Information processing device, method, program

Abstract

There is a need for technologies that can provide even greater support to help people who want to seek advice to continue working on the issues they are seeking help for. [Solution] The memory unit is configured to store various information such as knowledge and systems related to support for working. The program causes one or more computer processors to receive input of consultation content from the consultant, determine whether to generate a first prompt that includes search results from various information stored in the memory unit according to the type of consultation content received, or generate a second prompt that includes a predetermined instruction without including the search results, generate either the first or second prompt according to the determination result, and present the consultant with an answer generated by the large-scale language model by providing the generated first or second prompt to the large-scale language model.

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus, method, and program.

Background Art

[0002] There is a technology for assisting people who want to consult. For example, while working, one may want to consult about caregiving or childcare as something related to people around them. Also, one may want to consult about their own career as a worker.

[0003] These consultation contents may cover a long period of time on the time axis. One may not always be able to focus on the content one wants to consult. For example, for a business caregiver who cares while working, it is necessary to collect information according to the caregiving situation and also face the mental burden caused by daily caregiving.

[0004] There is also a system that performs natural language processing such as question-and-answer using a large language model (LLM: Large Language Models). Patent Document 1 describes a technique for dealing with the problem that "inputting a prompt generated by adding reference information to a question sentence into an LLM is effective in that it is easier to obtain an appropriate answer sentence compared to the case where no reference information is added. However, since there is a character count limit for the input of the Completion model of the LLM, the text generated based on the input question sentence cannot be added to the reference information without limit."

[0005] Patent Document 1 describes a technique aimed at "being able to generate a prompt by adding an effective sentence as reference information to the input question sentence within the determined character count limit."

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

[0007] It is important for both government agencies and organizations such as businesses to support people who want to seek advice in this way. For example, when working while also dealing with caregiving, childcare, etc., while various forms of support such as financial assistance are available, given that the subject of the consultation may be a long-term endeavor, it is even more important to provide support so that they can continue working while also continuing to work on the issues they are seeking advice on.

[0008] Therefore, there is a need for technologies that provide even greater support to help people who want to seek advice to continue working on the issues they are seeking help for. [Means for solving the problem]

[0009] According to one embodiment shown in this disclosure, a program is provided for operating a computer having one or more computer processors. The memory unit is configured to store various information such as knowledge and systems related to support for work. The program causes one or more computer processors to perform the following steps: receive input of consultation content from a consultant; determine whether to generate a first prompt including search results from various information stored in the memory unit, or a second prompt including a default instruction without search results, depending on the type of consultation content received; generate either the first or second prompt according to the determination result; provide the generated first or second prompt to a large-scale language model to obtain an answer generated by the large-scale language model in accordance with the instructions shown in the prompt; and present the answer generated by the large-scale language model to the consultant. [Effects of the Invention]

[0010] This disclosure will provide even greater support to those who wish to seek advice, enabling them to continue working on the issues they wish to address. [Brief explanation of the drawing]

[0011] [Figure 1] Figure 1 shows the configuration of System 1. [Figure 2] Figure 2 shows the configuration of server 20. [Figure 3] Figure 3 shows the configuration of terminal 10. [Figure 4] Figure 4 shows the data structure of the user database 211. [Figure 5] Figure 5 shows the data structure of the support system and service information database 212. [Figure 6] Figure 6 shows the data structure of the consultation history database 213. [Figure 7] Figure 7 shows the processing flow when the client inputs the details of their consultation, and the server 95 of the large-scale language model service generates an answer based on the results of searches in various databases, or generates an answer tailored to the client's needs. [Figure 8] Figure 8 shows an example of an operation screen that receives input from the user regarding their consultation and displays the response. [Modes for carrying out the invention]

[0012] The embodiments of this disclosure will be described below with reference to the drawings. In all the drawings illustrating the embodiments, common components are denoted by the same reference numerals, and repeated explanations are omitted. The following embodiments are not intended to unduly limit the content of this disclosure as described in the claims. Not all components shown in the embodiments are necessarily essential components of this disclosure. Also, each drawing is a schematic diagram and is not necessarily a strict illustration.

[0013] Also, in the following description, a "processor" is one or more processors. At least one processor is typically a microprocessor such as a CPU (Central Processing Unit), but may also be other types of processors such as a GPU (Graphics Processing Unit). At least one processor may be single-core or multi-core.

[0014] Also, at least one processor may be a processor in a broad sense, such as a hardware circuit (e.g., FPGA (Field-Programmable Gate Array) or ASIC (Application Specific Integrated Circuit)) that performs part or all of the processing.

[0015] Also, in the following description, expressions such as "xxx table" may be used to describe information from which an output is obtained for an input. This information may be data of any structure or a learning model such as a neural network that generates an output for an input. Therefore, "xxx table" can be referred to as "xxx information".

[0016] Also, in the following description, the configuration of each table is an example. One table may be divided into two or more tables, or all or part of two or more tables may be one table.

[0017] Also, in the following description, the "program" may be used as the subject to describe the processing. However, since the program is executed by the processor to perform the defined processing while appropriately using the storage unit and / or the interface unit, etc., the subject of the processing may be the processor (or a device such as a controller having that processor).

[0018] The program may be installed in a device such as a computer, or may be on, for example, a program distribution server or a computer-readable (e.g., non-transitory) recording medium. Also, in the following description, two or more programs may be realized as one program, or one program may be realized as two or more programs.

[0019] Also, in the following description, an identification number is used as identification information for various objects, but other types of identification information (e.g., an identifier including letters or symbols) may be adopted.

[0020] Also, in the following description, when describing elements of the same type without distinction, reference signs (or common signs among reference signs) are used, and when describing elements of the same type by distinction, the identification numbers (or reference signs) of the elements may be used.

[0021] Also, in the following description, the control lines and information lines show what is considered necessary for explanation, and not necessarily all control lines and information lines on the product are shown. All configurations may be interconnected.

[0022] <Outline of the Embodiment> In the following embodiments, a system that responds to various consultation modes of a person who wants to consult will be described.

[0023] (1) While working, consultations mainly related to interactions with people close by, such as "caregiving" and "child-rearing" While working, one may be mainly engaged in interactions with family members such as caregiving and child-rearing. For example, family members may take care of a parent, or parents may rear children, or multiple people may engage in caregiving and child-rearing together.

[0024] Government organizations and businesses sometimes provide support systems (such as subsidies) to enable employees to continue working while also managing caregiving or childcare responsibilities. However, the difficulty in even knowing about these support systems and finding one that suits one's needs can mean that these systems may not be utilized by those who need them.

[0025] Furthermore, while specific services may be offered (such as temporary childcare), a lack of information about capacity may lead to users not utilizing these services even when they are available, potentially resulting in the services themselves going unused.

[0026] Furthermore, individuals may encounter various problems while engaging in "caregiving" or "childcare." When multiple causes of these problems overlap, it becomes difficult to identify them on one's own, and these problems may not be resolved, potentially leading to a decline in motivation to continue engaging in "caregiving" or "childcare."

[0027] (2) Consultations concerning the workers themselves and their work environment, including consultations on "career" and "harassment". While employees themselves may think about their "careers," they may not understand the root cause of their career concerns, be unable to find options, or fail to take action to adopt those options. Support is also being provided to help them design their "careers" in this way.

