Human resources data decoding support system, method thereof, and operating program for the human resources data decoding support system

Abstract

We will organize the metrics required by law and the optional metrics disclosed in the guidelines to grasp the overall picture and clarify what needs to be done. Then, we will provide a system that supports the formulation of strategies such as the administrative support functions for disclosing information to securities reporting requirements and integrated reporting requirements, and quantitative understanding and analysis of human resource information, regarding HR analytics functions. [Solution] The personnel data decoding support system includes: an item-specific personnel-related data acquisition unit that acquires personnel-related data stored by distinguishing it into multiple items related to personnel management; a distribution-type fixed phrase storage unit that stores fixed phrases to be presented to the user in advance according to the distribution type of personnel-related data acquired by item; a fixed phrase acquisition unit that acquires fixed phrases stored based on distribution type identification information; and a fixed phrase output unit that outputs the acquired fixed phrases in association with the acquired item-specific personnel-related data.

Description

Technical Field

[0001] The present invention relates to a personnel data interpretation support system for assisting in the visualization of human capital.

Background Art

[0002] Since the release of the International Standard on Human Capital Management (ISO30414) in 2018, the disclosure of human capital information under the US Securities Law has been incorporated into the requirements. In Japan, the disclosure of the "gender wage gap" since 2023, which is one of the measures of "New Capitalism", has been made mandatory, and it has also been made mandatory to describe it in the securities report under the Financial Instruments and Exchange Act for investors. The visualization of human resource investment in enterprises is being promoted. In addition, in the securities report, the expansion regarding corporate governance is being carried out. For example, Corporate Governance Code 3-1(3), 4-2(2), 5-1(3), etc. Under such circumstances, while the cases of disclosing human capital information in integrated reports and sustainability reports are increasing mainly among large enterprises, the original purpose of visualizing human capital is not just visualization. It is required to continuously quantify and analyze the gap between As is (the current situation) and To be (the desired state), and link it to the improvement of corporate value.

[0003] For example, Patent Document 1 aims to provide a technology for obtaining appropriate data analysis results at low cost and in a short time regardless of the level of the user's data analysis skills. A program is described that functions a computer as means for managing the progress state of the dialogue regarding data analysis conducted between the user and the computer, means for specifying the purpose of the user's analysis through the progress of the dialogue, means for analyzing the target data set according to the purpose of the analysis, and means for outputting the result of the analysis of the target data set.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

[0005] Patent Document 1, as an example, displays a line graph showing the relationship between the dependent variable "sales volume" and the independent variable "date," information indicating outliers as important points in the graph, and a suggestion that states, "Sales volume generally fluctuates between approximately 100 and 400 units / day. Outliers occur during the year-end / New Year holidays and Golden Week, with sales volume rising to approximately 600 units / day." This suggestion includes information summarizing the dependent variable and information explaining the outliers.

[0006] In other examples, a box plot showing the relationship between the dependent variable "sales volume" and the independent variable "neighborhood events" is displayed, along with information indicating the difference between the baseline (median sales volume with "none" neighborhood events) and the value of interest (median sales volume with "nationwide" neighborhood events) as important points in the graph, and the suggestion that "when there is a 'nationwide' neighborhood event, sales volume tends to be higher compared to when there is no neighboring event. When comparing medians, the increase in sales volume is 30 units / day." This suggestion also includes information explaining the relationship between the dependent and independent variables.

[0007] However, while visualization using dependent and independent variables, and quantitative identification of the "as is"-"to be" gap are supported, it has not yet reached the point of supporting analysis, consideration of countermeasures, and implementation of policies to address that gap.

[0008] While the content and calculation methods for information disclosure mandated by law are predetermined, voluntary information disclosure is difficult to standardize across companies, even with the publication of the "Human Capital Visualization Guidelines" in August 2022, as disclosure is based on each company's management and human resource strategies. In other words, it is necessary to organize the indicators required by law and the voluntary disclosure indicators disclosed in the guidelines to grasp the overall picture and clarify what needs to be done. Furthermore, there is a need for HR (Human Resources) analytics functions that go beyond mere dashboard displays, providing administrative support functions for information disclosure in securities reporting and integrated reporting, and systems that support strategic formulation such as quantitative understanding and analysis of human resource information. [Means for solving the problem]

[0009] To solve the above problems, the first invention provides a personnel data decoding support system comprising: a personnel data storage unit that stores personnel-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to personnel management; an item-specific personnel data acquisition unit that acquires personnel-related data item by item; a distribution type-specific fixed phrase storage unit that stores fixed phrases to be presented to the user in advance according to distribution type identification information for identifying the n-dimensional distribution type of the personnel-related data acquired item by item; a distribution type identification information judgment rule storage unit that stores distribution type identification information judgment rules, which are rules for acquiring distribution type identification information from the distribution of acquired item-specific personnel-related data; a distribution type identification information acquisition unit that acquires distribution type identification information of acquired item-specific personnel-related data based on the acquired item-specific personnel-related data and the stored distribution type identification information judgment rules; a fixed phrase acquisition unit that acquires fixed phrases stored based on the acquired distribution type identification information; and a fixed phrase output unit that outputs the acquired fixed phrases in association with the acquired item-specific personnel-related data.

[0010] The second invention provides a personnel data decoding support system as described in the first invention, comprising: a target value holding unit that holds target values ​​which are target values ​​for at least some of the aforementioned items; a distribution type target-specific fixed phrase holding unit that holds fixed phrases to be presented to the user in advance according to distribution type identification information and the target values; and a target-dependent fixed phrase acquisition means that acquires the held fixed phrases based on the distribution type identification information and, if there is a target value assigned to the item for which the distribution type identification information is obtained, that target value.

[0011] The third invention provides a personnel data decoding support system as described in the first or second invention, further comprising: a cluster analysis rule holding unit that holds cluster analysis rules for clustering personnel-related data acquired item by item; a cluster analysis unit that clusters the personnel-related data acquired item by item based on the acquired personnel-related data and the held cluster analysis rules; and a cluster analysis result output unit that outputs the cluster analysis results.

[0012] The fourth invention provides a personnel data decoding support system as described in the first or second invention, further comprising: a correlation analysis rule holding unit that holds correlation analysis rules for performing correlation analysis on personnel-related data acquired item by item; a correlation analysis unit that performs correlation analysis on personnel-related data acquired item by item based on the personnel-related data acquired item by item and the held correlation analysis rules; and a correlation analysis result output unit that outputs the correlation analysis results.

[0013] The fifth invention provides a personnel data decoding support system as described in the first or second invention, further comprising: a feature word analysis rule holding unit that holds feature word analysis rules for analyzing natural language data when the personnel-related data acquired item by item includes natural language data; a feature word analysis unit that analyzes the natural language data based on the natural language data and the held feature word analysis rules; and a feature word analysis result output unit that outputs the feature word analysis results.

[0014] The sixth invention provides a human resources data decoding support system described in the first invention, which includes at least one of the following items: compliance and ethics information, cost information, diversity information, leadership information, organizational culture information, health management information, productivity information, recruitment, transfer and turnover information, skills and abilities information, succession planning information, and workforce information.

[0015] The seventh invention provides a personnel data decoding support system described in the first invention, further comprising at least one of the following items: information on the number of personnel and personnel composition, information on overtime hours, information on paid leave utilization rates, information on training, information on skills, information on competencies, and information on goal setting and evaluation.

[0016] The eighth invention provides a personnel data decoding support system as described in the first invention, wherein the distribution type identification information is distribution type identification information of at least one of the following graphs: bar graph, histogram, pie chart, line graph, scatter plot, radar chart, cross-tabulation, and scalar chart.

[0017] The ninth invention provides a method executed by the CPU of a computer-based personnel data decoding support system, comprising: a personnel-related data storage step in which personnel-related data including statistically processable n-dimensional (n≧2) data is stored by classifying it into multiple items related to personnel management; an item-specific personnel-related data acquisition step in which personnel-related data is acquired item by item; a distribution-type-specific fixed phrase storage step in which fixed phrases are held in advance to be presented to the user according to distribution-type identification information for identifying the n-dimensional distribution type of the personnel-related data acquired item by item; a distribution-type identification information judgment rule storage step in which distribution-type identification information judgment rules are rules for acquiring distribution-type identification information from the distribution of acquired item-specific personnel-related data; a distribution-type identification information acquisition step in which distribution-type identification information of acquired item-specific personnel-related data is acquired based on the acquired item-specific personnel-related data and the stored distribution-type identification information judgment rules; a fixed phrase acquisition step in which fixed phrases are held based on the acquired distribution-type identification information; and a fixed phrase output step in which the acquired fixed phrases are output in association with the acquired item-specific personnel-related data.

[0018] The tenth invention provides a method executed by the CPU of a computer-based personnel data decoding support system, comprising: a target value holding step of holding target values ​​which are target values ​​for at least some of the aforementioned items; a distribution type target-specific fixed phrase holding step of holding fixed phrases to be presented to the user in advance according to distribution type identification information and the target values, wherein the fixed phrase acquisition step further comprises a target-dependent fixed phrase acquisition substep of acquiring the held fixed phrase based on the distribution type identification information and, if there is a target value assigned to the item from which the distribution type identification information was obtained, that target value.

[0019] The eleventh invention provides a method executed by the CPU of a computer-based personnel data decoding support system, further comprising: a cluster analysis rule holding step of holding cluster analysis rules for cluster analysis of personnel-related data acquired item by item; a cluster analysis step of cluster analysis of personnel-related data acquired item by item based on the acquired personnel-related data and the held cluster analysis rules; and a cluster analysis result output step of outputting the cluster analysis results.

[0020] The twelfth invention provides a method executed by the CPU of a computer-based personnel data decoding support system, further comprising: a correlation analysis rule holding step of holding correlation analysis rules for correlating personnel-related data acquired item by item; a correlation analysis step of performing correlation analysis on personnel-related data acquired item by item based on the acquired personnel-related data and the held correlation analysis rules; and a correlation analysis result output step of outputting the correlation analysis results.

[0021] The thirteenth invention provides a method executed by the CPU of a computer-based personnel data decoding support system, further comprising: a feature word analysis rule holding step of holding feature word analysis rules for analyzing natural language data when personnel-related data acquired item by item includes natural language data; a feature word analysis step of analyzing the natural language data based on the natural language data and the held feature word analysis rules; and a feature word analysis result output step of outputting the feature word analysis results.

[0022] As a fourteenth invention, there is provided a method executed by a CPU in a personnel data decoding support system which is a computer, wherein the plurality of items include at least one or more of information related to compliance and ethics, information related to costs, information related to diversity, information related to leadership, information related to organizational culture, information related to healthy management, information related to productivity, information related to recruitment, transfer, and departure, information related to skills and capabilities, information related to successor planning, and information related to the workforce, as described in the ninth invention.

[0023] As a fifteenth invention, there is provided a method executed by a CPU in a personnel data decoding support system which is a computer, wherein the plurality of items further include at least one or more of information related to the number of personnel and personnel composition, information related to overtime hours, information related to the rate of paid leave acquisition, information related to training, information related to skills, information related to competencies, and information related to goal setting and evaluation, as described in the fourteenth invention.

[0024] As a sixteenth invention, there is provided a method executed by a CPU in a personnel data decoding support system which is a computer, wherein the distribution type identification information is distribution type identification information of at least one or more graphs among bar graphs, histograms, pie charts, line graphs, scatter plots, radar charts, cross tabulations, and scalar charts, as described in the ninth invention.

[0025] As the seventeenth invention, there is provided an operation program of a personnel data interpretation support system described in a personnel data interpretation support system so as to be readable and executable, which includes a personnel-related data accumulation step of accumulating personnel-related data including statistically processable n-dimensional (n≧2) data separately for a plurality of items related to personnel management, an item-by-item personnel-related data acquisition step of acquiring personnel-related data for each item, a fixed-form sentence holding step for each distribution type of holding fixed-form sentences presented to a user in advance according to distribution type identification information for identifying the n-dimensional distribution type of the personnel-related data acquired for each item, a distribution type identification information determination rule holding step of holding a distribution type identification information determination rule which is a rule for acquiring distribution type identification information from the distribution of the acquired item-by-item personnel-related data, a distribution type identification information acquisition step of acquiring distribution type identification information of the acquired item-by-item personnel-related data based on the acquired item-by-item personnel-related data and the held distribution type identification information determination rule, a fixed-form sentence acquisition step of acquiring the held fixed-form sentences based on the acquired distribution type identification information, and a fixed-form sentence output step of outputting the acquired fixed-form sentences in association with the acquired item-by-item personnel-related data, and causing a computer having the above steps to execute the operation program of the personnel data interpretation support system.

[0026] As the eighteenth invention, there is provided an operation program of a personnel data interpretation support system described in a personnel data interpretation support system so as to be readable and executable, which includes a target value holding step of holding a target value which is a target value for at least some of the above items, and a fixed-form sentence holding step for each distribution type and target of holding fixed-form sentences presented to a user in advance according to distribution type identification information and the target value, and the fixed-form sentence acquisition step further includes a target-dependent fixed-form sentence acquisition sub-step of acquiring the held fixed-form sentences based on the distribution type identification information and, if there is a target value assigned to the item from which the distribution type identification information is obtained, the target value, and causing a computer described in the seventeenth invention to execute the operation program of the personnel data interpretation support system.

[0027] The nineteenth invention provides an operation program for a personnel data decoding support system, which is a computer described in the seventeenth or eighteenth invention, that is written in a readable and executable manner for the personnel data decoding support system, and further comprises: a cluster analysis rule holding step for holding cluster analysis rules for cluster analysis of personnel-related data acquired item by item; a cluster analysis step for cluster analysis of personnel-related data acquired item by item based on the acquired personnel-related data and the held cluster analysis rules; and a cluster analysis result output step for outputting the cluster analysis results.

[0028] The twentieth invention provides an operation program for a personnel data decoding support system, which is a computer described in the seventeenth or eighteenth invention, that is written in a readable and executable manner for a personnel data decoding support system, and further comprises: a correlation analysis rule holding step for holding correlation analysis rules for correlating personnel-related data acquired item by item; a correlation analysis step for performing correlation analysis on personnel-related data acquired item by item based on the acquired personnel-related data and the held correlation analysis rules; and a correlation analysis result output step for outputting the correlation analysis results.

