​40. Each of the following processes may be applied independently, or at least a part of one process may be applied in combination with at least a part of another process.

[0096] (1) Extraction of conflicting clauses In this example, the AI ​​40 extracts, from the target log, clauses of the organization's predetermined rules that may conflict with the behavior related to the behavior log. The AI ​​40 includes a machine learning model trained using training data that includes identification information for the rule or rules similar to the rule, the behavior of the relevant person, and clauses of the rule that may conflict with the behavior. The training data may include the organization's own rules, or may include rules of a different organization (e.g., the employment rules of a competitor with similar attributes, such as industry or size).

[0097] In one example, the target log is an action log newly registered in the log database 111 (i.e., newly input by the user). In this case, when a new action log is input, the information processing device 10 determines whether the action log satisfies the conditions for document creation. If it is determined that the conditions for document creation are met, the information processing device 10 inputs the action log to the AI ​​40. The AI ​​40 extracts and outputs information indicating clauses of previously learned rules (e.g., the company's work rules) with which the action related to the action log may conflict. In another example, the target log is an action log belonging to a specific group of action logs grouped in the log database 111. In this case, for example, when a user inputs an instruction to check whether a certain case complies with the rules, the information processing device 10 inputs a series of action logs belonging to the case to the AI ​​40. The AI ​​40 extracts and outputs information indicating clauses of the corresponding rules with which the action related to the action log may conflict. In this way, the number of target logs may be one or more.

[0098] (2) Cost estimation In this example, the AI ​​40 outputs the cost of resolving the problem related to the action log for the target log. The AI ​​40 includes a machine learning model trained using training data including information on the actions of the people involved and the cost of resolving the problem related to the actions. More specifically, the information processing device 10 inputs the target log and related information to the AI ​​40. The AI ​​40 calculates the cost (predicted value) of resolving the problem related to the target log from the target log and related information. The related information is information related to the cost of solving the problem, and includes, for example, attribute information of the parties involved (observer Us, observed Ut, human resources personnel, etc.). If necessary, the information processing device 10 extracts job title, rank data, etc. of each user from the user database 113 and inputs it to the AI ​​40 as related information.

[0099] The costs of resolving the problem include at least one of the following: "direct costs," such as the labor costs of employees involved in training or interviews; "indirect costs," such as losses due to reduced productivity for the observed individual or surrounding employees; and "external costs," such as expert fees paid to the advisor (Ua). In one example, the AI ​​40 outputs the amount for each of these multiple types of costs. For example, the AI ​​40's machine learning model uses training data to learn correlations between numerous past case data and performance data (e.g., sales, production volume, project progress delay rate, etc.) of the subject and their department for the corresponding period. This makes it possible to statistically predict the amount of loss due to potential productivity losses when a specific type of problematic behavior is recorded.

[0100] The predicted costs are recorded in a database such as the user database 113. The breakdown of costs may be visualized using a pie chart, bar graph, or the like, so that management and human resources personnel can intuitively grasp the scale of the damage. Furthermore, if multiple cases occur regarding a specific observed person Ut, the accumulated costs may be displayed in chronological order to indicate the seriousness of the problem, or a future predicted loss amount may be presented based on past similar cases and trends in behavior logs, such as "If no improvement is seen, there is a possibility of an additional loss of approximately XXX yen over the next year," to support more in-depth decision-making.

[0101] (3) Preparation of written documents In this example, the AI ​​40 outputs a document created in a predetermined format based on the accumulated action log or information output from the AI ​​40 (or other models). The data format of the document can be specified by the user, and is, for example, a text file, a word processor file, or a PDF file. The document in the predetermined format may include, for example, at least one of the following: This can be done.

[0102] (a) Proposal of an action plan to resolve issues related to the behavior log (see 4-4). (i) A proposal for a development plan or education plan for relevant parties (see 4-11). (c) Evidence packages used in legal proceedings regarding problem behavior (see 4-6). (d) Analysis report of the behavioral characteristics and psychological tendencies of the people involved (see 4-10). In order to output each of the above documents, the AI ​​40 has trained using sample documents of each format or documents from past cases as training data. These documents may include information obtained by other functions of the AI ​​40. For example, these documents may include information indicating costs estimated by the AI ​​40 (e.g., a breakdown of costs, predicted future losses, etc.).

[0103] (4) Support for rule revisions In this example, AI 40 compares problematic behavior with rules (e.g., work rules) and proposes how to revise the rules to prevent or reduce the occurrence of problems. For example, the rules database 114 records the work rules of multiple companies that are users of the information processing system 1. AI 40 compares problematic behavior with its own work rules and further references the work rules of other companies to propose revisions for clauses that are missing from its own work rules or wording or expressions that should be revised. This allows the information processing system 1 to evolve its own work rules into living rules that are more in line with reality, based on actual data and by referring to examples from other companies. Note that the user may be able to arbitrarily select the work rules of other companies to reference from those for which permission to use has been obtained in advance.