[0028] (3) Consultations received by the business entity from customers and other demanders and users. When a business provides products or services to customers, if a defect occurs in those products or services, customers may become dissatisfied and emotionally need to resolve the issue. In addition to wanting the defective product or service repaired and restored to its original condition, customers may also have questions about how to use the product or service, even if it's not a major complaint.

[0029] In order to address the various types of consultations described above, the following description of the embodiment states that, depending on the type of consultation, (i) information necessary for the consultant may be provided by searching for information sources to refer to and summarizing the search results using a large-scale language model,

[0030] (ii) When the content of the consultation is such as "I want to vent" or "I want to express my dissatisfaction," and the person seeking advice is not even clear about what their problem is, then, in order to provide a message that is empathetic to the person seeking advice, tailored to their situation (such as their caregiving situation), This section describes the system that performs this task.

[0031] <1.1 System Configuration Diagram> Figure 1 shows the configuration of System 1.

[0032] System 1 may be a system in which an individual operates a terminal 10 to use services provided by server 20 (for example, a government organization may operate a website to support people who provide care, etc.), or it may be a system in which one or more employees, such as a business company, are employed and these employees use the services provided by server 20.

[0033] System 1, shown in Figure 1, includes a server 20, employee terminals 10, a server 95 for large-scale language modeling services, a local government server 94, a human resources and labor services server 96, an information provision services server 98, and a human resources staff terminal 12. These devices communicate with each other via a network 80.

[0034] In the illustrated example, one of the terminals 10 is shown as the terminal used by users of the service provided by server 20, but each user is operating their own terminal.

[0035] In this embodiment, each device (terminal device, server, etc.) can also be considered as an information processing device. That is, the collection of each device can be considered as a single "information processing device," and System 1 may be formed as a collection of multiple devices. The way in which the multiple functions required to realize System 1 according to this embodiment are distributed to one or more hardware can be appropriately determined in view of the processing capacity of each hardware and / or the specifications required for System 1.

[0036] Terminal 10 is a device operated by the user. Terminal 10 can be implemented, for example, as follows: • Handheld mobile devices such as smartphones and tablets • Stationary PCs (Personal Computers), Laptop PCs • Wearable devices worn by the user (watch-type, glasses-type, etc.) Terminal 10 includes a communication interface (IF) 12, an input device 13, an output device 14, memory 15, storage 16, and a processor 19.

[0037] The communication interface 12 is an interface for inputting and outputting signals so that terminal 10 can communicate with an external device.

[0038] The input device 13 is a device for receiving input operations from the user (for example, a touch panel, touchpad, pointing device such as a mouse, keyboard, etc.).

[0039] The output device 14 is a device (such as a display or speaker) for presenting information to the user.

[0040] Memory 15 is for temporarily storing programs and data processed by programs, etc., and is a volatile memory such as DRAM (Dynamic Random Access Memory).

[0041] Storage 16 is for storing data, and can be, for example, flash memory or an HDD (Hard Disk Drive).

[0042] The processor 19 is hardware for executing the instruction set described in the program, and consists of an arithmetic unit, registers, peripheral circuits, etc.

[0043] Server 20 is a device that provides users with a service that, while providing information on support systems and the like to the person seeking advice, also takes into account the person's complaints and other circumstances when they want to vent, and responds to them using language processing.

[0044] The server 20 includes a communication interface 22, an input / output interface 23, memory 25, storage 26, and a processor 29.

[0045] Communication IF22 is an interface for inputting and outputting signals so that the server 20 can communicate with external devices.

[0046] Input / Output IF23 functions as an interface between an input device for receiving user input operations and an output device for presenting information to the user.

[0047] Memory 25 is for temporarily storing programs and data processed by programs, etc., and is a volatile memory such as DRAM (Dynamic Random Access Memory).

[0048] Storage 26 is for storing data, and can be, for example, flash memory or an HDD (Hard Disk Drive).

[0049] The processor 29 is hardware for executing the instruction set described in the program, and consists of an arithmetic unit, registers, peripheral circuits, etc.

[0050] Server 95 of the Large-Scale Language Model Service is a server that executes language processing tasks using language models built through learning processes including artificial intelligence (AI). An LLM (Large Language Model) is a model that has been pre-trained on large amounts of data (such as text data), for example, a large amount of web content on the internet, or a large amount of data stored in a designated database, and can perform various language processing tasks by being given a task.

[0051] The server 95 of the large-scale language model service accepts prompt input in the form of text, images, audio, etc., and generates and responds with answers to those prompts. Examples of LLMs include GPT-3, GPT-4, and GPT-4o developed by OpenAI, and Gemini developed by Google.

[0052] Server 94, managed by the local government, is a device used to provide websites operated by administrative organizations and contains information on various support programs and other related matters.

[0053] The HR and Labor Services Server 96 is a system that provides employee management and labor management services.

[0054] The HR and labor services server 96 maintains a list of employees and manages information such as hiring, termination, arrival times, work schedules, and leave requests.

[0055] Server 98 of the information provision service is a device that transmits information such as the status of service provision through a website in order to support various users.

[0056] Terminal 12 for the HR staff is a terminal operated by the HR staff.

[0057] <1.2 Functional Configuration of Server 20> Figure 2 shows the configuration of server 20. As shown in Figure 2, server 20 functions as a communication unit 201, a storage unit 202, and a control unit 203.

[0058] The communications unit 201 performs processing to enable the server 20 to communicate with external devices.

[0059] The memory unit 202 stores various databases, such as the user database 211, the support system / service information database 212, and the consultation history database 213.

[0060] User database 211 is a database used for managing users.

[0061] The user database 211 contains information necessary for using the services provided by server 20, such as the user's name and attributes like gender. Further details will be provided later.

[0062] The Support System and Service Information Database 212 is a database of information sources provided to those seeking assistance. Further details will be provided later.

[0063] The consultation history database 213 is a database that manages the history of consultations made by users to the services provided by server 20. Further details will be described later.

[0064] The control unit 203 is realized when the processor 29 reads a program stored in the memory unit 202 and executes instructions contained in the program. By operating according to the program, the control unit 203 performs the functions shown as the receive control module 2041, the transmit control module 2042, the user management module 2043, the support information storage module 2044, the consultation processing module 2045, and the prompt creation module 2046.

[0065] The receive control module 2041 controls the process by which the server 20 receives signals from external devices according to a communication protocol.

[0066] The transmission control module 2042 controls the process by which the server 20 transmits signals to external devices according to a communication protocol.

[0067] The user management module 2043 is a module for managing information for each user using System 1. Specifically, the user management module 2043 accepts registration of each user's information and updates the user database 211.

[0068] The support information storage module 2044 is a program module that accepts registration of information to be provided to the person seeking advice and processes the update of the support system / service information database 212.

[0069] More specifically, the support information storage module 2044 updates the support system / service information database 212 by accepting registrations of support system information from operators of services provided by server 20 or organizations that provide support systems, or by automatically collecting support system information through crawling. For example, server 20 may also update the support system / service information database 212 by acquiring information stored on local government servers 94, information provision service servers 98, etc.

[0070] The consultation processing module 2045 is a program module that receives input of consultation details from the consultant, has the server 95 of the large-scale language model service generate an answer, and then presents that answer to the consultant.