[0029] The 21st invention provides an operation program for a personnel data decoding support system, which is a computer described in the 17th or 18th invention, that is written in a readable and executable manner for a personnel data decoding support system, and further comprises: a feature word analysis rule holding step of holding feature word analysis rules for analyzing natural language data when the personnel-related data acquired item by item includes natural language data; a feature word analysis step of analyzing the natural language data based on the natural language data and the held feature word analysis rules; and a feature word analysis result output step of outputting the feature word analysis results.

[0030] The 22nd invention provides an operating program for a personnel data decoding support system, which is a computer described in the 17th invention, that is executed by the personnel data decoding support system, wherein the plurality of items include at least one of the following: information on compliance and ethics, information on costs, information on diversity, information on leadership, information on organizational culture, information on health management, information on productivity, information on recruitment, transfers and turnover, information on skills and abilities, information on succession planning, and information on the workforce.

[0031] The 23rd invention provides an operation program for a personnel data decoding support system, which is a computer described in the 22nd invention, which is written in a readable and executable manner for the personnel data decoding support system, wherein the plurality of items further include at least one of the following: information on the number of personnel and personnel composition, information on overtime hours, information on paid leave acquisition rates, information on training, information on skills, information on competencies, and information on goal setting and evaluation.

[0032] The 24th invention provides an operation program for a personnel data decoding support system, which is a computer described in the 17th invention, which is an operation program for a personnel data decoding support system that is written in a readable and executable manner for the personnel data decoding support system, wherein the distribution type identification information is distribution type identification information for at least one of the following graphs: bar graph, histogram, pie chart, line graph, scatter plot, radar chart, cross tabulation, and scalar chart. [Effects of the Invention]

[0033] According to the present invention, it is possible to provide a personnel data decoding support system, a method thereof, and an operating program for the personnel data decoding support system that guides the user through the process from observing the OODA loop (situation confirmation and detailed confirmation) to making a situation judgment (automatic dynamic analysis and confirmation of analysis results) by having the user answer questions (teaching the key points of data analysis). [Brief explanation of the drawing]

[0034] [Figure 1] Conceptual diagram showing an example of hardware configuration in an embodiment of the present invention. [Figure 2] Conceptual diagram showing an example of the functional configuration of the personnel data decoding support system of Embodiment 1. [Figure 3] A model diagram showing an example of the overall workflow related to the visualization of human capital information in Embodiment 1 of the present invention. [Figure 4] A diagram illustrating an example of instructions for creating a bar graph in Embodiment 1. [Figure 5] A diagram illustrating an example of histogram instructions in Embodiment 1. [Figure 6] A diagram illustrating an example of instructions for creating a pie chart in Embodiment 1. [Figure 7] A diagram illustrating an example of instructions for creating a line graph in Embodiment 1. [Figure 8] A diagram illustrating an example of instructions for creating a scatter plot in Embodiment 1. [Figure 9] A diagram illustrating an example of radar chart instructions in Embodiment 1. [Figure 10] A diagram illustrating an example of cross-tabulation instructions in Embodiment 1. [Figure 11] A diagram illustrating an example of instructions for a scalar chart in Embodiment 1. [Figure 12] Conceptual diagram showing an example of the hardware configuration of the system in Embodiment 1. [Figure 13] Flowchart showing an example of the processing flow of the system in Embodiment 1 [Figure 14] Conceptual diagram explaining OODA [Figure 15] Conceptual diagram of a system that supports strategy execution by incorporating the OODA loop. [Figure 16] A diagram illustrating an example of a problem area in observing the OODA loop. [Figure 17] Diagram illustrating the operation of the instruction questioning phase in Example 1. [Figure 18] Diagram illustrating the operation of the instruction comment phase in Example 1. [Figure 19] Conceptual diagram showing an example of the functional configuration of the personnel data decoding support system of Embodiment 2. [Figure 20] Conceptual diagram showing an example of the hardware configuration of the system in Embodiment 2. [Figure 21] Flowchart showing an example of the processing flow of the system in Embodiment 2 [Figure 22] This diagram illustrates an example of a problematic area in the situational judgment process of the OODA loop. [Figure 23] Conceptual diagram showing an example of the functional configuration of the personnel data decoding support system of Embodiment 3. [Figure 24] This figure illustrates an example of a cluster analysis method in Embodiment 3. [Figure 25] Conceptual diagram showing an example of the hardware configuration of the system in Embodiment 3. [Figure 26] Flowchart showing an example of the processing flow of the system in Embodiment 3 [Figure 27]Conceptual diagram showing an example of the functional configuration of the personnel data decoding support system of Embodiment 4. [Figure 28] A diagram illustrating an example of the correlation analysis method in Embodiment 4. [Figure 29] Conceptual diagram showing an example of the hardware configuration of the system in Embodiment 4. [Figure 30] Flowchart showing an example of the processing flow of the system in Embodiment 4 [Figure 31] Conceptual diagram showing an example of the functional configuration of the personnel data decoding support system of Embodiment 5. [Figure 32] This figure illustrates an example of a feature word analysis method in Embodiment 5. [Figure 33] Conceptual diagram showing an example of the hardware configuration of the system in Embodiment 5. [Figure 34] Flowchart showing an example of the processing flow of the system in Embodiment 5 [Figure 35] This figure illustrates an example of cluster analysis performed in Example 2. [Figure 36] Figure illustrating an example of correlation analysis performed in Example 2. [Figure 37] Figure illustrating a detailed display example of correlation analysis in Example 2. [Figure 38] This figure illustrates an example of feature word analysis performed in Example 3. [Figure 39] A flowchart illustrating an example of instructions for creating a line graph in Embodiment 1. [Figure 40] A diagram showing some of the disclosure indicators for ISO 30414. [Figure 41] This diagram shows examples of items related to human resource management that are additionally addressed in the present invention.

[0035] <Regarding hardware that may constitute the present invention> Figure 1 shows the hardware configuration to which the present invention is applied. This invention, in principle, utilizes an electronic computer, but can be realized through software, hardware, and the collaboration of software and hardware. Hardware that realizes all or part of the constituent elements of this invention consists of the basic components of a computer, such as a CPU, memory, bus, input / output devices, various peripheral devices, and a user interface. Various peripheral devices include storage devices, internet interfaces, internet devices, displays, keyboards, mice, speakers, cameras, video cameras, televisions, various sensors for monitoring production status in laboratories or factories (flow sensors, temperature sensors, weight sensors, liquid volume sensors, infrared sensors, shipment counting machines, package counting machines, foreign object inspection devices, defective product counting machines, radiation inspection devices, surface condition inspection devices, circuit inspection devices, human presence sensors, worker work status monitoring devices (video, ID, PC workload, etc.)), CD players, DVD players, Blu-ray players, USB memory sticks, USB memory interfaces, removable hard disks, general hard disks, projector devices, SSDs, telephones, fax machines, copiers, printers, movie editing devices, and various sensor devices. Furthermore, this system does not necessarily have to consist of a single enclosure; it may be composed of multiple enclosures connected by communication. The communication may be LAN, WAN, WiFi, Bluetooth®, infrared communication, or ultrasonic communication, and some parts may be installed across national borders. Furthermore, each of the multiple enclosures may be operated by a different entity, or they may be operated by a single entity. The operating entity of the system of this invention may be singular or plural. The invention can also be considered a system that includes terminals used by third parties, and even more terminals used by other third parties. These terminals may also be installed across national borders. In addition to this system and the aforementioned terminals, devices used for registering related information of third parties, devices used for registering related individuals, and devices used for databases to record registration details may be provided. These may be provided within this system, or they may be provided outside this system, and this system may be configured to make this information available.

[0036] As shown in Figure 1, the computer consists of a chipset, CPU, non-volatile memory, main memory, various buses, BIOS, various interfaces such as USB, HDMI®, and LAN, and a real-time clock, all configured on a motherboard. These work in cooperation with the operating system, device drivers (for various interfaces such as USB and HDMI®, and for embedding various devices such as cameras, microphones, speakers or headphones, and displays), and various programs. The various programs and data constituting the present invention are configured to efficiently utilize these hardware resources to perform various processes.

[0037] ≪Chipset≫ A "chipset" is a set of large-scale integrated circuits (LSIs) mounted on a computer's motherboard that integrates the functionality of bridging—that is, the communication function between the CPU's external bus and the standard bus connecting memory and peripheral devices. There are two chipset configurations and one chipset configurations. The northbridge is located closer to the CPU and main memory, while the southbridge is located further away and serves as the interface for relatively slower external I / O.

[0038] (Northbridge) The northbridge includes the CPU interface, memory controller, and graphics interface. The CPU can handle most of the functions traditionally performed by the northbridge. The northbridge connects to the main memory slots via a memory bus and to the graphics card slots via a high-speed graphics bus (AGP, PCI Express).

[0039] (Southbridge) The southbridge connects to the PCI interface (PCI slot) via the PCI bus and handles I / O functions and sound functions for ATA (SATA) interface, USB interface, Ethernet interface, etc. Circuits supporting PS / 2 ports, floppy disk drives, serial ports, parallel ports, and ISA buses, which do not require or are impossible to operate at high speeds, can be separated from the southbridge chip and handled by a separate LSI called a super I / O chip, as this would hinder the speed of the chipset itself. Buses are used to connect the CPU (MPU) to peripherals and various control units. These buses are linked by the chipset. For the memory bus used to connect to main memory, a channel structure may be adopted instead to achieve higher speeds. Either a serial bus or a parallel bus can be used. While a serial bus transfers data one bit at a time, a parallel bus transmits the original data itself or multiple bits extracted from the original data as a single unit, simultaneously over multiple communication channels. A dedicated line for the clock signal runs parallel to the data lines to synchronize data demodulation at the receiving end. It is also used as a bus to connect the CPU (chipset) to external devices, and examples include GPIB, IDE / (parallel)ATA, SCSI, and PCI. Due to limitations in speed increases, in improved versions of PCI such as PCI Express and improved versions of parallel ATA such as Serial ATA, the data lines can be a serial bus.

[0040] ≪CPU≫ The CPU sequentially reads, interprets, and executes instruction sequences called programs located in main memory, outputting signal-based information back to the main memory. The CPU functions as the central hub for performing calculations within the computer. The CPU consists of a CPU core, which is the core of the calculations, and its peripheral parts. Inside the CPU are registers, cache memory, an internal bus connecting the cache memory and the CPU core, a DMA controller, timers, and an interface to the bus connecting to the northbridge. A single CPU (chip) may have multiple CPU cores. In addition to the CPU, processing may also be performed by a graphics interface (GPU) or FPU. The description in this embodiment is for a 2-core type, but it is not limited to this. Furthermore, the CPU can also have a program embedded within it.

[0041] Non-volatile memory (HDD) The basic structure of a hard disk drive consists of a magnetic disk, a magnetic head, and an arm that houses the magnetic head. The external interface can be SATA (formerly ATA). A high-performance controller, such as SCSI, supports communication between hard disk drives. For example, when copying a file to another hard disk drive, the controller can read sectors, transfer them to the other hard disk drive, and write them. This process does not access the host CPU's memory, thus avoiding increased CPU load.

[0042] Main Memory The CPU directly accesses and executes various programs in main memory. Main memory is volatile memory, and DRAM is used. Programs in main memory are loaded from non-volatile memory into main memory upon receiving a program execution command. Subsequently, the CPU executes the program according to various execution commands and procedures within the program.

[0043] Operating System (OS) An operating system is used to manage the resources on a computer for applications to use, manage various device drivers, and manage the computer hardware itself. In small computers, firmware may be used as the operating system.

[0044] ≪BIOS≫ The BIOS is the component that instructs the CPU to start the computer hardware and run the operating system. Most typically, it is the first piece of hardware the CPU reads when it receives a computer startup command. The BIOS contains the addresses of the operating system stored on the disk (non-volatile memory), and the BIOS, deployed by the CPU, sequentially loads the operating system into main memory, bringing it into operation. The BIOS also has a check function that checks for the presence of various devices connected to the bus. The results of the check are saved in main memory and made available to the operating system as appropriate. The BIOS may also be configured to check for external devices. The above applies to all embodiments.

[0045] As shown in Figure 1, the present invention can basically be composed of a general-purpose computer program and various devices. The computer basically operates by loading a program stored in non-volatile memory into main memory, and then executing processing between the main memory, the CPU, and various devices. Communication with devices is performed via an interface connected to a bus line. Possible interfaces include display interfaces, keyboards, and communication buffers. Embodiments of the present invention will be described below with reference to the illustrations.

[0046] <Satisfaction of the applicability of natural laws in this invention> The present invention functions through the collaboration of a computer, communication equipment, and software. Specifically, it relates to a personnel data decoding support system that stores pre-defined texts to be presented to the user according to distribution type identification information for identifying the n-dimensional distribution type of personnel-related data acquired item by item, and outputs the stored pre-defined texts based on the distribution type identification information in association with the acquired personnel-related data by item. Various information and data are exchanged between multiple user terminals, multiple administrator terminals, and multiple information sources via a network using hardware resources. Therefore, from this perspective, judging the present invention based on the matters described in the claims and specification and the common technical knowledge related to those matters, the present invention as a whole utilizes natural laws and falls under the category of a computer software-related invention.

[0047] <The significance of utilizing natural laws as required by patent law> The use of natural laws required under patent law is based on the purpose of the law, and is required to ensure that an invention is industrially useful, from the perspective that the invention must have industrial applicability and contribute to the development of industry. In other words, it requires that the invention be industrially useful, that is, that the effects of the invention declared in the application can be reproduced with a certain degree of certainty by implementing the invention. From this perspective, the applicability of natural laws is interpreted as meaning that the functions exhibited by each of the inventive features (constituent elements of the invention), which are the components of the invention that exert the effects of the invention, are exerted by utilizing natural laws. Furthermore, the effect of the invention only needs to have the potential to provide a certain level of usefulness to the user who uses the invention, and should not be viewed from the perspective of how the user feels or thinks about that usefulness. Therefore, even if the effect that the user obtains from this system is a psychological effect, the effect itself is an event outside the scope of the required applicability of natural laws. [Modes for carrying out the invention]

[0048] Embodiments of the present invention will be described below with reference to the drawings. However, the present invention is not limited in any way to these embodiments, and can be implemented in various ways without departing from its essence.