[0104] 4. Variations The present invention is not limited to the above-described embodiment, and various modifications are possible. Some modifications will be described below. Two or more of the following features may be applied in combination.

[0105] 4-1. Entering the action log The input method and input items of the behavior log are not limited to those exemplified in the embodiment. For example, at least a part of the behavior log may be input using a voice input technique. In this case, the user terminal 20 has a voice input means (not shown) that accepts voice input. The voice input means is, for example, a microphone mounted on the user terminal 20. The user can input the details of the problem behavior by voice into the user terminal 20. The user terminal 20 transmits the input voice to the information processing device 10 as a stream of voice data. The information processing device 10 converts this voice data into text data using a voice recognition engine. The voice recognition engine may be an internal system of the information processing device 10 or an external system. The information processing device 10 writes the text data obtained in this manner into the log database 111 as a behavior log.

[0106] In this case, the information processing device 10 may have an annotation means for adding annotations to the text data. An annotation refers to additional information added to a part or all of the text data, and may include, for example, the impression of the observer U when he / she inputs the action log. The information includes at least one of the following: the information about the event, the importance of the action log, and keywords in the action log.

[0107] If emotions are to be added as annotations, the information processing device 10 uses an emotion analysis engine that analyzes emotions from voice data to assign emotions to the action log. This emotion analysis engine includes, for example, a machine learning model. This machine learning model is, for example, a model that performs machine learning by providing a set of voice data and the emotions at that time as training data, with the voice data provided to an input layer and the emotions provided to an output layer. This training data does not have to be specialized for action logs, or may be specialized for action logs. Alternatively, the importance analysis engine may include an LLM (Large Language Model). Emotions can be obtained by inputting an instruction to this LLM to output emotions from the text data of the action log.

[0108] If importance is assigned as an annotation, the information processing device 10 assigns importance to the behavior log using an importance analysis engine that analyzes the importance of the behavior log from text data. This importance analysis engine includes, for example, a machine learning model. This machine learning model is, for example, a model that performs machine learning by providing a set of behavior logs and their importance (importance assigned by humans) as training data, with the behavior logs as an input layer and the importance as an output layer. Alternatively, the importance analysis engine may include an LLM. The importance can be obtained by inputting an instruction to the LLM to output the importance from the behavior log. In one example, the following factors may affect the importance: (a) the type of problematic behavior, (b) the frequency of occurrence of the problematic behavior, (c) the content of the problematic behavior, (d) the attributes of the person involved (e.g., position, age, past experience, etc.), and (e) relevance to work regulations. The information processing device 10 may analyze the importance by taking into account at least some of these factors.

[0109] If keywords are to be added as annotations, the information processing device 10 extracts keywords from the action log using a keyword extraction engine that extracts keywords from the action log from text data. This keyword extraction engine may be based on, for example, machine learning. The importance analysis engine may include an LLM. Keywords can be obtained by inputting an instruction to the LLM to output keywords from the behavior log.

[0110] The sentiment analysis engine, importance analysis engine, and keyword extraction engine may be internal systems of the information processing device 10 or external systems. The annotations added to the action log are recorded in the log database 111.

[0111] When these annotations are added, the information processing device 10 may display the contents of the annotations on a list display screen or a detailed display screen of the action logs. Alternatively, when selecting a corresponding action log to generate a case, the information processing device 10 may select the action log by referring to the annotations added to each action log. In one example, the information processing device 10 refers to the annotations in such a way that it selects an action log containing a specific keyword for a specific type of case, and selects an action log with a specific importance for another specific type of case.

[0112] That is, the example in which annotations are used is an example in which the information processing device 10 has a text conversion means for converting speech into text and an annotation means for adding annotations to the text based on speech.

[0113] The input items of the behavior log may include at least one of the group's actions and the warning to the observed person, instead of or in addition to the examples described in the embodiments. The group's actions indicate the action taken by the group regarding the behavior. The warning to the observed person indicates the fact that the observer or other related parties have warned the observed person, or the content of that warning. By including these pieces of information, it is possible to record a wider variety of information.

[0114] 4-2. Presentation of related information When or after inputting an action log, the information processing device 10 may present information related to the action log. The related information refers to information related to the action log among the information previously registered in the rule database 114 (not shown). The rule database 114 records at least one of the organization's relevant laws and regulations and the organization's own rules. The organization's own rules include, for example, company rules such as work rules. If the rule database 114 includes work rules, the information processing device 10 extracts clauses of the work rules that may be in conflict with the action log. The information processing device 10 presents the extracted information to the relevant person. For example, the information processing device 10 provides information such as "This action may be in conflict with Article 3, Paragraph 2 of the work rules" to the relevant person.