[0071] The prompt creation module 2046 is a program module that performs processing related to the use of LLM, such as generating prompts to be sent to information processing systems such as the server 95 of the Large-Scale Language Model Service. More specifically, the prompt creation module 2046 generates prompts as follows: ·Generate a prompt by including elements such as element A, element B, element C, etc. in a predetermined instruction. Here, generating a prompt may involve combining the above-mentioned predetermined instruction and each element such as element A to generate a prompt. Also, in a given text, a prompt may be generated by replacing or inserting the predetermined instruction and each element such as element A. Generating a prompt containing the predetermined instruction and elements such as element A in this way is included in generating a prompt by including elements such as element A in the predetermined instruction as described above.

[0072] In this embodiment, the prompt creation module 2046 may generate a prompt including a predetermined instruction by using, as the above elements, any one or an arbitrary combination of the information of each item in the user database 211, the information of each item in the support system / service information database 212, and the information of each item in the consultation history database 213. In this case, the following prompts may be generated.

[0073] (1) Example 1 of a prompt # If relevant materials to the consultation content are found in the genre you want to consult Based on the profile of the consultant described later, please generate an answer to the consultation content from the knowledge in the list of relevant materials that meets the answer generation conditions.

[0074] <Profile of the consultant> {Occupation and attributes of the consultant} < / Profile of the consultant> <Consultation content> {Consultation content} < / Consultation content> <List of relevant materials> <Material name={Material name}>{Excerpt of material content}< / Material> < / List of relevant materials> <Answer generation conditions> <Condition>Within 1000 characters< / Condition> <Condition>In plain language< / Condition> <Condition>Be kind to the counselor< / Condition> < / Generation conditions of the answer> Here, at a predetermined location of the above prompt, the prompt is generated by replacing information such as the attributes specified by the user. For example, as in the screen example of FIG. 8 described later, for the user of the counselor, information of items specified about the attributes, occupation, etc. of the counselor (the counselor attribute display area 804, the related person attribute display area 810, etc. of FIG. 8 described later) may be reflected in the above {counselor's occupation and attributes} (for example, "30-year-old male" in "Part-time job").

[0075] In the above example, the content input by the user of the counselor about the consultation content (the posted content specification part 818 of FIG. 8 described later) may be reflected in the above {consultation content} (for example, input as "I am making a living from a part-time job and am worried about whether I can afford my parents' nursing care expenses. I want to know whether I can use assistance such as nursing care expenses.").

[0076] In the above example, based on the information of the attributes of the counselor and the attributes of the person related to the counselor (such as the attributes of the person receiving care), the information of the materials obtained by searching information sources such as the support system and service information database 212 may be included in the prompt for each material obtained (the above <material name={material name}>{extract of the material content}< / material> is repeated for each material).

[0077] In addition, when the type of the consultation of the consulter is a specific one (for example, in the case of a consultation of the type such as "wanting to complain"), the prompt creation module 2046 does not include the information held in the support system / service information database 212 as an element, but uses the content input by the consulter (corresponding to the above "wanting to complain", the content input by the consulter in free text) as an element, and further optionally selects or does not select the information of each item about the consulter held in the consultation history database 213 etc. as an element, and may generate a prompt including an instruction sentence for outputting a sentence to address the consulter in consideration of these elements. Also, even when searching the information sources such as the support system / service information database 212 etc. according to the attributes etc. of the consulter and no search result is obtained (when the evaluation value of the relevance with the search word is below a certain level and the relevance is low), it may also generate a prompt including an instruction sentence for outputting a sentence to address the consulter in the same way. In this case, the following prompt may be generated.

[0078] (2) Example 2 of prompt # When there is no genre to consult about or there is no material related to the consultation content Please generate an answer to the consultation content that satisfies the answer generation conditions based on the profile of the consulter described later. <Profile of the consulter> {Occupation and attributes of the consulter} < / Profile of the consulter> <Consultation content> {Consultation content} < / Consultation content> <Answer generation conditions> <Condition>Within 1000 characters< / Condition> <Condition>In kind words< / Condition> <Condition>Kindly to the consulter< / Condition> < / Answer generation conditions> The prompt creation module 2046 may also create a prompt that combines the above (1) Example 1 of prompt and (2) Example 2 of prompt.

[0079] <1.3 Configuration of Terminal 10> Figure 3 shows the configuration of terminal 10.

[0080] As shown in Figure 3, terminal 10 includes multiple antennas (antenna 111, antenna 112), communication units corresponding to each antenna (first communication unit 120, second communication unit 121), an input device 130 (including a touch-sensitive device 131), a display 132, an audio processing unit 140, a microphone 141, a speaker 142, a position information sensor 150, a camera 160, a motion sensor 170, a storage unit 180, and a control unit 190. Terminal 10 also has functions and configurations not specifically shown in Figure 3 (for example, a battery for maintaining power, a power supply circuit for controlling the supply of power from the battery to each circuit, etc.). As shown in Figure 3, each block included in terminal 10 is electrically connected by a bus or the like.

[0081] Antenna 111 radiates signals emitted by terminal 10 as radio waves. Antenna 111 also receives radio waves from space and provides the received signals to first communication unit 120.

[0082] Antenna 112 radiates signals emitted by terminal 10 as radio waves. Antenna 112 also receives radio waves from space and provides the received signals to the second communication unit 121.

[0083] The first communication unit 120 performs modulation and demodulation processing, etc., for the terminal 10 to transmit and receive signals via the antenna 111 in order to communicate with other wireless devices. The second communication unit 121 also performs modulation and demodulation processing, etc., for the terminal 10 to transmit and receive signals via the antenna 112 in order to communicate with other wireless devices. The first communication unit 120 and the second communication unit 121 are a communication module that includes a tuner, an RSSI (Received Signal Strength Indicator) calculation circuit, a CRC (Cyclic Redundancy Check) calculation circuit, a high-frequency circuit, etc. The first communication unit 120 and the second communication unit 121 perform modulation and demodulation, frequency conversion, etc., of the wireless signals transmitted and received by the terminal 10, and provide the received signal to the control unit 190.

[0084] The input device 130 has a mechanism for receiving user input operations. Specifically, the input device 130 is configured as a touchscreen and includes a touch-sensitive device 131. The touch-sensitive device 131 receives user input operations of the terminal 10. The touch-sensitive device 131 detects the user's contact position with the touch panel, for example, by using a capacitive touch panel. The touch-sensitive device 131 outputs a signal indicating the user's contact position detected by the touch panel to the control unit 190 as an input operation.

[0085] The display 132 displays data such as images, videos, and text in accordance with the control of the control unit 190. The display 132 is implemented by, for example, an LCD or an organic EL display.

[0086] The audio processing unit 140 modulates and demodulates the audio signal. The audio processing unit 140 modulates the signal received from the microphone 141 and provides the modulated signal to the control unit 190. The audio processing unit 140 also provides the audio signal to the speaker 142. The audio processing unit 140 is implemented, for example, by an audio processing processor. The microphone 141 receives an audio input and provides the audio signal corresponding to that audio input to the audio processing unit 140. The speaker 142 converts the audio signal received from the audio processing unit 140 into sound and outputs the sound to the outside of the terminal 10.

[0087] The location information sensor 150 is a sensor that detects the location of the terminal 10, and is, for example, a GPS (Global Positioning System) module. A GPS module is a receiving device used in a satellite positioning system. In a satellite positioning system, signals are received from at least three or four satellites, and the current location of the terminal 10, which is equipped with a GPS module, is detected based on the received signals.

[0088] Camera 160 is a device that receives light using a photodetector and outputs it as an image. Camera 160 is, for example, a depth camera that can detect the distance from camera 160 to the object being photographed.