[0049] The functional configurations of each embodiment described below can be realized as a combination of hardware and software, which will be discussed later. Furthermore, each embodiment described herein can be realized not only as an apparatus or system, but also, in whole or in part, as an operating method. In addition, a part of such an apparatus can be configured as software. Moreover, software products used to run such software on a computer, and recording media on which such products are fixed, are naturally included in the technical scope of each embodiment described herein (and the same applies throughout this specification).

[0050] Embodiment 1 mainly describes claims 1, 6-8, 9, 14-16, 17, and 22-24. Embodiment 2 mainly describes claims 2, 10, and 18. Embodiment 3 mainly describes claims 3, 11, and 19. Embodiment 4 mainly describes claims 4, 12, and 20. Embodiment 5 mainly describes claims 5, 13, and 21.

[0051] <Embodiment 1 (mainly corresponding to claims 1, 6-8, 9, 14-16, 17, 22-24)> <Overview of Embodiment 1> The goal of human capital visualization is to visualize the state of human capital and to invest in human capital while continuously quantitatively understanding and analyzing the targets and indicators set by companies according to their industry, business model, and strategy, thereby aiming to improve corporate value and achieve sustainable growth in the medium to long term. Generally, OODA is known as an exemplary action process suitable for data-driven approaches. OODA is a theory of decision-making and action consisting of Observe, Orient, Decide, and Act. However, even if users unfamiliar with data analysis check the current situation using dashboard graphs, simply displaying graphs and analysis results will not allow them to execute the subsequent OODA loop; it will end up being nothing more than a visualization of "nice graphs." Embodiment 1 provides a human resources data decoding support system that guides the user through the OODA loop from observation (situation confirmation and detailed confirmation) to situation judgment (behavioral analysis and confirmation of analysis results) by having the user answer questions from the human resources data decoding support system (teaching the key points of data analysis).

[0052] <Embodiment 1 Functional Configuration> Figure 2 is a conceptual diagram showing an example of the functional configuration of the personnel data decoding support system 0200 according to Embodiment 1. As shown in the figure, the "personnel data decoding support system" 0200 includes at least a "personnel-related data storage unit" 0201, an "item-specific personnel-related data acquisition unit" 0202, a "standard text storage unit for each distribution type" 0203, a "distribution type identification information receiving unit" 0204, a "standard text acquisition unit" 0205, and a "standard text output unit" 0206.

[0053] <Description of each component in Embodiment 1> (Embodiment 1: Human Resources Data Storage Unit) The personnel-related data storage unit 0201 is configured to store personnel-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to personnel management.

[0054] The various items related to human resource management include, for example, information on compliance and ethics as defined in ISO 30414, information on costs, information on diversity, information on leadership, information on organizational culture, information on health management, information on productivity, information on recruitment, transfers and turnover, information on skills and abilities, information on succession planning, and information on the workforce. (See Figure 40 (excerpt))

[0055] Compliance and ethics information includes the number and types of complaints (on-site complaints, harassment, etc.), the number and types of disciplinary actions taken, the percentage of employees who have completed compliance and ethics training, disputes referred to external parties (labor disputes), and the number, types, and sources of external audit results and resolutions arising from these.

[0056] Cost information includes all labor costs (total personnel costs or gross labor costs), costs associated with external labor, average compensation and compensation rates, total employment costs (salary, social insurance, retirement benefit obligations, human resource development expenses, etc.), cost per employee (recruitment), recruitment costs, and turnover costs.

[0057] Information regarding diversity includes (workforce) diversity (a) age, b) gender, c) disability, d) other diversity measures (nationality, length of service, etc.), and management diversity (age, gender, disability, nationality, etc.).

[0058] Leadership-related information includes leadership credibility, scope of management (number of subordinates per person), and leadership development (percentage of employees who participate in training). Information related to organizational culture includes engagement / employee satisfaction / commitment and retention rates.

[0059] Information related to health management includes lost time (time taken off) due to work-related accidents, the number of work-related accidents, the number of fatalities during work (work-related accidents), and the percentage of employees who participated in training.

[0060] Productivity-related information includes EBIT / revenue / sales / profit per employee and human capital ROI (return on invested capital).

[0061] Information regarding recruitment, transfers, and departures includes: Recruitment (IN): Number of candidates per position, quality per employee (expectations before recruitment and evaluation after recruitment), average time (a) time required to fill a position, b) time required to fill important job openings), employee competence assessment and future potential (talent pool); Transfers (THROUGH): Percentage of positions filled with internal personnel, percentage of positions in important businesses filled internally, percentage of job openings in important businesses, percentage of job openings in important businesses relative to all job openings, internal transfer rate, employee replacement capacity (internal talent supply capacity); Departures (OUT): Resignations, voluntary retirement rate (excluding retirement), percentage of important voluntary retirements (voluntary retirements by employees who are reluctant to leave), reasons for resignation / departure, and the number of resignations / departures by reason.

[0062] Information regarding skills and abilities includes: 1) all costs related to human resource development and training, learning and growth (a) the percentage of employees who participated in training out of the total number of employees over the year, b) the average training hours per employee as defined in the training program (average training time), c) the percentage of employees who participated in various categories of training as defined in the training program, and employee competency rates (competency assessment, ability assessment, etc.).

[0063] Information regarding succession planning includes succession effectiveness, succession coverage, and succession readiness (a) Succession depth: Ready, b) Succession depth: Ready within 1-3 years, c) Succession depth: Ready within 4-5 years.

[0064] Labor force information includes the number of employees, the number of full-time equivalent (FET) employees, temporary labor (a) freelancers (number of contractors, consultants, gig workers, etc.), b) temporary labor (number of employees with employment contracts for a fixed period of time), and absentees.

[0065] In addition to these, it is also possible to include information on themes that are strongly requested by users, such as the number and composition of personnel, information on overtime hours, information on paid leave utilization rates, information on training, information on skills, information on competencies, and information on goal setting and evaluation (see Figure 41).

[0066] (Embodiment 1: Itemized Personnel-Related Data Acquisition Unit) The item-specific personnel-related data acquisition unit 0202 is configured to acquire personnel-related data item by item from the personnel-related data storage unit. The personnel-related data may be distributed across multiple personnel-related data storage units. The Item-Specific HR Data Acquisition Unit acquires HR-related data corresponding to the items for which human capital information has been visualized. The following provides a supplementary explanation of an example workflow for visualizing human capital information, referring to Figure 3. In Figure 3, the "aggregated data" used in "(e) Information Visualization Task" corresponds to "personnel-related data corresponding to the items for which human capital information has been visualized."

[0067] (a) Basic data creation work Based on personnel-related data acquired by category, "point-in-time data" is created and managed historically. This becomes the data used for aggregation. The functional requirements for the system are: • Users can specify the information they want to aggregate. • The ability to create point-in-time information using various reference dates to match the user's aggregation characteristics. (Example) Number of employees: as of the end of the fiscal year, Paid leave taken: April 1st to March 31st, Employment of people with disabilities: as of June 1st Includes.

[0068] (i) Data correction services The aggregated data is corrected as needed by manually entering data or importing previously managed data from previous handovers. The functional requirements for the system are: • The ability to process and save data as of the reference date. • Ability to carry over past processing data • The ability to import (store) information managed externally. • The ability to import data from manual input or spreadsheet tools, and to add, modify, and delete basic data. Includes.

[0069] (c) Aggregation and calculation tasks Calculations are performed only for the most recent year, and stored figures are used for past years. This becomes the aggregated data. The functional requirements for the system are: • The calculation formula and aggregation unit can be specified to match the user's aggregation characteristics. • Only the most recent year can be aggregated and calculated using a defined logic. • Ability to register and modify the latest year's aggregate calculation results. • Ability to correct calculation results for past fiscal years. • The aggregated data can be displayed on the screen in a layout defined by the user. • The aggregated data can be output to a summary table in any spreadsheet tool defined by the customer. Includes.

[0070] (e) Offline work using input / output with spreadsheet tools The HR department can download the data for aggregation to their work PC and manually correct the values ​​as needed. The corrected data can then be uploaded and saved as aggregated data. The functional requirements for the system are: • The ability to output aggregated calculation results in a spreadsheet tool using a user-defined layout. • The ability to import numerical values ​​directly into the output spreadsheet tool and manage them within the system. Includes.

[0071] <(O) Information visualization work> The aggregated data is graphed and displayed on a dashboard to visualize human capital information. The functional requirements for the system are: • The aggregated calculation results can be displayed in graphs and dashboards using a layout defined by the customer. • Ability to break down information • It can be made available not only to the HR department, but also to management and executives. • The ability to explain the points and analytical methods that the displayed graph represents. Includes.

[0072] (Embodiment 1: Standard text storage unit for each distribution type) The standard text storage unit 0203 for each distribution type is configured to store standard text for users for each distribution type.

[0073] In one embodiment, the standard phrases can be stored in a standard phrase storage unit for each distribution type, in a form that is referenced for each distribution type from the instructions corresponding to the type of graph. In another configuration, the standard phrases can be stored in a standard phrase storage unit for each distribution type, embedded in the comment section of the instruction corresponding to the graph type, for each distribution type.

[0074] Instructions are information that describes the process flow for guiding users through the OODA loop, from observation (situation confirmation and detailed confirmation) to situation judgment (behavioral analysis and confirmation of analysis results). Figure 4 shows an example of instructions for a bar graph. Figure 5 shows an example of instructions for a histogram. Figure 6 shows an example of instructions for a pie chart. Figure 7 shows an example of instructions for a line graph. Figure 8 shows an example of instructions for a scatter plot. Figure 9 shows an example of instructions for a radar chart. Figure 10 shows an example of instructions for a cross-tabulation. Figure 11 shows an example of instructions for a scalar chart.

[0075] "Distribution type" refers to the type of distribution when human capital information is visualized on the dashboard. Examples include "achievement type" and "time series type" in bar graphs, "isolated island type" and "two-peak type" in histograms, "repeating type" and "upward type" in line graphs, and "concentrated type" in scatter plots.

[0076] A standard phrase is, for example, if the distribution type is an "isolated island" type in the histogram, "For data that is significantly different from what is needed, there are often special circumstances that require individual attention. Let's examine the data in question and identify the cause." Let's perform a correlation analysis and investigate the items that have a strong relationship with the [X-axis items]. • Data that is significantly different often has special circumstances and requires individual attention. Check the data in question and identify the cause. Let's perform a correlation analysis and investigate the items that have a strong relationship with the [X-axis items]. This is the sentence. Furthermore, if the distribution type is a "bimodal" type in the histogram, "There may be multiple underlying causes. Let's perform cluster analysis and investigate the trends for each "peak" in the [X-axis items]." Let's perform a correlation analysis and investigate the items that have a strong relationship with the [X-axis items]. This is the sentence.

[0077] A standard phrase is, for example, if the distribution type is "repeating type" in a line graph, "The same trend is repeated every [period]. If you have an idea of ​​the cause and it is possible to improve it, let's consider measures to address this periodicity." Let's perform a correlation analysis on the time-series data and investigate the items that are strongly related to the [Y-axis items]. This is the sentence. Furthermore, if the distribution type is "upward" in the line graph, "• The trend is improving. If you need to improve quickly, perform a correlation analysis of the time-series data and investigate the items that are strongly related to the [Y-axis items]." This is the sentence.

[0078] A standard phrase is, for example, if the distribution type is "concentrated" in a scatter plot, There may be multiple underlying causes. Let's perform cluster analysis to investigate the trends within the group. Let's perform a correlation analysis and investigate the items that are strongly related to the [Y-axis items]. This is the sentence.

[0079] (Embodiment 1: Distribution Type Identification Information Receiving Unit) The distribution type identification information receiving unit 0204 is configured to receive distribution type identification information from the distribution of the acquired personnel-related data by item. For example, the histogram instructions in Figure 5 contain standardized text (comments) corresponding to four distribution types. Each of the four distribution types is also assigned distribution type identification information. For instance, a distribution that combines "isolated island" and "two-peak" types is assigned the distribution type identification information "11," a distribution consisting only of "isolated island" types is assigned the distribution type identification information "10," a distribution consisting only of "two-peak" types is assigned the distribution type identification information "01," and a distribution that is neither "isolated island" nor "two-peak" is assigned the distribution type identification information "00."

[0080] The distribution type identification information receiving unit receives distribution type identification information indicating which distribution type the graph displayed on the dashboard belongs to. For example, in the histogram instructions in Figure 5, the distribution type is identified based on the user's answers to question 2 and question 3, and distribution type identification information is received accordingly. As another example, the instructions for the line graph in Figure 7 contain standardized text (comments) corresponding to three distribution types. Each of the three distribution types is assigned distribution type identification information. For example, a "repeating" distribution is assigned "1X" as its distribution type identification information, an "ascending" distribution is assigned "01" as its distribution type identification information, and a distribution that is neither "repeating" nor "ascending" is assigned "00" as its distribution type identification information.

[0081] For example, in the instructions for the line graph in Figure 7, the answer to question 2, "Is there periodicity?", is the result of detecting periodicity using the EPA method (for example, calculating the standard deviations of seasonal variation (S) and irregular variation (I), and determining that there is periodicity if "standard deviation of S > standard deviation of I"). The answer to question 3, "Is there an improving trend?", is the result of automatically analyzing the trend (TC) using the EPA method (for example, calculating the derivative and second derivative of the most recent period (e.g., the most recent 10 data points), and determining that there is an improving trend if all derivatives are positive. If the derivative of the first half of the data is negative, and the data in question is less than half of the total data, and the second derivative is positive, then it is determined that there is an improving trend. In other cases, it is determined that there is no improving trend). The distribution type may be identified based on these automatic judgments, and distribution type identification information may be accepted accordingly. Here, the instructions for the line graph in Figure 7 can be represented as a flowchart, as shown in Figure 39.