[0115] To extract related information, for example, a machine learning model is used. For example, if the related information is work regulations, this machine learning model is a model in which a set of an action log and a clause of the work regulations that the action log conflicts with is given as training data to the input layer and the clause to the output layer, respectively, and machine learning is performed. Alternatively, the related information may be extracted using LLM. The extracted related information can be obtained by inputting an instruction to this LLM to extract (extract) the part of the related information that is related to the action log from the action log and the related information. The information processing device 10 automatically associates the extracted information of the clause that may conflict with the corresponding action log or case and records it. This accumulates an objective record of which rules, etc. each action log conflicts with, and can be used later when creating documents, etc. When presenting the extracted related information to the relevant person, the information processing device 10 highlights (decorates by underlining, bolding, changing color, etc.) the extracted part while displaying the entire text of the related information. Alternatively, the entire relevant information may be displayed without being added to the relevant information, or only the extracted portion may be displayed. The extracted relevant information may be presented to either the observer Us or the observed Ut. In this way, the information processing device 10 extracts and highlights relevant information (e.g., clauses of work rules) from the input behavior log, allowing the user to easily understand the relationship between the problematic behavior and the work rules. This allows the relevant parties to take appropriate measures against the problematic behavior.

[0116] Furthermore, the information processing device 10 may not only present potentially conflicting clauses but also more specifically indicate the basis for its determination. In this case, an extraction means using AI (e.g., natural language processing technology or LLM) closely compares the contents of the cased log with the work rules, various policies, relevant laws, and regulations stored in the rules database 114. Then, rather than simply presenting the conflicting clause (e.g., "Work Rules, Article X, Service Discipline"), the information processing device 10 may automatically generate a sentence explaining the basis for the conflict, such as "The statement 'Leaving the workplace without permission' in the log specifically corresponds to the provision 'Prohibition of leaving the workplace without permission' in this clause," and include this in a report, etc. Furthermore, the AI ​​may learn from past judicial precedents or data on disciplinary actions taken within an organization, and automatically score and visualize the legal risk and importance of the case in the form of "high," "medium," or "low." In addition, the AI ​​may search and extract similar cases from the past "Case Database 112" and provide reference information such as "In similar cases in the past, 'reprimands' were given," thereby assisting the judgments of the observer Us and the advisor Ua.

[0117] In other words, this example is one in which the information processing device 10 has an access means for accessing the organization's work rules, an extraction means for extracting clauses from the work rules that may conflict with the behavior log, and a presentation means for presenting the extracted clauses.

[0118] 4-3. Assistance with entering action logs In the above-described embodiment, an example has been described in which a fixed screen (i.e., a fixed set of input items) is used as the input screen for the behavior log. However, the information processing device 10 may have multiple templates as input screens and suggest to the person concerned to use a template appropriate for the situation. In one example, each template corresponds to one of multiple types of problematic behavior. For example, a dedicated input screen for lateness / absence is provided for tardiness / absence, and a dedicated input screen for harassment is provided for harassment.

[0119] The information processing device 10 takes into consideration the behavior log of the observed person Ut up to that point. The observer Us is asked to enter a behavior log on the input screen of the template according to the circumstances. In one example, this "situation" or "circumstance" may be the This includes the transition or context of the behavior log in the problem behavior. This can be said to suggest that there is a certain pattern (or frame or story) in the problem behavior, and that the observer Us should record the behavior log according to that pattern. For example, It may be known from past experience that there is a high possibility that a certain type of problematic behavior (for example, absenteeism without permission) will occur after a certain type of problematic behavior (for example, lateness) occurs with a frequency above a threshold. In other words, this is equivalent to predicting the future behavior log of a certain person concerned based on the past track record of the transitions in the behavior logs of multiple (preferably many) people concerned. In addition, at this time, knowledge may have been accumulated that "if it was known that a certain type of problematic behavior (for example, absenteeism without permission) would occur in the future, it would have been better to record this information even before the problematic behavior actually occurred." For example, when a certain type of problematic behavior occurs with a frequency above a threshold, the information processing device 10 notifies the observer Us that a certain type of problematic behavior is likely to occur. Based on this knowledge, the observer Us may be suggested to enter a behavior log using an input screen that includes input items corresponding to a particular type of problem behavior.

[0120] That is, in this example, the information processing device 10 receives a log input including items according to the type of behavior. In this example, the relationship between the transition or context of the action log and the template may be defined on a rule basis, or may be determined by AI (Artificial Intelligence) technology such as a machine learning model or LLM. stomach.

[0121] Instead of or in addition to proposing an input template, the information processing device 10 may provide information that assists input of an action log. The information that assists input is, for example, an example sentence of a sentence to be entered into a specific item (e.g., "content") of the action log when that item is entered. The example sentence may be registered in a database, or may be generated by referring to past action logs using AI technology such as LLM.