[0089] The motion sensor 170 includes an acceleration sensor, an angular velocity sensor, etc., and detects the movement of the terminal 10.

[0090] The storage unit 180 is composed of, for example, flash memory and stores data and programs used by the terminal 10. The various types of information stored in the storage unit 180 will be described later.

[0091] The control unit 190 controls the operation of the terminal 10 by reading the program stored in the memory unit 180 and executing the instructions contained in the program. The control unit 190 is, for example, an application processor. By operating according to the program, the control unit 190 performs the functions of an operation reception unit 191, a transmission / reception unit 192, a data processing unit 193, a notification control unit 194, and a memory control unit 195.

[0092] The operation reception unit 191 processes input operations from the user to an input device such as a touch-sensitive device 131. Based on the coordinate information of the touch-sensitive device 131 where the user's finger or the like has made contact, the operation reception unit 191 determines the type of operation, such as whether the user's operation is a flick operation, a tap operation, or a drag (swipe) operation.

[0093] The transmitting / receiving unit 192 performs processing to enable the terminal 10 to send and receive data with an external device such as a server 20 in accordance with a communication protocol.

[0094] The data processing unit 193 performs calculations on the data received as input by the terminal 10 according to the program and outputs the calculation results to memory or other locations.

[0095] The notification control unit 194 performs the following processes: displaying the display image on the display 132, outputting sound to the speaker 142, and generating vibrations.

[0096] The memory control unit 195 controls the storage of data to the memory unit 180.

[0097] The various types of information stored by the memory unit 180 will now be explained. In a given situation, the memory unit 180 stores various types of information, such as user information 181.

[0098] User information 181 is information about a user who uses the services of server 20. User information 181 includes, for example, the user's name.

[0099] <2 Data Structure> Figure 4 shows the data structure of the user database 211. The user database 211 includes the following items: "User ID", "Name", "Affiliated Organization", "Attribute - Gender", "Attribute - Age Group", "Affiliated Department", "Position", "Job Title", "Hiring Date", "Resignation Date", "Working Days", and "Employment Type". The server 20 may also update the user database 211 by communicating with the human resources service server 96 to obtain information on each employee managed by the human resources service server 96.

[0100] The "User ID" field is information that identifies the user.

[0101] The "Name" field contains the user's name information.

[0102] The "Affiliated Organization" field is information that identifies the organization to which the user belongs.

[0103] The item "Attributes - Gender" contains information about the user's gender as an attribute.

[0104] More specifically, the item "Attributes - Gender" is the gender information provided by the user. If the user has indicated that they will not provide their gender, that information may be retained.

[0105] The item "Attributes - Age Group" contains information about the user's age group as part of their attributes.

[0106] More specifically, the item "Attributes - Age Group" contains information about the age group specified by the user. If the user specifies that they will not provide an age group, that information may be retained.

[0107] The "Department" field contains information about the department or division to which the user belongs within an organization such as a business company.

[0108] The "Job Title" field indicates the user's job title within the organization to which they belong.

[0109] The "Job Title" field contains information about the user's job title.

[0110] The "Date of Joining the Company" field indicates when the user joined an organization such as a business company (for example, when they became an employee).

[0111] The "Resignation Date" field indicates the time when the user ceased to belong to an organization such as a business company (for example, when the employment relationship ended, or when they were no longer a contractor).

[0112] The item "Working Days" is information that indicates the amount of work (number of days) a user performs over a certain period (for example, weekly).

[0113] The "Work Style" item contains information indicating the manner in which the user performs their work.

[0114] Specifically, the item "Work Style" includes the following information regarding the manner in which the user performs their work: • Go to the office • Remote work? • Is it a combination of working in the office and working remotely? (Breakdown of the number of days) In addition to the above, the user database 211 may also retain the following as user information: • Occupation. This may include, for example, unemployed, student, civil servant, company employee, non-regular employee, part-time worker, professional, self-employed, company owner, and other occupations. Figure 5 shows the data structure of the support system / service information database 212. The support system / service information database 212 includes the following items: "Content ID", "Support Theme", "Information Type", "Information Title", "Information Content", "Information Source Link", "Information Registration Date and Time", and "Access Permission Required / Not Required".

[0115] The "Content ID" field is information that identifies each piece of information provided to the user.

[0116] The item "Support Theme" contains information about the attributes of the support themes provided to the user.

[0117] The item "Support Theme" may include the following as information about the attributes of the support theme for the user: • Support related to relationships with close family members, relatives, and other people close to the person seeking advice, such as "elderly care" and "childcare." • Like "career," it primarily provides support regarding the individual's work style, lifestyle, and future activities. This service provides support for problems primarily arising from interpersonal relationships within an organization, work procedures, etc., such as "harassment." • This refers to tasks such as "customer support in a contact center," which involve assisting customers with their inquiries and resolving their problems in customer support operations. The "Information Type" field contains information about the attributes of the information that supports the user.

[0118] The "Information Type" field may include the following as attribute information of information that supports the user: • Various support programs provided by government organizations, businesses, etc. • Providing information on services that address the support theme (e.g., the operational status of childcare support services). • Providing information on knowledge and skills related to the support theme (e.g., knowledge, techniques, and know-how regarding caregiving). The "Information Title" field contains information about the name of the information provided to the user.

[0119] The item "Information Title" may include the following as the name of the information provided to the user: In the case of "nursing care," possible topics include the long-term care insurance system, the provision of nursing care services, counseling techniques, labor standards law, career counseling, and mental health. In the case of "childcare," this could include company support systems, the availability of childcare services in the community, career counseling, mental health, etc. In the case of "career," this could include career counseling (including introduction of role models), various personnel systems, mental health, etc. The item "Information Content" refers to the specific content of the information provided to the user.

[0120] The item "Information Source Links" contains information about the links that serve as sources for the information provided to the user.

[0121] The item "Source Links for Information" may include the following as information about the links that serve as the source of the information: • URL of the webpage containing information about various support programs published by administrative organizations. • A URL for accessing files that an organization, such as a business company, makes available to its members, such as employees. The item "Information Registration Date and Time" refers to the time when the information provided to the user was registered in the support system / service information database 212, or the time when the information was made public.

[0122] More specifically, the item "Information Registration Date and Time" may refer to the date and time when information on support systems, etc., was posted by administrative organizations, etc., or it may also include the date and time when this information was registered in the support system / service information database 212.

[0123] The item "Access Permission Required" indicates the status of access permissions set for the information.

[0124] More specifically, the item "Access Permission Required" includes not only items that can be viewed without access permissions, but also items for which access permissions are set so that only members of organizations such as business companies can view them.

[0125] Figure 6 shows the data structure of the consultation history database 213. The consultation history database 213 includes the following items: "Consultation Thread ID", "Consultant User ID", "Post ID", "Consultant Attributes", "Attributes of Persons Involved with the Consultant", "Type of Consultation", "Post Content", "Related Posts", "Post Date and Time", "Searchable Content", "Prompt", "LLM Response", and "User Rating".

[0126] The "Consultation Thread ID" field is information that identifies each thread containing answers to the consultations made by the user who initiated the consultation.

[0127] The item "Consultant User ID" is information that identifies the user who is seeking advice.

[0128] The item "Consultant User ID" may be associated with the item "User ID" in user database 211.

[0129] The "Post ID" field is information that identifies each post made by the user seeking advice.

[0130] The item "Consultant's Attributes" contains information about the attributes of the user who provided the consultation, as answered by the user.