[0082] (Embodiment 1: Standard text acquisition unit) The standard text acquisition unit 0205 is configured to acquire standard texts that are held based on the acquired distribution type identification information. For example, based on the acquired distribution type identification information, it is possible to retrieve the boilerplate text stored in the boilerplate text storage unit for each distribution type, which is referenced (or embedded) in the comment section of the instruction corresponding to the graph type.

[0083] (Embodiment 1: Standard text output unit) The standard text output unit 0207 is configured to output the acquired standard text in association with the acquired item-specific personnel-related data.

[0084] Here, the boilerplate text can include parameter specifications for string substitution. Parameter specifications are made by enclosing the parameter name in "[" and "]". In the example of the standard text provided in the "Standard Text Storage Unit for Each Distribution Type," the [X-axis item], [Y-axis item], and [period] correspond to parameter specifications. The standard phrase output unit 0207, when it receives a standard phrase that includes parameter specifications, such as the standard phrase for a distribution type that combines "isolated island type" and "two-peak type" as shown in Figure 5, "There may be multiple underlying causes. Let's perform cluster analysis and investigate the trends for each 'peak' of [X-axis item]," associates it with the acquired personnel-related data by item, and outputs "There may be multiple underlying causes. Let's perform cluster analysis and investigate the trends for each 'peak' of [average overtime hours]." Furthermore, in the "repeating" standard phrase in Figure 7, "The same trend is repeated every [period]. If you have an idea of ​​the cause and it is possible to improve it, let's consider measures to address this periodicity," for example, "[period]" is replaced with a period value detected using the EPA method in the acquired personnel-related data by item, such as "[6 months]," and the output becomes "The same trend is repeated every [6 months]. If you have an idea of ​​the cause and it is possible to improve it, let's consider measures to address this periodicity."

[0085] <Embodiment 1: Human Resources Data Decoding Support System: Hardware Configuration> The hardware configuration of the personnel data decoding support system in Embodiment 1 will be explained with reference to a diagram.

[0086] Figure 12 shows the hardware configuration of the personnel data decoding support system in Embodiment 1. As shown in this figure, the information provision system in Embodiment 1 includes a "CPU (Central Processing Unit)" 1201 that performs various calculations, and a "main memory" 1202. It also includes a "non-volatile memory" 1203 that holds predetermined information, and a "network I / F (interface)" 1204 that sends and receives information with multiple user terminals 1206, multiple administrator terminals 1207, personnel system data 1208, personnel system data 1209, and other personnel-related data 1210. These are interconnected by data communication paths such as a "bus" 1205 to send and receive information and perform processing.

[0087] Main memory is read by the CPU to execute programs that perform various processing tasks, and at the same time, it provides a work area that serves as the working area for those programs. In addition, both main memory and non-volatile memory are assigned multiple addresses, and programs executed by the CPU can exchange data with each other and perform processing by identifying and accessing these addresses. In Embodiment 1, the programs stored in "main memory" include a personnel-related data storage program, a personnel-related data acquisition program by item, a standard text retention program for each distribution type, a distribution type identification information reception program, a standard text acquisition program, and a standard text output program. Furthermore, "main memory" and "non-volatile memory" store personnel-related data, personnel-related data by item, standard phrases, and distribution type identification information.

[0088] The CPU executes the personnel-related data storage program stored in main memory to store personnel-related data in main memory and non-volatile memory. It also executes the item-specific personnel-related data acquisition program stored in main memory to store item-specific personnel-related data in main memory and non-volatile memory. Furthermore, it executes the distribution type-specific boilerplate text retention program stored in main memory to store boilerplate text in main memory and non-volatile memory. It also executes the boilerplate text acquisition program stored in main memory to store boilerplate text in main memory and non-volatile memory based on distribution type identification information. Finally, it executes the boilerplate text output program stored in main memory to output boilerplate text associated with item-specific personnel-related data.

[0089] <Embodiment 1: Human Resources Data Decoding Support System: Processing Flow> Figure 13 shows the processing flow when using the personnel data decoding support system in Embodiment 1. As shown in the figure, the processing method consists of a personnel-related data storage step S1301, an item-specific personnel data acquisition step S1302, a standard text holding unit step S1303 for each distribution type, a distribution type identification information reception step S1304, a standard text acquisition step S1305, and a standard text output step S1306.

[0090] These processing methods are executed by a personnel data decoding support system comprising: a personnel data storage unit that stores personnel-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to personnel management; an item-specific personnel data acquisition unit that acquires personnel-related data item by item; a distribution type-specific fixed phrase storage unit that holds fixed phrases to be presented to the user in advance, according to distribution type identification information for identifying the n-dimensional distribution type of the personnel-related data acquired item by item; a distribution type identification information receiving unit that receives distribution type identification information from the distribution of the acquired item-specific personnel-related data; a fixed phrase acquisition unit that acquires fixed phrases held based on the acquired distribution type identification information; and a fixed phrase output unit that outputs the acquired fixed phrases in association with the acquired item-specific personnel-related data.

[0091] The "HR-related data accumulation step" S1301 is the stage in which HR-related data, including statistically processable n-dimensional (n≧2) data, is accumulated by categorizing it into multiple items related to HR management.

[0092] "Item-specific personnel-related data acquisition step" S1302 is the stage in which personnel-related data is acquired item by item.

[0093] The "Step to retain standard text for each distribution type" S1303 is a step in which standard text is retained in advance to be presented to the user, according to the distribution type identification information used to identify the n-dimensional distribution type of personnel-related data acquired for each item.

[0094] The "Distribution Type Identification Information Reception Step" S1304 is the stage where distribution type identification information is received from the acquired item-specific personnel-related data distribution.

[0095] The "standard text acquisition step" S1305 is the stage in which standard texts are acquired based on the acquired distribution type identification information.

[0096] The "Standard Text Output Step" S1306 is the stage in which the acquired standard text is output in association with the acquired personnel-related data by item.

[0097] <Embodiment 2 (mainly corresponding to claims 2, 10, and 18)> <Overview of Embodiment 2> Embodiment 2 is based on Embodiment 1, and includes a target value holding unit that holds target values ​​for at least some of the aforementioned items, a distribution type target-specific standard phrase holding unit that holds standard phrases to be presented to the user in advance according to the distribution type identification information and the target value, and a target-dependent standard phrase acquisition means that acquires the held standard phrases based on the distribution type identification information and, if there is a target value assigned to the item for which this distribution type identification information is obtained, that target value. This makes it possible to quantitatively grasp the gap between As is (current state) and To be (desired state). Furthermore, to achieve this, we provide a method executed by the CPU in the information provision system, which is a computer, and an operating program for the information provision system written in a readable and executable format for the information provision system, which is a computer.Hereafter, the same functional configuration, hardware configuration, and processing flow as in Embodiment 1 will be omitted from explanation as appropriate.

[0098] <Embodiment 2 Functional Configuration> Figure 19 is a conceptual diagram showing an example of the functional configuration of the personnel data decoding support system 1900 according to Embodiment 2. As shown in the figure, the "personnel data decoding support system" 1900 includes at least a "personnel-related data storage unit" 1901, an "item-specific personnel-related data acquisition unit" 1902, a "standard text holding unit for each distribution type" 1903, a "distribution type identification information receiving unit" 1904, a "standard text acquisition unit" 1905, a "target-dependent standard text acquisition means" 1905A, a "standard text output unit" 1906, a "target value holding unit" 1907, and a "standard text holding unit for each distribution type target" 1908.

[0099] <Description of each component in Embodiment 2> (Embodiment 2: Target Value Holding Unit) The target value holding unit 1907 is configured to hold target values, which are target values ​​for at least some of the items. The term "item" is the same as described in the explanation for the personnel-related data storage unit in Embodiment 1, so a repeated explanation will be omitted. For example, if the "item" is "average overtime hours," then the value "24 hours" corresponds to the target value.

[0100] (Embodiment 2: Standard text storage unit for each distribution type target) The standard text storage unit 1908 for each distribution type and target is configured to store standard texts to be presented to the user in advance, according to the distribution type identification information and the target value. The term "distribution type" is the same as described in the standard text storage unit for each distribution type in Embodiment 1, so a repeated explanation will be omitted. The "distribution type identification information" is the same as described in the distribution type identification information receiving unit of Embodiment 1, so a repeated explanation will be omitted.

[0101] In one embodiment, the standard phrases can be stored in a standard phrase storage unit for each distribution type and target, in a form that is referenced according to the instructions corresponding to the type of graph, the distribution type identification information, and the target value. In another configuration, the standard phrases can be stored in a standard phrase storage unit for each distribution type, embedded in the comment field of the instruction corresponding to the graph type, along with the distribution type and target value.

[0102] A standard phrase is, for example, in a "achievement-type" bar graph, the target value is set for the [Y-axis item]. "—With a target value—Let's perform characteristic word analysis and investigate the causes by comparing it with the group that has achieved the target value (or the group used as a benchmark)." --No target value-- Let's perform a feature word analysis and investigate the cause by comparing categories with high and low [Y-axis items]. This is the sentence. Here, "--Target Value Included--" and "--Target Value Not Included--" correspond to tags in structured documents, and depending on "--Target Value Included--" and "--Target Value Not Included--", ​​the following standard phrases are stored in the standard phrase storage unit for each distribution type and target: "Perform feature word analysis and investigate the cause by comparing with the group that has achieved the target value (or the group used as a benchmark)." and "Perform feature word analysis and investigate the cause by comparing the categories with high and low [Y-axis items]."

[0103] A standard phrase is, for example, if the distribution type is "upward" in a line graph, the target value is set for the [Y-axis item]. "--Target value exists---There is an improving trend. If this trend continues, the target is expected to be reached around [Expected target achievement date]. If you need to improve quickly, perform a correlation analysis of the time series data and investigate the items that are strongly related to [Y-axis items]." --No target value-- The trend is improving. If you need to improve quickly, perform a correlation analysis of the time-series data and investigate the items that are strongly related to the [Y-axis items]. This is the sentence. Then, depending on whether it is "--Target Value Included--" or "--Target Value Not Included--", ​​the following standard phrases are stored in the standard phrase storage unit for each distribution type and target: "--There is an improving trend. If this trend continues, the target is expected to be reached around [Target Achievement Date]. If you need to improve quickly, perform a correlation analysis of the time series data and investigate items that are strongly related to [Y-axis items]." and "--There is an improving trend. If you need to improve quickly, perform a correlation analysis of the time series data and investigate items that are strongly related to [Y-axis items]."

[0104] (Embodiment 2: Means for obtaining target-dependent boilerplate text) The target-dependent boilerplate text acquisition means 1905A is configured to acquire a stored boilerplate text based on distribution type identification information and, if such information is obtained, the target value assigned to the item. For example, based on the acquired distribution type identification information and, if there is a target value assigned to the item from which this distribution type identification information was obtained, the standard text stored in the standard text storage unit for each distribution type target, which is referenced from (or embedded in) the comment field of the instruction corresponding to the graph type, can be retrieved.

[0105] <Embodiment 2: Human Resources Data Decoding Support System: Hardware Configuration> The hardware configuration of the personnel data decoding support system in Embodiment 2 will be explained with reference to a diagram.

[0106] Figure 20 shows the hardware configuration of the personnel data decoding support system in Embodiment 2. As shown in this figure, the information provision system in this embodiment includes a "CPU (Central Processing Unit)" 2001 that performs various calculations and a "main memory" 2002. It also includes a "non-volatile memory" 2003 that holds predetermined information, and a "network I / F (interface)" 2004 that sends and receives information with multiple user terminals 2006, multiple administrator terminals 2007, personnel system data 2008, personnel system data 2009, and other personnel-related data 2010.

[0107] Main memory is read by the CPU to execute programs that perform various processing tasks, and at the same time, it provides a work area that serves as the working area for those programs. In addition, both main memory and non-volatile memory are assigned multiple addresses, and programs executed by the CPU can exchange data with each other and perform processing by identifying and accessing these addresses. In Embodiment 2, the programs stored in "main memory" include a personnel-related data storage program, a personnel-related data acquisition program by item, a standard text retention program for each distribution type, a distribution type identification information reception program, a standard text acquisition program, a standard text output program, a target value retention program, a standard text retention program for each distribution type target, and a target-dependent standard text acquisition subprogram. Furthermore, the "main memory" and "non-volatile memory" store personnel-related data, personnel-related data by item, standard phrases, distribution type identification information, target values, and standard phrases for each distribution type target.

[0108] The CPU executes the personnel-related data storage program stored in main memory to store personnel-related data in main memory and non-volatile memory. It also executes the item-specific personnel-related data acquisition program stored in main memory to store item-specific personnel-related data in main memory and non-volatile memory. Furthermore, it executes the distribution-type-specific boilerplate text retention program stored in main memory to store boilerplate text in main memory and non-volatile memory. It also executes the boilerplate text acquisition program stored in main memory to store boilerplate text in main memory and non-volatile memory based on distribution-type identification information. Furthermore, it executes the boilerplate text output program stored in main memory to output boilerplate text associated with item-specific personnel-related data. Furthermore, it executes the target value retention program stored in main memory to store target values ​​in main memory and non-volatile memory. Furthermore, it executes the distribution-type-specific target-specific boilerplate text retention program stored in main memory to store distribution-type-specific target-specific boilerplate text in main memory and non-volatile memory. Furthermore, the program executes the target-dependent boilerplate text acquisition subprogram stored in "main memory" to obtain the boilerplate text.

[0109] <Embodiment 2: Human Resources Data Decoding Support System: Processing Flow> Figure 21 is a diagram showing the processing flow when using the personnel data decoding support system in Embodiment 2. As shown in the figure, the processing method consists of a personnel-related data storage step S2101, an item-specific personnel data acquisition step S2102, a standard text storage unit step S2103 for each distribution type, a distribution type identification information reception step S2104, a standard text acquisition step S2105, a standard text output step S2106, a target value storage step S2107, a standard text storage step S2108 for each distribution type target, and a target-dependent standard text acquisition substep S2105A.