[0122] 4-4. Behavioral advice The information processing device 10 may provide behavior advice to the person concerned. As described in the section "Proposal of input template", the information processing device 10 may provide behavior advice to the person concerned by proposing the following behavior advice from the behavior log of the observed person Ut up to that point. It is possible to predict problem behaviors that the observed person Ut is likely to exhibit. Advice for the person concerned may be generated based on this prediction. The advice for the person concerned is generated by taking into consideration, for example, at least one of the following factors: (a) the attributes of the person concerned (e.g., age, gender, position, personality, work experience, etc.), (b) the type of problematic behavior, (c) the circumstances under which the problematic behavior occurred (e.g., frequency, time of day, location, surrounding circumstances, etc.), (d) the explanation or reflection of the person concerned recorded in the log, (e) the content of rules such as work regulations and guidelines for behavior, and (f) effective advice based on similar past cases.

[0123] The person to whom the advice is given may be the observed person Ut, the observer Us, or any other person (for example, a colleague belonging to the same group). This can also be seen as a preventative measure to prevent problems from occurring. The advice provided to others other than the observed person (Ut) includes guidelines for curbing problematic behavior, such as "take a vacation" or "relieve stress." For example, to change the environment of the person being observed, such as "regulating overtime" or "transferring the person to a different position," In particular, from the pattern of the behavior log, it is possible to determine the influence of the problem behavior of the observed person Ut. If it is possible to identify a person concerned who is likely to be affected, the information processing device 10 may provide advice to the identified person concerned.

[0124] The information processing device 10 generates advice for the person concerned using an advice engine that generates advice from the history of the action log. This advice engine includes, for example, a machine learning model. This machine learning model is a model that performs machine learning by providing, for example, past action log transitions (a set of multiple action logs for the same person concerned) and a set of advice for the person concerned (e.g., advice provided by an expert), with the action log transitions provided to the input layer and the advice provided to the output layer as training data. Alternatively, the advice engine may include an LLM. Importance can be obtained by inputting an instruction to this LLM to output advice from the action log transitions.

[0125] By providing such advice, more effective guidance and advice can be given to the person concerned, and problems can be prevented from becoming more serious. Furthermore, the person concerned can reflect on his or her own actions and be encouraged to make improvements. The information processing device 10 may provide more personalized advice through interactive exchanges with the person concerned. For example, the advice engine may ask questions to the person concerned and adjust the content of the advice depending on the answers.

[0126] Furthermore, the information processing device 10 has developed the advice engine described above, and has a function of automatically generating a specific "improvement action plan proposal" based on the contents of the cased log. In this function, the analysis means using AI technology comprehensively analyzes information such as patterns of behavior logs related to the case (e.g., frequent occurrence of problematic behavior on specific days of the week or during specific time periods), attributes of the observed person recorded in the user database 113 (e.g., job title, past career history), and rules recorded in the rule database 114. Based on the analysis results, the creation means 17 generates an improvement action plan proposal that describes specific solutions. The proposal includes specific action plans to directly support the manager's decision-making, such as "holding weekly one-on-one meetings," "recommending participation in a training program for specific skills," and "temporarily changing work content."

[0127] 4-5. Decision on whether to develop a project The method of determining whether to case or not and the conditions for case determination are not limited to those exemplified in the embodiment. For example, the determination of whether to case or not may be made automatically or semi-automatically using AI technology. In one example, the determination of whether to case or not is made using LLM. In this case, the information processing device 10 determines whether the necessity of case or the requirements for case determination are met from the transition of the behavior log of a certain observed person Ut. An instruction to output the degree to which the observed person Ut meets the necessity of making the event into a case is input to this LLM. If the degree to which the observed person Ut meets the requirements for making the event into a case output from the LLM exceeds a threshold, the information processing device 10 determines that the observed person Ut's behavior log should be made into a case. In this case, The information processing device 10 may use RAG technology to input information used to determine whether to treat a case into the LLM. In one example, this RAG system has a database that records the history of behavior logs that have been treated as cases in the past in the organization. When the behavior log to be determined is identified, the information processing device 10 searches the database for past cases that are similar to the behavior log. Similar cases are cases with similar attributes related to the behavior log. The attributes related to the behavior log are the attributes (e.g., age, gender) of the observed person Ut who is the subject of the behavior log. The information processing device 10 instructs the LLM to attach the similar cases extracted from the database and the transition of the target action log, and to output the degree of necessity of turning the transition of the target action log into a case. Note that the database in the RAG system may include not only data on transitions of action logs that have been turned into cases in the past, but also data on transitions of action logs that have not been turned into cases. In this case, the information processing device 10 instructs the LLM to attach the similar cases extracted from the database, whether the similar cases have been turned into cases, and the transition of the target action log, and to output the degree of necessity of turning the transition of the target action log into a case.