[0131] The item "Consultant's Attributes" may include the following as information about the consultant's user attributes: • Gender (If the user specifies gender, possible options may include "Male," "Female," "Prefer not to answer," etc.) • Age range (If the user specifies an age range, options such as "Under 20," "20s," and "30s" may be provided to accept the user's specification.) • Occupation (For example, the user may specify from options such as "Unemployed" (meaning they intend to work but are not currently employed due to focusing on caregiving, etc.), "Student," "Civil Servant," "Company Employee," "Non-Regular Employee," "Part-time Worker," "Professional," "Self-Employed," "Company Owner," etc.) The user seeking advice may specify the user's attribute information when seeking advice, or they may refer to information such as the "Attribute - Gender" and "Attribute - Age Group" items in the user database 211. Alternatively, the attribute information specified by the user may be registered in the user database 211.

[0132] The item "Attributes of individuals the user is involved with" contains information about the attributes of individuals the user is involved with in relation to the content of the consultation.

[0133] The item "Attributes of the person the consultant is involved with" may include the following as information about the attributes or circumstances of the person the consultant user is involved with: • The relationship between the person seeking advice and the user (e.g., family, relatives; within an organization, superior, subordinate; not family, relatives, but acquaintances, etc.) • Age (for example, the age of the person receiving care) • Gender (for example, the gender of the person receiving care) • Living situation (for example, the current living situation of the person receiving care and the person seeking advice, such as "living together," "living separately (the caregiver lives in a different house)," "living separately (the person receiving care is in a facility)," "living separately (the person receiving care is hospitalized)," etc. In cases of living separately, classification according to the circumstances is also acceptable.) • In the case of long-term care, the care recipient's current level of care needs (for example, information on multiple levels such as "not certified as requiring care," "requiring support level 1," "requiring support level 2," "requiring care level 1," "requiring care level 2," "requiring care level 3," "requiring care level 4," and "requiring care level 5" may be stored). • In the case of long-term care, information regarding the care recipient's dementia may be stored at multiple stages (for example, "no dementia," "suspected dementia," "mild dementia," "moderate dementia," "severe dementia," etc.). As described above, it is also possible to retain information about the people involved with the person seeking advice, at multiple stages that represent the circumstances surrounding the consultation.

[0134] The item "Type of consultation" contains information about the type of consultation the user has specified.

[0135] The item "Type of consultation" may include the following information regarding the type of consultation you wish to have: Regarding financial care to address financial worries and anxieties Regarding the workplace environment, as a way to address interpersonal relationships and worries and anxieties about the future... • Services that can be utilized, such as long-term care insurance. • About work and the workplace • Regarding consultations about things you are worried about or things that are difficult for you. The "Post Content" field contains information about the user's inquiry, entered as free text or similar format.

[0136] The "Related Posts" item refers to information about other posts related to a user's post when that user makes a series of posts within a thread.

[0137] More specifically, the item "Related Posts" includes information that identifies the preceding post for a subsequent post, for example, when Server 20 provides a response to a preceding post made by a user, and the user then makes a subsequent post.

[0138] The "Posting Date and Time" field indicates the timing of the post made by the user seeking advice.

[0139] The item "Search Target Content" is information that identifies the information source on which the answer was generated when an answer is generated by searching for information sources such as the support system and service information database 212 in response to the user's inquiry.

[0140] The item "Searchable Content" may be associated with the item "Content ID" in the support system / service information database 212.

[0141] The item "Prompt" contains information about the prompt generated by server 95 of the large-scale language model service to provide an answer to a user's posted inquiry.

[0142] The item "LLM Response" contains information about the response output by server 95 of the Large-Scale Language Model Service to the content of the item "Prompt".

[0143] The "User Rating" item contains information about the user's evaluation of the response generated by server 95 of the large-scale language model service.

[0144] <3 operations> Figure 7 shows the processing flow when the client inputs their consultation details, and the server 95 of the large-scale language model service generates an answer based on the results of a search of various databases, or generates an answer tailored to the client's needs. The following explanation uses an example of responding to a client's consultation about caregiving.

[0145] In step S721, the consultation processing module 2045 of the server 20 outputs an operation screen to the terminal 10 that accepts input of the consultation details regarding caregiving.

[0146] Here, server 20 is configured to store various information such as knowledge and systems related to support for working in the support system and service information database 212.

[0147] In step S711, terminal 10 displays an operation screen and receives information from the user regarding the attributes of the person seeking advice, the attributes of the person the person seeking advice will be caring for, and the type of advice they wish to receive. In response to the user's actions, terminal 10 sends the information entered by the user to server 20.

[0148] In step S723, the consultation processing module 2045 of the server 20 receives input of the consultation content from the user seeking consultation. Depending on the type of consultation content, the consultation processing module 2045 determines whether to (1) search a database such as the support system / service information database 212, include the search results in the prompt, and generate an answer using the functions (LLM) provided by the server 95 of the large-scale language modeling service based on the search results, or (2) generate an answer using the functions (LLM) provided by the server 95 of the large-scale language modeling service without including the search results in the prompt, such as by not searching the database. The consultation processing module 2045 generates a prompt according to the determination result.

[0149] As described above, the consultation processing module 2045 determines, depending on the type of consultation content received from the user seeking advice, whether to generate a first prompt that includes search results obtained by searching various databases such as the support system / service information database 212, and instructions to generate an answer by referring to those search results (including instructions to summarize), or to generate a second prompt that does not include search results and includes default instructions. The prompt creation module 2046 generates either the first or second prompt according to the determination result.

[0150] More specifically, the consultation processing module 2045 receives input from the user seeking advice that expresses emotions or desires empathy, such as "I want someone to listen to my complaints" or "I want someone to listen to me," rather than searching for information such as knowledge or systems. The consultation processing module 2045 determines that the type of consultation is one in which the user wants someone to listen to their complaints or to listen to them, and then generates a second prompt. The prompt creation module 2046 generates the second prompt.

[0151] More specifically, in the processes of steps S721, S711, etc., the consultation processing module 2045 presents the consultant with an object on the operation screen that corresponds to the type of content they wish to consult about, and presents at least a first object that corresponds to consultations on knowledge, systems, etc., and a second object that corresponds to requests such as wanting someone to listen to complaints or wanting someone to listen to them. The consultation processing module 2045 accepts the specification of an object from the consultant. Depending on the type of object specified, if the first object is specified, the prompt creation module 2046 generates a first prompt, and if the second object is specified, the prompt creation module 2046 generates a second prompt.

[0152] Here, the prompt creation module 2046 may generate a second prompt, which may include the information of the person seeking advice received in the receiving step, and instructions to speak to the person seeking advice about their concerns, taking that information into consideration. In this way, the prompt creation module 2046 may generate prompts to create content such as text, images, etc., that can be spoken to according to the care situation and psychological state of the person seeking advice.

[0153] The consultation processing module 2045 receives input information about the consultant's attributes, such as age and gender, and the attributes of people involved with the consultant, and updates various databases such as the consultation history database 213 and the user database 211. The prompt creation module 2046 generates a second prompt, and may generate a second prompt that includes instructions to create content such as spoken words and images, taking into consideration the received information about the consultant's attributes and the attributes of people involved with the consultant.