[0110] These processing methods include: a personnel-related data storage unit that stores personnel-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to personnel management; an item-specific personnel-related data acquisition unit that acquires personnel-related data item by item; a distribution-type fixed phrase storage unit that holds fixed phrases to be presented to the user in advance, according to distribution-type identification information for identifying the n-dimensional distribution type of the personnel-related data acquired item by item; a distribution-type identification information receiving unit that receives distribution-type identification information from the distribution of the acquired item-specific personnel-related data; and a fixed phrase storage unit that acquires fixed phrases held based on the acquired distribution-type identification information. The system is executed by a personnel data decoding support system having a text acquisition unit, a standard text output unit that outputs the acquired standard text in association with the acquired personnel-related data for each item, a target value holding unit that holds target values ​​which are target values ​​for at least some of the aforementioned items, a distribution type target-specific standard text holding unit that holds standard text to be presented to the user in advance according to the distribution type identification information and the target value, and the standard text acquisition unit acquires the held standard text based on the distribution type identification information and, if there is a target value assigned to the item from which the distribution type identification information was obtained, that target value.

[0111] The "HR-related data accumulation step" S2101 is the stage in which HR-related data, including statistically processable n-dimensional (n≧2) data, is accumulated by categorizing it into multiple items related to HR management.

[0112] "Item-specific personnel-related data acquisition step" S2102 refers to the stage of acquiring personnel-related data item by item.

[0113] The "Step to retain standard text for each distribution type" S2103 is a step in which standard text is retained in advance to be presented to the user, according to the distribution type identification information used to identify the n-dimensional distribution type of personnel-related data acquired for each item.

[0114] The "Distribution Type Identification Information Reception Step" S2104 is the stage where distribution type identification information is received from the acquired item-specific personnel-related data distribution.

[0115] The "standard text acquisition step" S2105 is the stage in which standard texts are acquired based on the acquired distribution type identification information.

[0116] The "Target-Dependent Boolean Text Acquisition Substep" S2105A is a dependent step that acquires stored boilerplate text based on distribution type identification information and, if applicable, the target value assigned to the item from which this distribution type identification information was obtained.

[0117] The "Standard Text Output Step" S2106 is the stage in which the acquired standard text is output in association with the acquired personnel-related data by item.

[0118] The "target value holding step" S2107 is a step in which a target value is held for at least some of the aforementioned items.

[0119] The "Step to retain a fixed phrase for each distribution type target" S2108 is a step in which a fixed phrase to be presented to the user in advance is retained according to the distribution type identification information and the target value.

[0120] <Example 1> The following describes an example (Example 1) based on Embodiment 1 and Embodiment 2. Figure 14 is a conceptual diagram of OODA, an exemplary action process suitable for data-driven approaches. As illustrated, OODA is a theory of decision-making and action consisting of Observe, Orient, Decide, and Act. OODA is a theory originally proposed for military operations where the battle situation changes moment by moment, but it is now being proposed to be applicable to the fields of politics and business as well. Compared with the well-known action process PDCA, the characteristic of OODA is that action begins not with "planning," but with "observation" to accurately understand the current situation. OODA prevents plans from being formulated with predetermined conclusions or with preconceived notions about the current situation. The OODA loop involves grasping the current situation as it is, deductively deriving strategies from that, and then implementing them precisely to improve the situation.

[0121] Next, we will explain a system that incorporates this OODA loop to support strategy execution, referring to Figure 15. Figure 15 is a conceptual diagram of a system that incorporates the OODA loop to support strategy execution.

[0122] <Observe - Observe the market, competitors, etc., and gather information> As explained with reference to Figure 2, the item-specific personnel-related data acquisition unit 0202 is configured to acquire personnel-related data item by item from the personnel-related data storage unit. (a) Basic data creation work Based on personnel-related data acquired by category, "point-in-time data" is created and managed historically. This becomes the data used for aggregation. (i) Data correction services The aggregated data is corrected as needed by manually entering data or importing previously managed data from previous handovers. (c) Aggregation and calculation tasks Calculations are performed only for the most recent year, and stored figures are used for past years. This becomes the aggregated data. (e) Offline work using input / output with spreadsheet tools The HR department can download the data for aggregation to their work PC and manually correct the values ​​as needed. The corrected data can then be uploaded and saved as aggregated data. <(O) Information visualization work> The aggregated data is graphed and displayed on a dashboard to visualize human capital information. The visualization of HR-related data is performed through the process described above.

[0123] However, even if users unfamiliar with data analysis check the current situation using dashboards, simply displaying graphs and analysis results is often insufficient to execute the OODA loop shown in Figures 3 and 15, and the process ends up being merely a "visualization of human capital information." This is likely because, as shown in Figure 16, the key points to look at when assessing the current situation are not understood. As a solution, the present invention provides an environment in which HR personnel can execute (or continue) the OODA loop without ending up with just a "nice graph" by following instructions from an interactive HR data interpretation support system. The following describes in detail the feature of the present invention, which is "the ability to present the points and analysis methods that the displayed graph represents."

[0124] The HR data interpretation support system according to Example 1 is configured to support "points to look at for checking the current situation" based on "instructions" (see Figures 4 to 11) provided for each type of graph displayed on the dashboard.

[0125] Typical graphs displayed on the dashboard include bar graphs, histograms, pie charts, line graphs, scatter plots, radar charts, cross-tabulations, and scalar charts. Here, we will explain in detail using histograms as an example. Although the standard phrases differ for each type of distribution in other graphs, the flow of the conversation is similar, so detailed explanations will be omitted. It goes without saying that this can also be applied to graphs other than bar graphs, histograms, pie charts, line graphs, scatter plots, radar charts, cross-tabulations, and scalar charts by preparing appropriate standard phrases for each type of distribution.

[0126] Standard phrases for each classification type are stored in a tabular format in the comment section for each distribution type, as shown in Figures 4 to 11, for example, for bar graphs, histograms, pie charts, line graphs, scatter plots, radar charts, cross-tabulations, and scalar charts. For example, referring to the histogram in Figure 5, if a case shows a combination of "isolated island" and "two-peak" patterns, the following standard phrases are maintained: "• Data that are significantly separated often have special circumstances and require individual attention. Let's examine the target data and identify the cause. • Multiple causes may be underlying. Let's perform cluster analysis and investigate the trends for each 'peak' of the [X-axis items]. • Let's perform correlation analysis and investigate the items that have a strong relationship with the [X-axis items]."

[0127] Figure 17 shows the distribution of average overtime hours visualized as a histogram and displayed on the left side of the dashboard as human capital information. For example, if the standard text storage unit 0203 for each distribution type retrieves the instruction for "histogram," and the HR data decoding support system asks question 1, and a target value (e.g., 24 hours) is set for average overtime hours, the system will detect that there are employees whose average overtime hours exceed 24 hours and will prompt the user with "[Average overtime hours] exceed the target value. Let's analyze and develop improvement measures." If the HR department staff member selects "Analyze this problem," they will have answered "Yes" to question 1, and a dialogue with the HR data decoding support system will begin on the right side of the screen. If no specific target values ​​have been set, the system will ask, "Is the distribution (shape of the graph) in this graph different from what you expected, or are there any points that concern you?" If the HR department representative selects "Analyze this issue," it will be considered that they have answered "Yes" to question 1, and a dialogue with the system will begin on the right side of the screen.

[0128] First, the distribution type identification information receiving unit 0204 receives distribution type identification information based on the answers received to the questions stored in the instructions corresponding to the type of graph displayed on the left side of the screen (histogram in Figure 17). The dialogue proceeds according to the predefined texts (user guidance, questions, and comments) stored by the distribution type-specific predefined text storage unit 0203. As the second question, I would guide the student by illustrating a typical example, asking, "Are there any data points that are far apart in this graph?" If the HR department representative selects "Yes" following the instructions, then, as shown in Figure 18, they will be prompted with Question 3, "Does this graph have multiple peaks?", illustrating a typical example. In this process, HR personnel can learn that, in the case of a histogram, the key points to observe (understand the current situation) are whether there are data points that are far apart and whether there are multiple "peaks".

[0129] Then, if the answer to both question 2 and question 3 is "yes", the standard text output unit 0206 outputs the standard text stored in the histogram instructions based on the distribution type identification information "11" received by the distribution type identification information receiving unit 0205. For example, the dialogue screen will show: "The content of your response" There is data that is separated. There are multiple mountains. This graph may have multiple underlying causes. Please analyze the data to identify the cause. correspondence (1) Data that is significantly different often has special circumstances and requires individual attention. Check the data in question and identify the cause. (2) There may be multiple underlying causes. Let's perform cluster analysis and investigate the trends in each "peak" of [average overtime hours]. (3) Let's conduct a correlation analysis and investigate the items that have a strong relationship with [average overtime hours]. The display of guidance and comments allows HR personnel to learn to proceed with situational judgment (automatic analysis and prediction of factors related to areas for improvement) based on observation (understanding the current situation), that is, to execute the subsequent OODA loop.

[0130] <Summary> Based on the above, the present invention provides a personnel data decoding support system that can output standard phrases by obtaining standard phrases based on distribution type identification information for identifying the n-dimensional distribution type of personnel-related data obtained item by item, and by associating the obtained standard phrases with the acquired personnel-related data item by item. Furthermore, the present invention provides a personnel data decoding support system that can obtain predefined texts to be presented to the user in advance based on distribution type identification information and target values ​​which are target values ​​for each item, and output predefined texts by associating the obtained predefined texts with the acquired personnel-related data for each item.

[0131] <Embodiment 3 (mainly corresponding to claims 3, 11, and 19)> <Overview of Embodiment 3> Embodiment 3 is based on Embodiment 1 and includes a function to guide situational judgment based on observation of the OODA loop. As shown in Figure 22, even if the points to look at during the current situation confirmation are understood and the situational judgment can be made, it often occurs that the subsequent OODA loop cannot be executed because "it is unclear how to perform the analysis" or "it is unclear what points (key points) to look at in the analysis results." Embodiment 3 can resolve this situation and provides a function that enables the analysis of the gap between As is (current state) and To be (desired state). Furthermore, to achieve this, we provide a method executed by the CPU in the information provision system, which is a computer, and an operating program for the information provision system written in a readable and executable format for the information provision system, which is a computer.Hereafter, the same functional configuration, hardware configuration, and processing flow as in Embodiment 1 will be omitted from explanation as appropriate.

[0132] <Embodiment 3 Functional Configuration> Figure 23 is a conceptual diagram showing an example of the functional configuration of the personnel data decoding support system 2300 according to Embodiment 3. As shown in the figure, the "personnel data decoding support system" 2300 includes at least a "personnel-related data storage unit" 2301, an "item-specific personnel-related data acquisition unit" 2302, a "standard text storage unit for each distribution type" 2303, a "distribution type identification information receiving unit" 2304, a "standard text acquisition unit" 2305, a "standard text output unit" 2306, a "cluster analysis rule storage unit" 2307, a "cluster analysis unit" 2308, and a "cluster analysis result output unit" 2309.

[0133] <Description of each component in Embodiment 3> (Embodiment 3: Cluster Analysis Rule Holding Unit) In Embodiment 3, cluster analysis is provided as a data analysis method for classifying data and presenting key insights to the user based on the analysis of personnel-related data. Cluster analysis groups similar data together and classifies them into multiple clusters. By grouping employees with similar attributes into one cluster, it becomes easier to perform analysis based on employee types. The cluster analysis rule holding unit 2307 is configured to hold cluster analysis rules for analyzing personnel-related data acquired item by item. Cluster analysis rules involve, for example, "clustering similar data." Personnel information often contains data that is difficult to interpret based on single items alone, such as personality assessment results or aptitude test results. There is a need for a function that performs clustering and secondary analysis on such data.

[0134] (Embodiment 3: Cluster Analysis Unit) The cluster analysis unit 2308 is configured to analyze the personnel-related data acquired item by item based on the acquired personnel-related data and the stored cluster analysis rules. For example, in cluster analysis, as shown in Figure 24, the following three clustering algorithms are used to perform data clustering. (1) Gaussian Mixture Model Multiple normal distributions are fitted to the graph, and clustering is performed for each normal distribution. Clustering can be performed without specifying the number of classes. (2) Hierarchical clustering This method clusters elements in order of their similarity, with the intermediate steps represented hierarchically and the final result shown as a tree diagram. It also allows for dynamic adjustments to the number of clusters. (3) K-means method This algorithm divides data into a specified number (k) of clusters, calculates the average value for each cluster, and then repeatedly clusters the data points that are close to that average value to arrive at appropriate clusters.

[0135] (Embodiment 3: Cluster Analysis Result Output Unit) The cluster analysis result output unit 2309 is configured to output the cluster analysis results.

[0136] <Embodiment 3: Human Resources Data Decoding Support System: Hardware Configuration> The hardware configuration of the personnel data decoding support system in Embodiment 3 will be explained with reference to a diagram.

[0137] Figure 25 shows the hardware configuration of the personnel data decoding support system in Embodiment 3. As shown in this figure, the information provision system in this embodiment includes a "CPU (Central Processing Unit)" 2501 that performs various calculations and a "main memory" 2502. It also includes a "non-volatile memory" 2503 that holds predetermined information, and a "network I / F (interface)" 2504 that sends and receives information with multiple user terminals 2506, multiple administrator terminals 2507, personnel system data 2508, personnel system data 2509, and other personnel-related data 2510.

[0138] Main memory is read by the CPU to execute programs that perform various processing tasks, and at the same time, it provides a work area that serves as the working area for those programs. In addition, both main memory and non-volatile memory are assigned multiple addresses, and programs executed by the CPU can exchange data with each other and perform processing by identifying and accessing these addresses. In Embodiment 3, the programs stored in "main memory" include a personnel-related data storage program, a personnel-related data acquisition program by item, a standard text retention program for each distribution type, a distribution type identification information reception program, a standard text acquisition program, a standard text output program, a cluster analysis rule retention program, a cluster analysis program, and a cluster analysis result output program. Furthermore, "main memory" and "non-volatile memory" store personnel-related data, personnel-related data by item, standard phrases, distribution type identification information, cluster analysis rules, and more.