[0128] In another example, the determination of whether to make a case is performed using a machine learning model. This machine learning model is a model that performs machine learning using a set of data of past cases, including transitions in action logs, attributes related to the action logs, and whether the case has been made a case, as training data, in which the transitions in action logs and attributes related to the action logs are provided to an input layer and the data of whether the case has been made a case is provided to an output layer.

[0129] "Automatically using AI technology" means that the information processing device 10 uses the above-mentioned AI technology to decide whether to make a case or not without relying on instructions from the person involved. "Semi-automatically using AI technology" means that the information processing device 10 uses the above-mentioned AI technology to output the necessity of making a case or the degree to which the requirements for making a case are met from the transition of the action log, and the person involved refers to this information to give instructions to make a case.

[0130] Alternatively, the decision to issue an incident may be made automatically or semi-automatically using rule-based technology rather than AI technology. In this case, the conditions for issue identification are defined in relation to the transition of the behavior log. For example, the conditions for issue identification may be "if the employee is late four or more times within two weeks, an issue will be identified."

[0131] According to these examples, the necessity of projecting or the degree to which the requirements for projecting are met is a key factor in information processing. Since the data is output from the management system 1, it is possible to make objective judgments without relying on the subjective judgments of the people involved.

[0132] The determination of whether to make a case may be omitted in the information processing system 1. That is, the behavior logs may be grouped according to a predetermined criterion (for example, by observed person), and it may be determined for each group whether the conditions for creating a written document are met (step S303), and if the conditions are met, a written document may be automatically created (step S304).

[0133] In the embodiments (particularly the first embodiment), an example has been described in which a subsequent process (for example, a decision on whether to prepare a document) is performed after a decision on whether to create a case has been made, but the decision on whether to create a case may be omitted. That is, the information processing device 10 may perform a process such as a decision on whether to prepare a document without making a decision on whether to create a case for the input action log.

[0134] 4-6. Preparation of documents The conditions for creating a document are not limited to those exemplified in the embodiments. For example, conditions such as a predetermined time having passed since an action log that satisfies a specific condition was recorded, or a predetermined amount of action logs that satisfy a specific condition have been accumulated, may be adopted as conditions for creating a document. Furthermore, the document to be created is not limited to the examples in the embodiments. Any document may be created as long as it contributes to solving problems in the organization that employs the information processing system 1.

[0135] As a specific example of the documents to be created, the information processing device 10 may have a function to automatically generate an "evidence package for legal and official proceedings" specifically designed to enhance the evidentiary value of documents intended for use in official proceedings such as disciplinary actions and litigation. In this function, the creation means 17 automatically collects and integrates all action logs, attached audio and image files, instruction history, and message history between parties related to the cased event from the log database 111 and the case database 112. The collected information is organized in strict chronological order, and a report is generated that objectively reconstructs the overall picture and history of the event. At this time, each log and file may be assigned a timestamp or hash value, which may be included in the report as information to prove their authenticity. Furthermore, audio logs may be automatically converted to text and recorded along with speaker information. Furthermore, to facilitate review by the advisor Ua, such as a lawyer, an analysis means using AI technology may highlight important points or automatically generate summaries of potential legal issues. In addition, based on the collected evidence, it is possible to automatically generate drafts of various forms required for legal procedures, such as a "Notice of Disciplinary Action," a "Notice of Opportunity to Defense," and a "Notice of Termination."

[0136] 4-7. Collaboration with Advisor Ua The cooperation with the advisor Ua, particularly with the external expert, is not limited to the example of the embodiment. When a new case is created, the processing device 10 notifies the advisor Ua that a new case has been created. The "notification by push" means that information is sent from the information processing device 10 to the advisor Ua voluntarily, for example, by sending information by e-mail or by using a dedicated account. In this case, the advisor Ua who is the notification recipient is the user database. The advisor Ua to be notified may be defined for each type of case. good.

[0137] The push notification from the information processing device 10 to the advisor Ua includes at least one of the following information: (a) a link to display the case details screen for that case, (b) a list of action logs for that case, (c) a link to an input screen for entering advice regarding that case, and (d) a link to hold a video conference regarding that case.

[0138] The information processing device 10 may, instead of or in addition to the push notification to the advisor Ua, On the display screen or the detailed display screen, the advisor Ua can chat (i.e., exchange messages). In this example, the group relationship may provide UI objects for closer communication with the advisor Ua. You can take it.