[0154] The consultation processing module 2045 stores the input content of consultations of the type "I want someone to listen to my complaints" or "I want someone to listen to me" received in steps S721 and S711 in the consultation history database 213 of the storage unit 202. The consultation processing module 2045 may also summarize the input content of consultations of the type "I want someone to listen to my complaints" or "I want someone to listen to me" that has been stored in this way, and present the summarized result to the person who entered the consultation content. In this way, as the person enters the consultation content by saying things like "I want someone to listen to my complaints," the person's worries themselves are put into words and organized, which can reduce the psychological burden on the person who is worried about caregiving.

[0155] In steps S721 and S711, the consultation processing module 2045 receives input from the user regarding the content of their consultation about support systems provided by local governments, the national government, private companies, etc. When the user inputs the content of their consultation about support systems, the prompt creation module 2046 searches for information about the support systems in its storage unit and generates a first prompt that includes the search results and instructions to summarize the search results.

[0156] In this way, the consultation processing module 2045 responds to the user seeking advice by creating a prompt that includes instructions to search for information sources and have the server 95 of the large-scale language model service summarize the search results, etc., if the user seeking advice wants to know about the above-mentioned knowledge, systems, etc. On the other hand, if the user seeking advice wants someone to listen to their complaints, the consultation processing module 2045 can provide an answer that is more empathetic to the user's concerns by creating a default prompt that includes instructions to speak to the user considering their situation, rather than summarizing the search results of the information sources.

[0157] The following describes an example of the processing of the consultation processing module 2045, following a use case.

[0158] (1) Care Server 20 is intended to provide information about caregiving to those seeking advice.

[0159] The support system and service information database 212 stores information on at least one of the following support systems: the long-term care insurance system, career counseling, and mental health.

[0160] Furthermore, the support system and service information database 212 stores information related to long-term care, including at least one of the following: the status of long-term care service provision or counseling techniques.

[0161] Furthermore, the support system and service information database 212 stores information on the Labor Standards Act as part of the information on laws, ordinances, and guidelines.

[0162] In steps S721 and S711, if the consultation content received from the consultant concerns at least one of the following: information on support systems, information on long-term care, laws, ordinances, or guidelines, the consultation processing module 2045 causes the prompt creation module 2046 to generate a first prompt.

[0163] If the consultation module receives input from a caller indicating that they want someone to listen to their complaints or talk to them, the consultation processing module 2045 will have the prompt creation module 2046 generate a second prompt.

[0164] (2) Childcare Server 20 is intended to provide information about childcare to those seeking advice.

[0165] The support system and service information database 212 stores information on at least one of the following support systems: company-provided childcare support systems, support systems provided by government agencies, career counseling, and mental health services.

[0166] Furthermore, the support system and service information database 212 stores information regarding the availability of childcare support services as part of its childcare-related information.

[0167] In steps S721 and S711, if the consultation processing module 2045 receives input from the consultant regarding consultation content that concerns at least one of the following: information on support systems or information on childcare, the consultation processing module 2045 causes the prompt creation module 2046 to generate a first prompt.

[0168] If the consultation module receives input from a caller indicating that they want someone to listen to their complaints or talk to them, the consultation processing module 2045 will have the prompt creation module 2046 generate a second prompt.

[0169] (3) Career Server 20 is intended to provide the client with information about their career.

[0170] In the support system and service information database 212, information on support systems is stored that includes information on at least one of the following: company personnel systems, career counseling (including introduction of career role models), and mental health.

[0171] In steps S721 and S711, if the consultation module receives input regarding information about support systems as the consultation content from the consultant, the consultation processing module 2045 causes the prompt creation module 2046 to generate a first prompt.

[0172] If the consultation module receives input from a caller indicating that they want someone to listen to their complaints or talk to them, the consultation processing module 2045 will have the prompt creation module 2046 generate a second prompt.

[0173] In step S725, the consultation processing module 2045 of server 20 sends the generated prompt to the server 95 of the large-scale language model service, receives the response generated by the server 95 of the large-scale language model service, and updates the consultation history database 213.

[0174] In this way, the consultation processing module 2045 provides the server 95 of the large-scale language model service with a first or second prompt generated by the prompt creation module 2046, and obtains a response generated by the server 95 of the large-scale language model service in accordance with the instructions indicated in the prompt.

[0175] In step S727, the consultation processing module 2045 of server 20 presents the response generated by the server 95 of the large-scale language model service to the user making the consultation.

[0176] In step S713, terminal 10 outputs the response generated by the server 95 of the large-scale language model service.

[0177] <4. Screen Example> Figure 8 shows an example of an operation screen that receives input from the user regarding their consultation and displays the response.

[0178] The consultation operation screen 800 is an operation screen that allows users to gather information about caregiving and to seek advice on their concerns.

[0179] The consultation operation screen 800 corresponds to each process, such as step S721 in Figure 7.

[0180] Account display area 802 is the area that displays the client's account information.

[0181] In the illustrated example, the account display area 802 displays some of the information of the user logging into the service.

[0182] The client attribute display area 804 is an area that displays information about the client's attributes.

[0183] In the illustrated example, the consultant attribute display area 804 displays an operating element for receiving the consultant's attribute specification, and accepts attribute specification from the consultant user. In the illustrated example, the consultant user's attributes are set to accept specification of "gender" and "age," but it is not limited to these, and other attributes such as "occupation" may also be accepted. Furthermore, the response for the consultant user's attributes may include an option to "do not answer." The prompt creation module 2046 of the server 20 may create a prompt without including the attribute for which the user has selected "do not answer" as an element, or it may create a prompt with the attribute for which "do not answer" has been selected as an element.

[0184] The gender selection unit 806 is an operating component that accepts the gender of the person seeking advice.

[0185] In the illustrated example, the gender specification section 806 shows an example where the user specifies the attribute using a dropdown menu. Options include "Male," "Female," and "Prefer not to answer."

[0186] The generation selection unit 808 is an operating component that accepts the generation of the person seeking advice.

[0187] In the illustrated example, the generation selection unit 808 shows an example where the user specifies the generation using a pull-down menu. The options could be set to select from predetermined age ranges, such as "Teens" and "Twenties," or there could be an option such as "Do not answer."

[0188] The related party attribute display area 810 is an area that displays information about the related parties who will be providing care to the person seeking advice.

[0189] In the illustrated example, the related party attribute display area 810 displays an operating component for receiving a specification from the consultant regarding the relationship between the consultant and the person receiving care, and accepts the specification from the user. The following are possible options for accepting the specification of the relationship between the consultant and the person receiving care. • "Parents (including in-laws)", "Spouse", "Other relatives", "Children", "Acquaintances" In addition to the examples shown, the following specifications may also be accepted as attributes of the person receiving care. • Age of the person receiving care • The gender of the person receiving care (e.g., "male," "female," "both male and female") • Current living situation of the person receiving care ("Living together", "Living separately (the person receiving care lives in a different house from the person seeking advice)", "Living separately (the person receiving care is in a care facility)", "Living separately {the person receiving care is hospitalized}") • The current level of care required by the person receiving care (multiple options may be available, such as "Not certified as requiring care," "Requires support level 1," "Requires support level 2," "Requires care level 1," "Requires care level 2," "Requires care level 3," "Requires care level 4," or "Requires care level 5.") • Dementia of the person receiving care (Multiple options may be available, such as "No dementia," "Suspected dementia," "Mild dementia," "Moderate dementia," and "Severe dementia.") The relationship designation unit 812 is an operating component that accepts the designation of the relationship between the person seeking advice and the person providing care.