[0139] The CPU executes the personnel-related data storage program stored in main memory to store personnel-related data in main memory and non-volatile memory. It also executes the item-specific personnel-related data acquisition program stored in main memory to store item-specific personnel-related data in main memory and non-volatile memory. Furthermore, it executes the distribution-type-specific boilerplate text retention program stored in main memory to store boilerplate text in main memory and non-volatile memory. It also executes the boilerplate text acquisition program stored in main memory to store boilerplate text in main memory and non-volatile memory based on distribution-type identification information. Furthermore, it executes the boilerplate text output program stored in main memory to output boilerplate text associated with item-specific personnel-related data. Furthermore, it executes the cluster analysis rule retention program stored in main memory to store cluster analysis rules in main memory and non-volatile memory. Furthermore, it executes the cluster analysis program stored in main memory to perform cluster analysis. Finally, it executes the cluster analysis result output program stored in main memory to output the cluster analysis results.

[0140] <Embodiment 3: Human Resources Data Decoding Support System: Processing Flow> Figure 26 is a diagram showing the processing flow when using the personnel data decoding support system in Embodiment 3. As shown in the figure, the processing method consists of a personnel-related data storage step S2601, an item-specific personnel data acquisition step S2602, a standard text holding unit step S2603 for each distribution type, a distribution type identification information reception step S2604, a standard text acquisition step S2605, a standard text output step S2606, a cluster analysis rule holding step S2607, a cluster analysis step S2608, and a cluster analysis result output step S2609.

[0141] These processing methods are executed by a personnel data decoding support system comprising: a personnel data storage unit that stores personnel-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to personnel management; an item-specific personnel data acquisition unit that acquires personnel-related data item by item; a distribution type-specific fixed phrase storage unit that holds fixed phrases to be presented to the user in advance, according to distribution type identification information for identifying the n-dimensional distribution type of the personnel-related data acquired item by item; a distribution type identification information receiving unit that receives distribution type identification information from the distribution of the acquired item-specific personnel-related data; a fixed phrase acquisition unit that acquires fixed phrases held based on the acquired distribution type identification information; a fixed phrase output unit that outputs the acquired fixed phrases in association with the acquired item-specific personnel-related data; a cluster analysis rule storage unit that holds cluster analysis rules for clustering the personnel-related data acquired item by item; a cluster analysis unit that clusters the personnel-related data acquired item by item based on the acquired personnel-related data and the held cluster analysis rules; and a cluster analysis result output unit that outputs the cluster analysis results.

[0142] The "HR-related data accumulation step" S2601 is the stage in which HR-related data, including statistically processable n-dimensional (n≧2) data, is accumulated by categorizing it into multiple items related to HR management.

[0143] "Item-specific personnel-related data acquisition step" S2602 is the stage in which personnel-related data is acquired item by item.

[0144] The "Step to retain standard text for each distribution type" S2603 is a step in which standard text is retained in advance to be presented to the user, according to the distribution type identification information used to identify the n-dimensional distribution type of personnel-related data acquired for each item.

[0145] The "Distribution Type Identification Information Reception Step" S2604 is the stage where distribution type identification information is received from the distribution of the acquired personnel-related data by item.

[0146] The "standard text acquisition step" S2605 is the stage in which standard texts are acquired based on the acquired distribution type identification information.

[0147] The "Standard Text Output Step" S2606 is the stage in which the acquired standard text is output in association with the acquired personnel-related data by item.

[0148] The "cluster analysis rule retention step" S2607 is the stage in which the cluster analysis rules for performing cluster analysis on personnel-related data acquired by item are retained.

[0149] The "cluster analysis step" S2608 is the stage in which the personnel-related data acquired item by item is clustered based on the acquired personnel-related data and the stored cluster analysis rules.

[0150] The "cluster analysis result output step" S2609 is the stage where the cluster analysis results are output.

[0151] <Embodiment 4 (mainly corresponding to claims 4, 12, and 20)> <Embodiment 4 Overview> Embodiment 4 is based on Embodiment 1 and includes a function to guide situational judgment based on observation of the OODA loop. As shown in Figure 22, even if the points to look at during the current situation confirmation are understood and the situational judgment can be made, it often occurs that the subsequent OODA loop cannot be executed because "it is unclear how to perform the analysis" or "it is unclear what points (key points) to look at in the analysis results." Embodiment 4 can resolve this situation and provides a function that enables the analysis of the gap between As is (current state) and To be (desired state). Furthermore, to achieve this, we provide a method executed by the CPU in the information provision system, which is a computer, and an operating program for the information provision system written in a readable and executable format for the information provision system, which is a computer.Hereafter, the same functional configuration, hardware configuration, and processing flow as in Embodiment 1 will be omitted from explanation as appropriate.

[0152] <Embodiment 4 Functional Configuration> Figure 27 is a conceptual diagram showing an example of the functional configuration of the personnel data decoding support system 2700 according to Embodiment 4. As shown in the figure, the "personnel data decoding support system" 2700 includes at least a "personnel-related data storage unit" 2701, an "item-specific personnel-related data acquisition unit" 2702, a "standard text storage unit for each distribution type" 2703, a "distribution type identification information receiving unit" 2704, a "standard text acquisition unit" 2705, a "standard text output unit" 2706, a "correlation analysis rule storage unit" 2707, a "correlation analysis unit" 2708, and a "correlation analysis result output unit" 2709.

[0153] <Description of each component in Embodiment 4> (Embodiment 4: Correlation Analysis Rule Holding Unit) In Embodiment 4, correlation analysis is provided as a data analysis method for analyzing personnel-related data and presenting key insights to the user. Correlation analysis numerically expresses how data relate to each other. This makes it possible to extract data that has a strong relationship with specific data, facilitating the analysis of the causes of events. The correlation analysis rule holding unit 2707 is configured to hold correlation analysis rules for analyzing personnel-related data acquired item by item. Correlation analysis rules include, for example, "measuring the correlation between data using MIC (Micro-Indicator)." Personnel information contains many categorical data, including employment categories and business locations, and non-linear correlations often exist between these data.

[0154] (Embodiment 4: Correlation Analysis Unit) The correlation analysis unit 2708 is configured to perform correlation analysis on the personnel-related data acquired item by item, based on the personnel-related data acquired item by item and the correlation analysis rules that are stored therein. For example, as shown in Figure 28, we can use the MIC (Maximum Information Coefficient) algorithm, which can measure correlations even between categorical and arbitrary data. The MIC captures the strength of the correlation better than Pearson's correlation coefficient.

[0155] (Embodiment 4: Correlation Analysis Result Output Unit) The correlation analysis result output unit 2709 is configured to output the correlation analysis results.

[0156] <Embodiment 4: Human Resources Data Decoding Support System: Hardware Configuration> The hardware configuration of the personnel data decoding support system in Embodiment 4 will be explained with reference to a diagram.

[0157] Figure 29 shows the hardware configuration of the personnel data decoding support system in Embodiment 4. As shown in this figure, the information provision system in this embodiment includes a "CPU (Central Processing Unit)" 2901 that performs various calculations and a "main memory" 2902. It also includes a "non-volatile memory" 2903 that holds predetermined information, and a "network I / F (interface)" 2904 that sends and receives information with multiple user terminals 2906, multiple administrator terminals 2907, personnel system data 2908, personnel system data 2909, and other personnel-related data 2910.

[0158] Main memory is read by the CPU to execute programs that perform various processing tasks, and at the same time, it provides a work area that serves as the working area for those programs. In addition, both main memory and non-volatile memory are assigned multiple addresses, and programs executed by the CPU can exchange data with each other and perform processing by identifying and accessing these addresses. In Embodiment 4, the programs stored in "main memory" include a personnel-related data storage program, a personnel-related data acquisition program by item, a standard text retention program for each distribution type, a distribution type identification information reception program, a standard text acquisition program, a standard text output program, a correlation analysis rule retention program, a correlation analysis program, and a correlation analysis result output program. Furthermore, the "main memory" and "non-volatile memory" store personnel-related data, personnel-related data by item, standard phrases, distribution type identification information, correlation analysis rules, and more.

[0159] The CPU executes the personnel-related data storage program stored in main memory to store personnel-related data in main memory and non-volatile memory. It also executes the item-specific personnel-related data acquisition program stored in main memory to store item-specific personnel-related data in main memory and non-volatile memory. Furthermore, it executes the distribution-type-specific boilerplate text retention program stored in main memory to store boilerplate text in main memory and non-volatile memory. It also executes the boilerplate text acquisition program stored in main memory to store boilerplate text in main memory and non-volatile memory based on distribution-type identification information. Furthermore, it executes the boilerplate text output program stored in main memory to output boilerplate text associated with item-specific personnel-related data. Furthermore, it executes the correlation analysis rule retention program stored in main memory to store correlation analysis rules in main memory and non-volatile memory. Furthermore, it executes the correlation analysis program stored in main memory to perform correlation analysis. Finally, it executes the correlation analysis result output program stored in main memory to output the correlation analysis results.

[0160] <Embodiment 4: Human Resources Data Decoding Support System: Processing Flow> Figure 30 shows the processing flow when using the personnel data decoding support system in Embodiment 4. As shown in the figure, the processing method consists of a personnel-related data storage step S3001, an item-specific personnel data acquisition step S3002, a standard text holding step S3003 for each distribution type, a distribution type identification information reception step S3004, a standard text acquisition step S3005, a standard text output step S3006, a correlation analysis rule holding step S3007, a correlation analysis step S3008, and a correlation analysis result output step S3009.

[0161] These processing methods are performed by a personnel data decoding support system comprising: a personnel data storage unit that stores personnel-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to personnel management; an item-specific personnel data acquisition unit that acquires personnel-related data item by item; a distribution type-specific fixed phrase storage unit that holds fixed phrases to be presented to the user in advance, according to distribution type identification information for identifying the n-dimensional distribution type of the personnel-related data acquired item by item; a distribution type identification information receiving unit that receives distribution type identification information from the distribution of the acquired item-specific personnel-related data; a fixed phrase acquisition unit that acquires fixed phrases held based on the acquired distribution type identification information; a fixed phrase output unit that outputs the acquired fixed phrases in association with the acquired item-specific personnel-related data; a correlation analysis rule storage unit that holds correlation analysis rules for performing correlation analysis on personnel-related data acquired item by item; a correlation analysis unit that performs correlation analysis on personnel-related data acquired item by item based on the acquired personnel-related data and the held correlation analysis rules; and a correlation analysis result output unit that outputs the correlation analysis results.

[0162] The "HR-related data accumulation step" S3001 is the stage in which HR-related data, including statistically processable n-dimensional (n≧2) data, is accumulated by categorizing it into multiple items related to HR management.

[0163] "Item-specific personnel-related data acquisition step" S3002 is the stage where personnel-related data is acquired item by item.

[0164] The "Step to retain standard text for each distribution type" S3003 is a step in which standard text is retained in advance to be presented to the user, according to the distribution type identification information used to identify the n-dimensional distribution type of personnel-related data acquired for each item.

[0165] The "Distribution Type Identification Information Reception Step" S3004 is the stage where distribution type identification information is received from the acquired item-specific personnel-related data distribution.

[0166] The "standard text acquisition step" S3005 is the stage in which standard texts are acquired based on the acquired distribution type identification information.

[0167] The "Standard Text Output Step" S3006 is the stage in which the acquired standard text is output in association with the acquired personnel-related data by item.

[0168] The "Correlation Analysis Rule Retention Step" S3007 is the stage in which the correlation analysis rules for performing correlation analysis on personnel-related data acquired item by item are retained.

[0169] The "correlation analysis step" S3008 is the stage in which personnel-related data acquired item by item is correlated with the stored correlation analysis rules.

[0170] The "Correlation Analysis Result Output Step" S3009 is the stage where the correlation analysis results are output.

[0171] <Embodiment 5 (mainly corresponding to claims 5, 13, and 21)> <Embodiment 5 Overview> Embodiment 5 is based on Embodiment 1 and includes a function to guide situational judgment based on observation of the OODA loop. As shown in Figure 22, even if the points to look at during the current situation confirmation are understood and the situational judgment can be made, it often occurs that the subsequent OODA loop cannot be executed because "it is unclear how to perform the analysis" or "it is unclear what points (key points) to look at in the analysis results." Embodiment 5 can resolve this situation and provides a function that enables the analysis of the gap between As is (current state) and To be (desired state). Furthermore, to achieve this, we provide a method executed by the CPU in the information provision system, which is a computer, and an operating program for the information provision system written in a readable and executable format for the information provision system, which is a computer.Hereafter, the same functional configuration, hardware configuration, and processing flow as in Embodiment 1 will be omitted from explanation as appropriate.

[0172] <Embodiment 5 Functional Configuration> Figure 31 is a conceptual diagram showing an example of the functional configuration of the personnel data decoding support system 3100 according to Embodiment 5. As shown in the figure, the "personnel data decoding support system" 3100 includes at least a "personnel-related data storage unit" 3101, an "item-specific personnel-related data acquisition unit" 3102, a "standard text storage unit for each distribution type" 3103, a "distribution type identification information receiving unit" 3104, a "standard text acquisition unit" 3105, a "standard text output unit" 3106, a "feature word analysis rule storage unit" 3107, a "feature word analysis unit" 3108, and a "feature word analysis result output unit" 3109.

[0173] <Description of each component in Embodiment 5> (Embodiment 5: Feature word analysis rule retention unit) In Embodiment 5, a data analysis method is provided for analyzing HR-related data and presenting key insights to the user: feature word analysis, which summarizes the data. Feature word analysis represents a large amount of data with a small amount of data. By enabling users to quickly grasp the overall picture of the data, the time required from reviewing the data to formulating measures can be significantly reduced. The feature word analysis rule holding unit 3107 is configured to hold feature word analysis rules for analyzing natural language data when the personnel-related data acquired item by item includes such data. Feature word analysis rules involve aggregating natural language data using methods such as co-occurrence networks or word clouds. Personnel information contains a certain amount of natural language data collected from goal management, performance evaluations, and surveys. A means is needed to aggregate this natural language data and visualize the relevant parts.