[0139] 4-8. List display screen The list display screen (or dashboard screen) of cases or action logs is not limited to the example of the embodiment. For example, the list display screen of action logs may include the aggregation result of problematic behaviors. The aggregation result of problematic behaviors is the result of aggregating the action logs recorded in the log database 111. The aggregation is performed according to a predetermined aggregation unit. The aggregation unit is, for example, a group in an organization (specifically, a department or section). The information processing device 10 aggregates problematic behaviors for each group. The aggregation result is displayed in a graph such as a bar graph. Alternatively, the aggregation result is shown by highlighting action logs whose aggregation value meets a condition (for example, those whose occurrence frequency exceeds a reference value) more than other action logs. The highlighting includes decorations such as adding an icon or badge, flashing, bolding, underlining, or changing the color. Alternatively, the aggregation of problematic behaviors may be performed for each observed person Ut. That is, in this example, the information processing device 10 aggregates problematic behaviors for each observed person Ut (when an organization has multiple groups). This is an example in which the system has a group, aggregating means for aggregating problematic behaviors in the log for each group, and a presentation means for presenting information showing the aggregation result of the problematic behaviors.

[0140] In another example, the list display screen includes the elapsed time of the problematic behavior. The elapsed time of the problematic behavior refers to the time elapsed since the time the behavior log was entered (or the time the behavior related to the behavior log occurred). The elapsed time is indicated, for example, by a numerical value (e.g., "10 hours," "2 days," etc.). Alternatively, the elapsed time is indicated by highlighting behavior logs whose elapsed time meets a condition (e.g., whose elapsed time exceeds a threshold) more than other behavior logs. That is, in this example, the information processing device 10 has a measuring means for measuring the elapsed time since the problematic behavior in the log and a presentation means for presenting information indicating the measured elapsed time.

[0141] By adding such information to the list display screen, the person in charge can easily understand the occurrence status and elapsed time of problem behavior. For example, by looking at the list display screen, the person in charge can understand which group has a lot of problem behavior and how much time has passed since the occurrence of each problem behavior.

[0142] 4-9. Obtaining behavior logs In the embodiment, an example in which the observer Us inputs the behavior log has been described. The method is not limited to this. The information processing device 10 may record information acquired from other systems as an action log automatically, or as an action log of the observer Us after confirmation by the observer Us. The other systems include, for example, at least one of an attendance management system, an IT security system, and a personnel evaluation system. In one example, the information processing device 10 is set up to link with these systems. For example, when the observer Us inputs an action log related to "lateness / absence" for a certain observed person Ut, the information processing device 10 acquires the record of the observed person Ut's arrival time from the attendance management system and stores it in the log database as an action log. Alternatively, when an employee is late in the attendance management SaaS, the information processing device 10 acquires a record of the lateness through the API of the system or the like, and records it as an action log in the log database 111. As another example, when a tool for acquiring operation logs of a PC used by an employee detects evidence that a specific file has been downloaded or a history of access to an inappropriate website, the information processing device 10 acquires the operation log and records it as an action log in the log database 111. Also, for example, when an observer Us inputs an action log regarding "lack of ability" for a certain observed person Ut, the information processing device 10 acquires a record of the personnel evaluation of the observed person Ut from a personnel evaluation system, and records it as an action log. In this example, the time and effort required for input by the observer Us is reduced to As another example, when the observer U inputs a behavior log relating to "poor grades" for a certain observed person Ut, the information processing device 10 acquires the PC operation log of the observed person Ut from the IT security system. The behavioral log is recorded in the log database 111 as a behavioral log. The relationship between the type of behavioral log (e.g., "lack of ability" or "poor performance") and the source of the log data (e.g., personnel evaluation system or IT security system) is defined. According to this example, the input of the observer Us is This can reduce the effort required.

[0143] The behavior log may include a log of the behavior of the observed person Ut after he / she leaves the company. In one example, the activity log is collected from a publicly available source (e.g., the profile of a particular social networking account). The information processing device 10 acquires the information from a preset account for the observed person Ut. The information processing system 1 periodically acquires information that matches the count, keyword, or behavioral pattern and records it as a behavior log. You can expand the range.

[0144] 4-10.Analysis of behavior logs In the embodiment, an example has been described in which a document in a predetermined format is created from the behavior log, but the method of using the behavior log is not limited to this example. For example, instead of or in addition to creating a document, the information processing device 10 may display the results of analyzing the contents of the behavior log on the user terminal 20. The analysis results include, for example, the number of occurrences of problematic behavior, the number of occurrences by type of problematic behavior, and the occurrence trend of problematic behavior.

[0145] As another example of behavior log analysis, the information processing device 10 may have a function for calculating and visualizing the costs of the activities based on recorded observation history and guidance history (e.g., a series of behavior logs and response records included in a case) using AI technology. In this function, the AI ​​extracts information from the behavior log, such as the time spent by the observer (e.g., supervisor) or human resources staff to respond to the case and the time the observed person took time off work, and multiplies this information by the unit labor cost corresponding to the position, etc., of each person involved, pre-recorded in the user database 113, etc., to calculate labor costs. The calculated costs are automatically added and recorded to the corresponding case information. This allows the invisible costs spent on responding to problematic behavior to be quantitatively visualized.