[0190] In the illustrated example, the relationship specification unit 812 allows the user to select the relationship between the person providing care and the user using a pull-down menu, and includes options for relationships with relatives, such as "parent (including spouse's parents)."

[0191] The consultation content display area 814 is an area that displays the content of the consultation by the person seeking advice.

[0192] In the illustrated example, the consultation content display area 814 displays an operating element for the user to specify the content of their consultation and accepts input of the consultation content.

[0193] The type designation section 816 is an operating member that accepts the designation of the type of content the user wishes to consult about.

[0194] In the illustrated example, the type designation unit 816 accepts designations for broad categories of the content to be discussed, such as "financial care" and "work environment." Depending on the client's concerns, the unit may also accept designations for categories corresponding to individual concerns, such as financial worries, future worries, relationship worries, and health worries. Alternatively, the unit may present options for the type of content to be discussed, as shown below. • Services that can be utilized, such as long-term care insurance. • About work and the workplace • Consultation about things you are worried about or things that are difficult for you. The type designation unit 816 displays the following as types of consultations related to caregiving and accepts designations from the user seeking consultation.

[0195] (1) To respond to inquiries from people who want to know more about the system. "Long-term care insurance system"

[0196] (2) To respond to inquiries from people who want to use specific services. "Care Service Information" (Contents of services provided, locations where services are provided, etc.)

[0197] (3) To answer questions about specific techniques and approaches to caregiving. "Counseling Techniques"

[0198] (4) To respond to inquiries from people who want to know about laws, guidelines, etc. "Labor Standards Act"

[0199] (5) A service that provides advice on future concerns and health concerns. "Career counseling" "Mental health"

[0200] (6) Services that respond to requests for advice such as wanting to vent about worries, complain, or just wanting someone to listen. "I just want to vent." (Before discussing specific solutions or systems, the consultation may be considered complete once the person expresses and verbalizes their inner thoughts and worries, or it may lead to the next step of presenting solutions.) In the example shown, the user seeking advice has selected "I want to vent," and this is highlighted to distinguish it from the other options.

[0201] The post content specification unit 818 is an operating component that accepts the specification of the consultation content in free text format.

[0202] In the illustrated example, the submission content specification section 818 accepts input in text format, but it may also accept input via voice input or other means and transcribe it.

[0203] The consultation posting operation unit 820 is an operation component that accepts the operation of a user who is seeking advice to post the content of the consultation they have specified.

[0204] The consultation posting operation unit 820 corresponds to the processing shown in step S711 in Figure 7. In response to user operations on the consultation posting operation unit 820, the terminal 10 transmits information about the consultation content specified by the user in the relationship specification unit 812, type specification unit 816, posting content specification unit 818, etc., to the server 20.

[0205] Furthermore, the server 20 performs the following processes, such as steps S723, S725, and S727, based on the information about the consultation content specified by these users.

[0206] The answer display area 822 is the area that displays the answer corresponding to the consultation content.

[0207] In the illustrated example, the answer display area 822 displays the input content from the user seeking advice in a right-aligned manner (for example, the input content display area 826 described later) while the answer created by the server 95 of the large-scale language model service is left-aligned (for example, the system information display area 824 and the LLM answer display area 828 described later), thereby distinguishing between the input consultation content and the corresponding answer.

[0208] The system information display area 824 is an area that displays a summary of information about systems that can support the user, obtained by searching for information sources in response to the user's consultation content.

[0209] In the illustrated example, the system information display area 824, in response to a user requesting information about the system, searches the information registered in the support system / service information database 212 and displays the answer created by the large-scale language model service server 95 in response to a prompt to summarize the search results. It also displays links to the information sources on which the answer was based (such as "Reference (1)" in the illustrated example), making it easy for the user to refer to more detailed information.

[0210] The input content display area 826 is an area that displays the results of a user requesting advice who has entered specific content (post content specification section 818) corresponding to the type of advice request being "I want to vent."

[0211] In the illustrated example, the input content display area 826 shows that "I want to complain" has been selected as the type of consultation. In the case of this type of consultation, as explained in step S723, the server 20 creates a prompt that includes the content entered by the consultant in the posting content specification unit 818, information about the consultant's attributes (gender specification unit 806, generation specification unit 808), information about the person the consultant is caring for (relationship specification unit 812), and instructions to speak to the consultant, taking this information into consideration. The server 20 sends the created prompt to the server 95 of the large-scale language model service, and receives a response from the server 95 of the large-scale language model service that includes the content of the message created in response to the complaint.

[0212] The LLM response display area 828 is an area that displays responses, including the content of responses to complaints, created by the server 95 of the Large-Scale Language Model Service.

[0213] The summary confirmation operation unit 830 is an operation component that accepts the operation of the user to confirm the result of summarizing the information accumulated when the user inputs the content of their consultation, such as "I want to vent."

[0214] In the illustrated example, the summary confirmation operation unit 830 searches for information stored as the user's consultation content (consultation history database) in response to the user's operation, and the server 20 creates a prompt that includes the search results and an instruction to summarize the search results. The server 20 sends the created prompt to the server 95 of the large-scale language model service. This allows the user to easily confirm the consultation content (including writing down complaints) that they have accumulated by entering information on the consultation operation screen 800, and for example, the user's worries may be visualized, and the causes of their worries may be put into words.

[0215] The consultation reservation operation unit 832 is an operating component that accepts the operation of a person making a reservation for a consultation.

[0216] In the illustrated example, the consultation reservation operation unit 832 makes a reservation for a consultation with an organization that supports people who provide care while working, in response to the user's operation. At this time, the information entered by the user through the consultation operation screen 800 (such as the user's attributes, relationship with the caregiver, information on systems researched, the content of the complaint entered, and a summary of the problem processed in response to the operation of the summary confirmation operation unit 830) may be transmitted to the support organization. This allows the support organization to increase the resolution of the user's information and facilitate smoother consultations with the user.

[0217] <Variation> In addition to the embodiments described above, the following may also be used.

[0218] (1) Content of consultations The above embodiment mainly describes examples of providing consultations on topics such as "elderly care," "childcare," and "career."

[0219] In addition, for example, when a business provides products or services to customers, for inquiries regarding how to use or configure the products or services, etc., for which items are available in help pages, manuals, etc., the business may search for information sources that address the questions and create prompts that include the search results and instructions to explain and summarize the search results for the customer.

[0220] On the other hand, if a customer wants to express dissatisfaction with a product or service, instead of having them search for information sources (or including search results in the prompt), it may be better to create a prompt that uses empathetic language, such as by including the customer's situation in the prompt.

[0221] As described above, depending on the client's problem, simply offering solutions may not be sufficient. For example, by organizing the client's messages, which are expressed based on their emotions, it may be possible to select an appropriate solution.

[0222] Thus, as a technical means to solve the problem, "a program is provided that operates a computer equipped with one or more computer processors. The memory unit is configured to store various information for providing information to the consultant. The program causes one or more computer processors to perform the following steps: receive input of consultation content from the consultant; determine whether to generate a first prompt containing search results from the various information stored in the memory unit, or a second prompt containing a default instruction without including the search results, depending on the type of consultation content received; generate either the first or second prompt according to the determination result; provide the generated first or second prompt to a large-scale language model to obtain an answer generated by the large-scale language model according to the instructions shown in the prompts; and present the answer generated by the large-scale language model to the consultant."

[0223] This allows for responses that refer to various sources, depending on the type of consultation received, as well as responses that are empathetic and supportive of the person seeking advice, thereby further facilitating the resolution of the problems the person is facing.