[0174] (Embodiment 5: Feature word analysis unit) The feature word analysis unit 3108 is configured to perform feature word analysis on natural language data based on the natural language data and the stored feature word analysis rules. For example, as shown in Figure 32, by creating a co-occurrence network to show the relationships between words that make up a text, and a word cloud to show the overall structure of the text, characteristic features of natural language data can be identified at a glance.

[0175] (Embodiment 5: Feature word analysis result output unit) The feature word analysis result output unit 3109 is configured to output the feature word analysis results.

[0176] <Embodiment 5: Human Resources Data Decoding Support System: Hardware Configuration> The hardware configuration of the personnel data decoding support system in Embodiment 5 will be explained with reference to a diagram.

[0177] Figure 33 shows the hardware configuration of the personnel data decoding support system in Embodiment 5. As shown in this figure, the information provision system in this embodiment includes a "CPU (Central Processing Unit)" 3301 that performs various calculations and a "main memory" 3302. It also includes a "non-volatile memory" 3303 that holds predetermined information, and a "network I / F (interface)" 3304 that sends and receives information with multiple user terminals 3306, multiple administrator terminals 3307, personnel system data 3308, personnel system data 3309, and other personnel-related data 3310.

[0178] Main memory is read by the CPU to execute programs that perform various processing tasks, and at the same time, it provides a work area that serves as the working area for those programs. In addition, both main memory and non-volatile memory are assigned multiple addresses, and programs executed by the CPU can exchange data with each other and perform processing by identifying and accessing these addresses. In Embodiment 5, the programs stored in "main memory" include a personnel-related data storage program, a personnel-related data acquisition program by item, a standard phrase storage program for each distribution type, a distribution type identification information receiving program, a standard phrase acquisition program, a standard phrase output program, a feature word analysis rule storage program, a feature word analysis program, and a feature word analysis result output program. Furthermore, the "main memory" and "non-volatile memory" store personnel-related data, personnel-related data by item, standard phrases, distribution type identification information, and characteristic word analysis rules.

[0179] The CPU executes the personnel-related data storage program stored in main memory to store personnel-related data in main memory and non-volatile memory. It also executes the item-specific personnel-related data acquisition program stored in main memory to store item-specific personnel-related data in main memory and non-volatile memory. Furthermore, it executes the distribution-type-specific boilerplate text retention program stored in main memory to store boilerplate text in main memory and non-volatile memory. It also executes the boilerplate text acquisition program stored in main memory to store boilerplate text in main memory and non-volatile memory based on distribution-type identification information. Furthermore, it executes the boilerplate text output program stored in main memory to output boilerplate text associated with item-specific personnel-related data. Furthermore, it executes the feature word analysis rule retention program stored in main memory to store feature word analysis rules in main memory and non-volatile memory. Furthermore, it executes the feature word analysis program stored in main memory to perform feature word analysis. Finally, it executes the feature word analysis result output program stored in main memory to output the feature word analysis results.

[0180] <Embodiment 5: Human Resources Data Decoding Support System: Processing Flow> Figure 34 shows the processing flow when using the personnel data decoding support system in Embodiment 5. As shown in the figure, the processing method consists of a personnel-related data storage step S3401, an item-specific personnel data acquisition step S3402, a standard text holding step S3403 for each distribution type, a distribution type identification information reception step S3404, a standard text acquisition step S3405, a standard text output step S3406, a correlation analysis rule holding step S3407, a correlation analysis step S3408, and a correlation analysis result output step S3409.

[0181] These processing methods are performed by a personnel data decoding support system having: a personnel data storage unit that stores personnel-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to personnel management; an item-specific personnel data acquisition unit that acquires personnel-related data item by item; a distribution type-specific fixed phrase storage unit that holds fixed phrases to be presented to the user in advance, according to distribution type identification information for identifying the n-dimensional distribution type of the personnel-related data acquired item by item; a distribution type identification information receiving unit that receives distribution type identification information from the distribution of the acquired item-specific personnel-related data; a fixed phrase acquisition unit that acquires fixed phrases held based on the acquired distribution type identification information; a fixed phrase output unit that outputs the acquired fixed phrases in association with the acquired item-specific personnel-related data; a feature word analysis rule storage unit that holds feature word analysis rules for analyzing natural language data when the acquired item-specific personnel-related data includes natural language data; a feature word analysis unit that analyzes the natural language data based on the natural language data and the stored feature word analysis rules; and a feature word analysis result output unit that outputs the feature word analysis results.

[0182] The "HR-related data accumulation step" S3401 is the stage in which HR-related data, including statistically processable n-dimensional (n≧2) data, is accumulated by categorizing it into multiple items related to HR management.

[0183] "Item-specific personnel-related data acquisition step" S3402 is the stage in which personnel-related data is acquired item by item.

[0184] The "Step to retain standard text for each distribution type" S3403 is a step in which standard text is retained in advance to be presented to the user, according to the distribution type identification information used to identify the n-dimensional distribution type of personnel-related data acquired for each item.

[0185] The "Distribution Type Identification Information Reception Step" S3404 is the stage where distribution type identification information is received from the acquired distribution of personnel-related data by item.

[0186] The "standard text acquisition step" S3405 is the stage in which standard texts are acquired based on the acquired distribution type identification information.

[0187] The "Standard Text Output Step" S3406 is the stage in which the acquired standard text is output in association with the acquired personnel-related data by item.

[0188] The "Feature Word Analysis Rule Retention Step" S3407 is the step in which, when the personnel-related data acquired item by item includes natural language data, the feature word analysis rules for analyzing that natural language data are retained.

[0189] The "feature word analysis step" S3408 is the stage in which the natural language data is subjected to feature word analysis based on the natural language data and the retained feature word analysis rules.

[0190] The "feature word analysis result output step" S3409 is the stage where the feature word analysis results are output.

[0191] <Example 2> The following describes an example (Example 2) based on Embodiments 3 to 5. First, we will explain a system that incorporates the OODA loop to support strategy execution, referring to Figure 22.

[0192] Next, the function of "being able to present the points and analysis methods that the displayed graph represents," which is a feature of the present invention, will be described in detail.

[0193] The personnel data interpretation support system according to Example 2 is configured to support "how to perform analysis" and "points to be seen in the analysis results" based on the "instructions" prepared for each type of graph displayed on the dashboard (see FIGS. 4 to 11).

[0194] Typical graphs displayed on the dashboard include bar graphs, histograms, pie charts, line graphs, scatter plots, radar charts, cross-tabulations, and scalar charts. Here, the histogram will be taken as an example for detailed explanation. Even for other graphs, although the fixed texts for each distribution type are different, the flow of the dialogue is the same, so detailed explanations will be omitted. Needless to say, it is also applicable to graphs other than bar graphs, histograms, pie charts, line graphs, scatter plots, radar charts, cross-tabulations, and scalar charts by appropriately preparing fixed texts for each distribution type.

[0195] The fixed texts for each classification type are held in the comment column in tabular form for each distribution type, as shown in FIGS. 4 to 11, for each of the bar graph, histogram, pie chart, line graph, scatter plot, radar chart, cross-tabulation, and scalar chart. For example, referring to the histogram in FIG. 5, when "isolated island type" and "double mountain type" are combined, the fixed text "· For data that is significantly separated, there are often cases where special circumstances exist and individual responses are required. Check the target data and identify the cause. · There may be multiple underlying causes. Perform cluster analysis and investigate the trends for each'mountain' of the [X-axis item]. · Perform correlation analysis and investigate items that are strongly related to the [X-axis item]." is held.

[0196] Figure 17 shows a state where the average overtime hours distribution is visualized as a histogram and displayed on the left side of the dashboard as human capital information. For example, the fixed-form sentence holding unit 0203 for each distribution type retrieves an instruction of "histogram", and as Question 1, when a target value (for example, 24 hours) is set for the average overtime hours, the personnel data interpretation support system detects that there are employees whose average overtime hours exceed 24 hours, and guides with "[Average overtime hours] exceed the target value. Let's conduct an analysis and formulate improvement measures.", and when the person in charge of the personnel department selects "Analyze this problem", it is considered that a Yes answer has been given to Question 1, and the dialogue with the personnel data interpretation support system starts on the right side of the screen. Particularly when no target value is set, it guides with "Is there anything unusual about the distribution (shape of the graph) of this graph?", and when the person in charge of the personnel department selects "Analyze this problem", it is considered that a Yes answer has been given to Question 1, and the dialogue with the system starts on the right side of the screen.

[0197] First, the distribution type identification information reception unit 0204 receives distribution type identification information based on the answer received for the question held in the instruction corresponding to the type of graph (in Figure 17, "histogram") displayed on the left side of the screen. The dialogue proceeds according to the fixed-form sentences (guidance sentences, question sentences, and comment sentences to the user) held by the fixed-form sentence holding unit 0203 for each distribution type. As Question 2, it guides while showing a typical example with "Are there any data points that are far apart in this graph?" Following the guidance, when the person in charge of the personnel department selects "Yes", as shown in Figure 18, as Question 3, it guides while showing a typical example with "Are there multiple 'peaks' in this graph?" In this process, the person in charge of the personnel department can learn that for the case of "histogram", whether there are data points that are far apart and whether there are multiple "peaks" are the points of observation (current situation recognition).

[0198] If the answer to both question 2 and question 3 is "yes," the standard text output unit 0306 outputs the standard text stored in the histogram instructions based on the distribution type identification information "11" received by the distribution type identification information receiving unit 0305.

[0199] As shown in Figure 18, for example, the dialogue screen includes: "The content of your response" There is data that is separated. There are multiple mountains. This graph may have multiple underlying causes. Please analyze the data to identify the cause. correspondence (1) Data that is significantly different often has special circumstances and requires individual attention. Check the data in question and identify the cause. (2) There may be multiple underlying causes. Let's perform cluster analysis and investigate the trends in each "peak" of [average overtime hours]. (3) Let's conduct a correlation analysis and investigate the items that have a strong relationship with [average overtime hours]. An explanatory text or comment will be displayed.

[0200] This allows HR personnel to learn that "individual analysis is suitable for data that are significantly different," "cluster analysis is suitable for investigating trends in each 'peak' of [average overtime hours]," and "correlation analysis is suitable for investigating items that are strongly related to [average overtime hours]."

[0201] Similarly, in the instructions for the "bar graph" in Figure 4, if the distribution type is "achievement type," the standard message "Perform correlation analysis and investigate items that are strongly related to [Y-axis items]" will be displayed. Furthermore, if the HR-related data obtained by item includes natural language data, the standard message "Perform feature word analysis and compare categories with high and low [Y-axis items] to investigate the cause." will be displayed. This allows HR personnel to learn that "correlation analysis is suitable for investigating items that are strongly related to [Y-axis items]" and "characteristic word analysis is suitable for comparing categories with high and low [Y-axis items]."

[0202] Similarly, in the instructions for the "Scatter Plot" in Figure 8, if the distribution type is "Concentrated," the following standard phrases will be displayed: "Multiple causes may be underlying. Let's perform cluster analysis to investigate the trends of the group." and "Let's perform correlation analysis to investigate items that are strongly related to [Y-axis items]." This allows HR personnel to learn that "cluster analysis is suitable for investigating group trends," and "correlation analysis is suitable for investigating items that are strongly related to the [Y-axis items]."

[0203] Returning to Figure 18 and selecting "Perform Analysis," cluster analysis and correlation analysis will be automatically executed according to the predefined text. Figure 35 shows an example of how the results of an automatically executed cluster analysis are displayed on the interactive screen. In addition to reporting the results of the cluster analysis, stating "Cluster analysis was performed. [Four groups were found in this overtime period]," it also displays the characteristics of each group and recommendations from the HR data interpretation support system, providing a function that highlights the key points (essentials) to look for in the analysis results. Figure 36 shows an example of how the results of the automatically executed correlation analysis are displayed on the interactive screen. It initially displays the graph with the strongest correlation, which was automatically generated, and also expands to display multiple graphs of the top correlations, with the message, "Correlation analysis has been performed. The following items have a strong relationship with [annual income]."

[0204] In addition, selecting "View Details" switches to a screen display similar to the image shown in Figure 37. This screen also displays comments such as, "--If there is a correlation--Overtime hours, performance evaluations, and training participation hours have a particularly strong correlation with annual income. By improving this relationship, you can create a compensation system that rewards outstanding employees. --If there is no correlation--No correlation was found in the personnel information. If this seems unusual, try analyzing it in conjunction with other information (for example, accounting information)." This feature helps to highlight the key points (essentials) of the analysis results.

[0205] <Example 3> Furthermore, an example (Example 3) based on Embodiments 2 and 5 will be described. The difference between Example 3 and Example 2 is that a "target value" is set. Assuming that the process described in Figure 3 has progressed to (e) information visualization and a bar graph is displayed on the dashboard, following the steps outlined in Figure 3. As previously explained, the HR data interpretation support system assists with the OODA loop from observation (Observe) to situational judgment (Orient) according to the bar graph instructions (Figure 4) (Figure 22). Here, the standard text storage unit 1908 for each distribution type and target is configured to store standard texts to be presented to the user in advance, according to the distribution type identification information and the target value.

[0206] A standard phrase is, for example, in a "achievement-type" bar graph, the target value is set for the [Y-axis item]. "—With a target value—Let's perform characteristic word analysis and investigate the causes by comparing it with the group that has achieved the target value (or the group used as a benchmark)." --No target value-- Let's perform a feature word analysis and investigate the cause by comparing categories with high and low [Y-axis items]. This is the sentence. Here, "——With target value——" and "——Without target value——" correspond to tags in the structured document. Depending on "——With target value——" and "——Without target value——", the stereotyped sentence "Perform keyword analysis and investigate the causes by comparing with the group that has achieved the target value (or the group used as a benchmark)." and the stereotyped sentence "Perform keyword analysis and investigate the causes by comparing the categories with high [Y-axis item] and low [Y-axis item]." are retained in the stereotyped sentence retention part for each distribution type target. That is, in the bar graph instruction (Figure 4), when the distribution type is "achievement type" and a "target value" is set, and the personnel-related data obtained by item contains natural language data, the target-dependent stereotyped sentence acquisition means 1905A acquires the stereotyped sentence "Perform keyword analysis and investigate the causes by comparing with the group that has achieved the target value (or the group used as a benchmark).", and the stereotyped sentence output unit 1906 outputs this stereotyped sentence and displays it on the dialogue screen. On the dialogue screen, similar to the average overtime distribution graph in Figure 18, buttons for "Analyze" and "Do not analyze" are displayed. When "Analyze" is selected, automatic analysis is performed. Figure 38 is an example display in which the results of the automatically executed keyword analysis are displayed on the dialogue screen. In this example too, not only does the keyword comparison analysis report the result as "Keyword analysis has been performed. Frequently occurring words have been linked to [target].", but by also displaying together, it tells us the key points (the crucial points) of the analysis results.