[0146] In analyzing the behavior log, the information processing device 10 may extract potentially problematic parties from the behavior log and the contents of the rule database 114. The extraction of potentially problematic parties is performed using, for example, AI technology. AI technology includes LLM or a machine learning model. When an LLM is used, the information processing device 10 inputs instructions to the LLM to extract potentially problematic parties from the behavior log and work rules. When a machine learning model is used, the machine learning model is a model trained by providing data indicating past behavior logs and the presence or absence of problematic behavior (or disciplinary action) as training data, with the behavior log as an input layer and the presence or absence of problematic behavior (or disciplinary action) as training data in an output layer. That is, this example is an example in which the information processing device 10 has extraction means for extracting potentially problematic parties from the log and work rules.

[0147] Furthermore, the information processing device 10 may have the function of not only aggregating events but also analyzing the underlying causes and psychological tendencies behind the series of recorded behavior logs, and automatically generating a "behavioral characteristics and psychological tendencies analysis report." In this function, an analysis means using AI technology analyzes the log history of a specific observed person Ut in chronological order, and detects correlations between behaviors and specific signs, such as, for example, "logs related to 'lack of ability and poor performance' increase the week after overtime hours exceed XX hours per month." In addition, emotions can be analyzed from the tone of voice during voice input as described in Variation 4-1 and words selected in the text log, and the results can be used to generate reports such as "dissatisfied or unhappy in the past month." Changes in psychological state are visualized in the form of, for example, "There has been a sharp increase in the frequency of keywords indicating 'stress.'" By generating a report that integrates these analysis results, observers Us and advisors Ua can understand not only the superficial problematic behavior but also its root causes, enabling them to take more fundamental and effective measures (such as setting up opportunities for dialogue or adjusting workloads).

[0148] The analysis of the action log may include evaluation of the action log. For example, the evaluation of the action log may be carried out by aggregating the content (e.g., type of positive or negative action) or number of action logs associated with a certain task item within a specific period. The aggregation result is then compared with the KPI (Key Performance Index) set for the task item, thereby determining the KPI. The degree of achievement is calculated quantitatively and the evaluation results are output as a report, making it possible to evaluate job performance based on objective data.

[0149] 4-11.Other Organizations that use the recording service are not limited to businesses. This recording service can be used by any organization, including non-business organizations such as government agencies, local governments, or non-profit organizations, educational institutions such as schools or cram schools, medical and nursing care facilities such as hospitals or nursing homes, and sports facilities such as baseball stadiums or soccer fields. For example, if the organization is a school, the observed person could be a student or parent, or if the organization is a web service provider, the observed person could be a user of the service.

[0150] The behaviors recorded in the recording service are not limited to problematic behaviors, but may also include behaviors that should be well-received in the organization or recommended behaviors.

[0151] When the recording target is expanded to include positive behaviors as described above, the information processing device 10 may have a function of automatically generating a "talent development proposal" (an example of a development plan) that supports the growth of individuals as an output of the analysis results. In this function, AI technology is used. The analysis means analyzes positive behavior logs, such as "active support for colleagues" and "suggestions for work improvement" recorded in the log database 111, to identify the potential strengths and competencies possessed by the person involved. Then, the creation means 17 generates a proposal including a specific development plan and career path, such as "recommending participation in leadership training" or "appointing the person as a sub-leader of a new project," based on the identified strengths. As a result, the information processing system 1 can be used not only to manage problematic behaviors, but also as a tool that actively contributes to human resource development and organizational development.

[0152] The classification of users is not limited to the example embodiment, and other classifications may be used instead of or in addition to observer, observed, and advisor.

[0153] The UI and data described in the embodiments are merely examples. Some or all of these may be replaced with UI or data in other forms. In addition, the databases described as multiple databases in the embodiments may be partially combined and operated as a single database.

[0154] The functional configuration and hardware configuration of the information processing system 1 described in the embodiments are merely examples. Some of the functional elements or hardware elements described in the embodiments may be omitted, or new functional elements or hardware elements may be added. Furthermore, the correspondence between the functional elements and hardware elements described in the embodiments is also merely an example. Each functional element may be implemented in any hardware as long as it can implement the functions required for the information processing system 1. For example, at least some of the functions described in the embodiments as being possessed by the information processing device 10 may be implemented in the user terminal 20. The information processing device 10 may be a physical server or a virtual server (including a so-called cloud).

[0155] At least a part of the processing described in the embodiment as being performed using AI may be implemented using a non-AI method, such as a rule-based method.

[0156] The program executed by a processing device such as CPU 101 may be distributed in a form recorded on a computer-readable non-transitory recording medium (e.g., a DVD-ROM), or may be provided in a form that can be downloaded from a server on a network. [Explanation of symbols]

[0157] 1...information processing system, 10...information processing device, 11...storage means, 12...receiving means, 13...writing means, 14...judgment means, 15...registration means, 16...output means, 17...creation means, 19...control means, 20...user terminal, 30...network, 91...area, 92...area, 93...area, 101...CPU, 102...memory, 103...storage device, 104...communication device, 107...bus, 111...log database, 112...case database, 11 3...user database, 114...rule database, 201...CPU, 202...memory, 203...storage device, 204...communication device, 205...input device, 206...output device, 207...bus, 601...area, 602...button, 603...button, 611...input field, 612...input field, 613...input field, 614...input field, 615...input field, 616...area, 617...checkbox, 618...button, Ua...advisor, Us...observer, Ut...observed

Claims

1. a receiving means for receiving an input of a log of the behavior of an observed person, which is any one of a related person in an organization, a robot, and an AI, the log including a designation of one of a plurality of defined types; a write means for writing the log into a log database; a first determination means for determining whether a target log among the logs recorded in the log database satisfies a condition for document creation; an extraction means for extracting clauses of rules relating to the organization that may be in conflict with the behavior, using (a) a generation AI including a large-scale language model that, when a log and the rule are input, outputs clauses in the rule that may be in conflict with the log, or (b) an AI including a machine learning model trained using training data that includes the rule or a rule similar to the rule, the behavior of the observed person, and identification information of clauses in the rule that may be in conflict with the behavior (in either case of (a) or (b), if the observed person is an AI, an AI other than the AI); a creation means for creating a document in a predetermined format from the contents of the target log recorded in the log database when it is determined that the target log satisfies the conditions for document creation, the document including the extracted clause; An information processing device having the above.

2. a second determination means for determining whether the logs registered in the log database satisfy a grouping condition; The target logs are logs that satisfy the grouping conditions. The information processing device according to claim 1 .

3. an estimation means for estimating a cost for solving the problem related to the target log; The information processing device according to claim 1 ,

4. The estimation means estimates the cost using an AI (if the observed person is an AI, an AI other than the AI) including a machine learning model trained using training data including information indicating the behavior of the observed person and the cost of solving a problem related to the behavior. The information processing device according to claim 3 .

5. The creating means creates the document including the information indicating the cost. The information processing device according to claim 3 .

6. a presentation means for presenting information indicating the cost; The information processing device according to claim 3 , further comprising:

7. The costs include at least one of the labor costs of the parties involved in resolving the problem, the loss of productivity caused by the problem, and the cost paid to an outside expert. The information processing device according to claim 3 .

8. The cost includes future additional losses incurred in resolving the problem. The information processing device according to claim 3 .

9. The creating means creates the document including an action plan for solving the problem related to the target log. The information processing device according to claim 1 .

10. The creating means creates the document including an analysis of the psychological tendencies of the parties. The information processing device according to claim 1 .

11. The creating means creates the document including the development plan for the person being observed. The information processing device according to claim 1 .

12. The creating means creates the document including an evidence package to be used in legal proceedings regarding problematic behavior related to the target log. The information processing device according to claim 1 .

13. A computer comprising: receiving an input of a log of the behavior of an observed person, which is any one of a related person in an organization, a robot, and an AI, the log including a designation of one of a plurality of defined types; writing the log to a log database; A step of determining whether a target log among the logs recorded in the log database satisfies a condition for writing; A step of extracting clauses of rules related to the organization that may conflict with the behavior using (a) a generative AI including a large-scale language model that, when a log and the rule are input, outputs clauses in the rule that may conflict with the log, or (b) an AI including a machine learning model trained using training data that includes the rule or a rule similar to the rule, the behavior of the observed person, and identification information of clauses in the rule that may conflict with the behavior (in either case (a) or (b), if the observed person is an AI, an AI other than the AI); a step of creating a document in a predetermined format from the contents of the target log among the logs recorded in the log database, when it is determined that the target log satisfies the conditions for creating a document, the document including the extracted clause; An information processing method comprising:

14. On the computer, receiving an input of a log of the behavior of an observed person, which is any one of a related person in an organization, a robot, and an AI, the log including a designation of one of a plurality of defined types; writing the log to a log database; A step of determining whether a target log among the logs recorded in the log database satisfies a condition for writing; A step of extracting clauses of rules related to the organization that may conflict with the behavior using (a) a generative AI including a large-scale language model that, when a log and the rule are input, outputs clauses in the rule that may conflict with the log, or (b) an AI including a machine learning model trained using training data that includes the rule or a rule similar to the rule, the behavior of the observed person, and identification information of clauses in the rule that may conflict with the behavior (in either case (a) or (b), if the observed person is an AI, an AI other than the AI); a step of creating a document in a predetermined format from the contents of the target log among the logs recorded in the log database, when it is determined that the target log satisfies the conditions for creating a document, the document including the extracted clause; A program to execute.

Images