[0207] In the above Embodiments 1 to 5 and Examples 1 to 3, generally, since OODA is known as an exemplary action process suitable for data-driven, the personnel data decoding support system applied to the observation (situation confirmation / detail confirmation) and situation judgment (action analysis / analysis result confirmation) of the OODA loop has been described in detail, but it is not limited to this.

[0208] As is well known, OODA focuses on decision-making in a competitive environment and is effective for making quick and accurate judgments and taking prompt actions. For example, Observe: Identify which job types have the most overtime hours. Orient (Situation Assessment): Sales staff in the XX business unit are working excessive overtime. Decision: Increase the number of sales staff in the XX business unit. Act: Increase the number of sales staff in the XX business unit by 2 through internal transfers. That is the case. In contrast, PDCA, for example, is a method used in quality control and project management, and is effective for achieving continuous improvement. Plan: Limit overtime to 20 hours per month. Do (Execution): Conduct work for 1 month and 3 months and track overtime hours. Check (Evaluation): Investigate organizations and job types where overtime hours exceed 20 hours per month, and analyze the causes. Action (Improvement): Increase the number of sales staff in the XX business unit by 2 through internal transfers. That is the case. In short, OODA allows for quicker identification of the problem of overwork and enables real-time, accurate decision-making and swift action, even in rapidly changing business fields. PDCA, on the other hand, involves planning, execution, evaluation, and improvement, which can take time to implement and sometimes lead to reactive measures, such as overwork resulting in employee turnover. Thus, although there are advantages and disadvantages depending on the application field, it should be noted that this does not in any way prevent the use of this method as a data decoding support system that applies the technical ideas of the present invention to the Do (execution) and Check (evaluation) phases of the PDCA loop. [Explanation of Symbols]

[0209] 0200, 1900, 2300, 2700, 3100 Human Resources Data Decoding Support System 0201, 1901, 2301, 2701, 3101 Human Resources Data Storage Department 0202, 1902, 2302, 2702, 3102 Itemized Personnel Data Acquisition Department 0203, 1903, 2303, 2703, 3103 Standard text storage unit for each distribution type 0204, 1904, 2304, 2704, 3104 Distribution Type Identification Information Reception Unit 0205, 1905, 2305, 2705, 3105 Fixed phrase acquisition department 0206, 1906, 2306, 2706, 3106 Standard text output unit 1905A Target-dependent fixed phrase acquisition part 1907 Target value holding unit 1908 Distribution Type Target-Specific Fixed Text Storage Unit 2307 Cluster analysis rule storage unit 2308 Cluster Analysis Department 2309 Cluster Analysis Result Output Unit 2707 Correlation analysis rule retention unit 2708 Correlation Analysis Department 2709 Correlation Analysis Result Output Unit 3107 Feature word analysis rule retention unit 3108 Feature word analysis unit 3109 Feature word analysis result output unit

Claims

1. A personnel-related data storage unit that stores personnel-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to personnel management, and The Item-Specific Personnel Data Acquisition Unit acquires personnel-related data by item, A distribution type-specific predefined text storage unit holds predefined texts to be presented to the user in advance, according to distribution type identification information for identifying the n-dimensional distribution type of personnel-related data acquired by item, A distribution type identification information receiving unit receives distribution type identification information from the distribution of acquired personnel-related data by item, A boilerplate text acquisition unit that acquires boilerplate text based on the acquired distribution type identification information, A standard text output unit that outputs the acquired standard text in association with the acquired personnel-related data by item, A personnel data decoding support system.

2. A target value holding unit that holds a target value which is a target value for at least some of the aforementioned items, A distribution type target-specific fixed text storage unit that stores a fixed text to be presented to the user in advance, according to the distribution type identification information and the target value, It has, The personnel data decoding support system according to claim 1, wherein the standard text acquisition unit has a target-dependent standard text acquisition means that acquires a stored standard text based on distribution type identification information and, if there is a target value assigned to the item from which the distribution type identification information was obtained, that target value.

3. A cluster analysis rule holder unit holds cluster analysis rules for performing cluster analysis on personnel-related data acquired by item, A cluster analysis unit performs cluster analysis on the HR-related data acquired item by item, based on the acquired HR-related data and the stored cluster analysis rules. A cluster analysis result output unit that outputs the cluster analysis results, The personnel data decoding support system according to claim 1 or claim 2, further comprising:

4. A correlation analysis rule holder unit that holds correlation analysis rules for performing correlation analysis on personnel-related data acquired by item, A correlation analysis unit performs correlation analysis on the personnel-related data acquired item by item, based on the personnel-related data acquired item by item and the correlation analysis rules that are maintained. A correlation analysis result output unit that outputs the correlation analysis results, The personnel data decoding support system according to claim 1 or claim 2, further comprising:

5. When the personnel-related data acquired item by item includes natural language data, the feature word analysis rule holder holds feature word analysis rules for analyzing that natural language data, A feature word analysis unit performs feature word analysis on the natural language data based on the aforementioned natural language data and the retained feature word analysis rules. A feature word analysis result output unit that outputs the feature word analysis results, The personnel data decoding support system according to claim 1 or claim 2, further comprising:

6. The personnel data decoding support system according to claim 1, wherein the aforementioned items include at least one of the following: information on compliance and ethics, information on costs, information on diversity, information on leadership, information on organizational culture, information on health management, information on productivity, information on recruitment, transfers and turnover, information on skills and abilities, information on succession planning, and information on the workforce.

7. The personnel data decoding support system according to claim 6, wherein the aforementioned items further include at least one of the following: information on the number of personnel and personnel composition, information on overtime hours, information on paid leave utilization rates, information on training, information on skills, information on competencies, and information on goal setting and evaluation.

8. The personnel data decoding support system according to claim 1, wherein the distribution type identification information is distribution type identification information of at least one of the following graphs: bar graph, histogram, pie chart, line graph, scatter plot, radar chart, cross-tabulation, and scalar chart.

9. A method executed by the CPU in a computer-based personnel data decoding support system, A human resources data storage step involves storing human resources-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to human resources management. Steps for acquiring personnel-related data by item, A step to store predefined text for each distribution type, which is stored according to distribution type identification information for identifying the n-dimensional distribution type of personnel-related data acquired by item, and a step to store predefined text to be presented to the user in advance. A distribution type identification information reception step that receives distribution type identification information from the distribution of acquired personnel-related data by item, A boilerplate text acquisition step that acquires boilerplate text based on the acquired distribution type identification information, A standard text output step that outputs the acquired standard text in association with the acquired HR-related data by item, A method of having.

10. A method executed by the CPU in a computer-based personnel data decoding support system, A target value holding step that holds a target value which is a target value for at least some of the aforementioned items, A step of storing a standard phrase for each distribution type target, which stores a standard phrase to be presented to the user in advance, according to the distribution type identification information and the target value, It has, The method according to claim 9, wherein the standard text acquisition step further comprises a target-dependent standard text acquisition substep that acquires a held standard text based on distribution type identification information and, if there is a target value assigned to the item from which the distribution type identification information was obtained, that target value.

11. A method executed by the CPU in a computer-based personnel data decoding support system, A cluster analysis rule retention step that holds cluster analysis rules for cluster analysis of HR-related data acquired by item, A cluster analysis step is performed to perform cluster analysis on the HR-related data obtained item by item, based on the HR-related data obtained item by item and the stored cluster analysis rules. A cluster analysis result output step that outputs the cluster analysis results, The method according to claim 9 or claim 10, further comprising:

12. A method executed by the CPU in a computer-based personnel data decoding support system, A correlation analysis rule retention step that holds correlation analysis rules for performing correlation analysis on personnel-related data acquired by item, A correlation analysis step is performed to correlate the HR-related data obtained item by item with the established correlation analysis rules. A correlation analysis result output step that outputs the correlation analysis results, The method according to claim 9 or claim 10, further comprising:

13. A method executed by the CPU in a computer-based personnel data decoding support system, When the HR-related data obtained item by item includes natural language data, the step of storing the feature word analysis rules for performing feature word analysis on that natural language data, A feature word analysis step in which the natural language data is analyzed based on the aforementioned natural language data and the retained feature word analysis rules, A feature word analysis result output step that outputs the feature word analysis results, The method according to claim 9 or claim 10, further comprising:

14. A method executed by the CPU in a computer-based personnel data decoding support system, The method according to claim 9, wherein the aforementioned items include at least one of the following: information on compliance and ethics, information on costs, information on diversity, information on leadership, information on organizational culture, information on health management, information on productivity, information on recruitment, transfers and turnover, information on skills and abilities, information on succession planning, and information on the workforce.

15. A method executed by the CPU in a computer-based personnel data decoding support system, The method according to claim 14, wherein the aforementioned items further include at least one of the following: information regarding the number of personnel and personnel composition, information regarding overtime hours, information regarding the rate of paid leave taken, information regarding training, information regarding skills, information regarding competencies, and information regarding goal setting and evaluation.

16. A method executed by the CPU in a computer-based personnel data decoding support system, The method according to claim 9, wherein the distribution type identification information is distribution type identification information of at least one of the following graphs: bar graph, histogram, pie chart, line graph, scatter plot, radar chart, cross-tabulation, and scalar chart.

17. An operating program for a personnel data decoding support system, which is written in a readable and executable manner for the personnel data decoding support system, A human resources data storage step involves storing human resources-related data, including statistically processable n-dimensional (n≧2) data, categorized into multiple items related to human resources management. Steps for acquiring personnel-related data by item, A step to store predefined text for each distribution type, which is stored according to distribution type identification information for identifying the n-dimensional distribution type of personnel-related data acquired by item, and a step to store predefined text to be presented to the user in advance. A distribution type identification information reception step that receives distribution type identification information from the distribution of acquired personnel-related data by item, A boilerplate text acquisition step that acquires boilerplate text based on the acquired distribution type identification information, A standard text output step that outputs the acquired standard text in association with the acquired HR-related data by item, An operating program for the personnel data decoding support system, which is a computer having the following characteristics, to be executed by the personnel data decoding support system.

18. An operating program for a personnel data decoding support system, which is written in a readable and executable manner for the personnel data decoding support system, A target value holding step that holds a target value which is a target value for at least some of the aforementioned items, A step of storing a standard phrase for each distribution type target, which stores a standard phrase to be presented to the user in advance, according to the distribution type identification information and the target value, It has, An operation program for a personnel data decoding support system, which is a computer according to claim 17, to be executed by the personnel data decoding support system, wherein the aforementioned boilerplate text acquisition step further comprises a target-dependent boilerplate text acquisition substep that acquires a held boilerplate text based on distribution type identification information and, if there is a target value assigned to the item from which the distribution type identification information was obtained, that target value.

19. An operating program for a personnel data decoding support system, which is written in a readable and executable manner for the personnel data decoding support system, A cluster analysis rule retention step that holds cluster analysis rules for cluster analysis of HR-related data acquired by item, A cluster analysis step is performed to perform cluster analysis on the HR-related data obtained item by item, based on the HR-related data obtained item by item and the stored cluster analysis rules. A cluster analysis result output step that outputs the cluster analysis results, An operation program for a personnel data decoding support system to be executed by the personnel data decoding support system, which is a computer according to claim 17 or claim 18, further comprising the above.

20. An operating program for a personnel data decoding support system, which is written in a readable and executable manner for the personnel data decoding support system, A correlation analysis rule retention step that holds correlation analysis rules for performing correlation analysis on personnel-related data acquired by item, A correlation analysis step is performed to correlate the HR-related data obtained item by item with the established correlation analysis rules. A correlation analysis result output step that outputs the correlation analysis results, An operation program for a personnel data decoding support system to be executed by the personnel data decoding support system, which is a computer according to claim 17 or claim 18, further comprising the above.

21. An operating program for a personnel data decoding support system, which is written in a readable and executable manner for the personnel data decoding support system, When the HR-related data obtained item by item includes natural language data, the step of storing the feature word analysis rules for performing feature word analysis on that natural language data, A feature word analysis step in which the natural language data is analyzed based on the aforementioned natural language data and the retained feature word analysis rules, A feature word analysis result output step that outputs the feature word analysis results, An operation program for a personnel data decoding support system, characterized in that it is executed by the personnel data decoding support system, which is a computer according to claim 17 or claim 18, further comprising the above.

22. An operating program for a personnel data decoding support system, which is written in a readable and executable manner for the personnel data decoding support system, An operation program for a personnel data decoding support system, which is a computer according to claim 17, to be executed by the personnel data decoding support system, wherein the aforementioned multiple items include at least one of the following: information on compliance and ethics, information on costs, information on diversity, information on leadership, information on organizational culture, information on health management, information on productivity, information on recruitment, transfers and turnover, information on skills and abilities, information on succession planning, and information on the workforce.

23. An operating program for a personnel data decoding support system, which is written in a readable and executable manner for the personnel data decoding support system, An operation program for a personnel data decoding support system, which is a computer according to claim 22, to be executed by the personnel data decoding support system, wherein the aforementioned multiple items further include at least one of the following: information on the number of personnel and personnel composition, information on overtime hours, information on paid leave utilization rates, information on training, information on skills, information on competencies, and information on goal setting and evaluation.

24. An operating program for a personnel data decoding support system, which is written in a readable and executable manner for the personnel data decoding support system, An operation program for a personnel data decoding support system, which is a computer according to claim 17, to be executed by the personnel data decoding support system, wherein the distribution type identification information is distribution type identification information for at least one of the following graphs: bar graph, histogram, pie chart, line graph, scatter plot, radar chart, cross tabulation, and scalar chart.

Images