[0224] (2) How to create a prompt In addition to the embodiments described above, there are systems that use large-scale language models to perform natural language processing such as question and answer, but one of the problems that the invention aims to solve is that it is not easy to create prompts that are suitable for consultations depending on the type of consultation content that the person seeking advice wants to discuss.

[0225] As a means to solve the above problems, as described in the above embodiment, "a program is provided for operating a computer equipped with one or more computer processors. The memory unit is configured to store various information for providing information to a consultant. The program causes one or more computer processors to perform the steps of: receiving input of consultation content from a consultant; generating a first prompt that includes search results obtained by searching various information stored in the memory unit according to the type of consultation content received as input, and including instructions to generate an answer by referring to the search results; or generating a second prompt that does not include search results but includes a default instruction; and generating either the first or second prompt according to the determination result."

[0226] This makes it even easier to create appropriate prompts based on the content of the caller's inquiry.

[0227] (3) Classification of the results of summarizing the content of the consultations of the people seeking advice In the description of the above embodiment, the summary confirmation operation unit 830 and the like described an example in which information accumulated by the user inputting the consultation content is summarized.

[0228] Here, the server 20 may perform a process to associate the type of consultation with the result of summarizing the content of the consultation as described above.

[0229] For example, the server may display an operating component that accepts the specification of the type of consultation, as shown in the type specification unit 816, while presenting the client with a summary of the client's consultation content, thereby accepting the client's specification of the type of consultation content. The server 20 may then store the type of consultation and the summary of the consultation content entered in this manner in the consultation history database 213, etc., and present the client with an answer obtained by searching for information sources through processes such as steps S723 and S725.

[0230] Alternatively, server 20 may create a prompt that includes a summary of the consultant's consultation content, information on the type of consultation, and an instruction to determine the type of consultation, and send this prompt to server 95 of the large-scale language model service, causing server 95 of the large-scale language model service to determine the type of consultation based on the summary of the consultation content. After determining the type of consultation in this way, the server may present the consultant with an answer that has searched for information sources through processes such as steps S723 and S725 as described above.

[0231] (4) A prompt may be generated that includes both the content of the first prompt and the content of the second prompt, and the server 95 of the large-scale language model service may be made to generate an answer.

[0232] For example, the prompt may include information such as the "attributes of the person seeking advice" and "attributes of people involved with the person seeking advice" items from the consultation history database 213, as well as information specified in the consultation attribute display area 804 and related party attribute display area 810 in Figure 8. If the type of consultation is "I want to vent" (type specification unit 816), the prompt may include an instruction to generate empathetic words for the person seeking advice, using these attributes of the person seeking advice and the content entered in the post content specification unit 818 as elements, and an instruction to summarize the systems and services available to the person seeking advice from information sources such as the support system / service information database 212, using the above-mentioned information on the attributes of the person seeking advice and the information on the attributes of people involved with the person seeking advice in caregiving, etc., as elements. The server 95 of the large-scale language model service generates a response based on this prompt, thereby providing empathetic words and useful information to the person seeking advice.

[0233] For example, if the user enters "unemployed" as their occupation and "I quit my job to dedicate myself to caregiving" as the free text for their consultation (specified post content section 818), the server 95 of the large-scale language model service may generate an answer such as, "Balancing caregiving and work is a very difficult challenge. I understand how you feel. You are currently focusing on caregiving, but if you are considering returning to work in the future, you may be able to utilize systems such as caregiving leave or caregiving time off. Telecommuting and flexible working hours may also be helpful in balancing work and caregiving." By presenting the results of the information search and answer generation, the service can provide the user with more detailed information (specific systems, etc.) on systems that are more suitable for their situation (re-employment support, vocational training programs), the use of long-term care insurance services, financial support systems such as caregiving allowances, recommendations to consult with a care manager instead of bearing the burden alone, and suggestions that they take care of their own health and mental well-being, given that caregiving is long-term, and that joining a caregiver support group is also an option.

[0234] As described above, regarding the challenge of not being able to easily create prompts that generate answers that are appropriate to the content of the consultation from the person seeking advice, as a technical means to solve this problem, "A program is provided for operating a computer having one or more computer processors. The memory unit is configured to store various information for providing information to a consultant. The program may also be configured to cause one or more computer processors to execute the following steps: receive input of consultation content from a consultant; determine whether the content of the prompt should include instructions that generate an answer by referencing the search results obtained by searching various information stored in the memory unit, depending on the type of consultation content received as input; include default instructions without including search results; or include default instructions that generate text by referencing the consultation content rather than the search results based on the input consultation content, and default instructions that output an answer by referencing the search results obtained by searching for information sources based on the input consultation content, and then generate a prompt according to the determination result."

[0235] A network consists of various mobile communication systems, such as the internet, LANs, and wireless base stations. For example, a network includes 3G, 4G, and 5G mobile communication systems, LTE (Long Term Evolution), and wireless networks that can connect to the internet via designated access points (e.g., Wi-Fi®). When connecting wirelessly, communication protocols include, for example, Z-Wave®, ZigBee®, and Bluetooth®. When connecting via a wired connection, the network also includes connections made directly via USB (Universal Serial Bus) cables, etc.

[0236] Furthermore, by distributing all or part of each hardware configuration across multiple computers and connecting them to each other via a network, computer 90 can be virtually realized. Thus, the concept of a computer includes not only computers housed in a single enclosure or case, but also virtualized computer systems.

[0237] A database, specifically a relational database, is used to manage and link together tabular data sets called masters, which are structurally defined by rows and columns. In a database, tables are called tables, masters are called masters, the columns of tables are called columns, and the rows of tables are called records. In a relational database, relationships can be established and linked between tables and masters.

[0238] Typically, each table and master has a primary key column to uniquely identify records, but setting a primary key column is not mandatory. The control unit can instruct the processor 901 to add, delete, or update records in specific tables and masters stored in the memory unit, according to various programs.

[0239] Furthermore, by storing data, various programs, and various databases in the memory unit, the information processing device and information processing system related to this disclosure can be considered to have been manufactured.

[0240] Furthermore, the databases and masters in this disclosure may include any data structures (lists, dictionaries, associative arrays, objects, etc.) in which information is structurally defined. Data structures also include data that can be considered as data structures by combining data with functions, classes, methods, etc., written in any programming language.

[0241] Furthermore, each of the above-mentioned configurations, functions, processing units, processing means, etc., may be implemented in hardware, either partially or entirely, by designing them as integrated circuits, for example. The present invention can also be implemented by software program code that realizes the functions of the embodiment. In this case, a storage medium on which the program code is recorded is provided to a computer, and the processor of that computer reads the program code stored in the storage medium. In this case, the program code read from the storage medium itself realizes the functions of the embodiment described above, and the program code itself and the storage medium on which it is stored constitute the present invention. Examples of storage media used to supply such program code include flexible disks, CD-ROMs, DVD-ROMs, hard disks, SSDs, optical disks, magneto-optical disks, CD-Rs, magnetic tapes, non-volatile memory cards, ROMs, and the like.

[0242] Furthermore, the program code that implements the functions described in this embodiment can be implemented in a wide range of programming or scripting languages, such as assembler, C / C++, Perl, Shell, PHP, and Java (registered trademark).

[0243] Furthermore, the program code for the software that implements the functions of the embodiment may be distributed via a network and stored in a storage means such as a computer's hard disk or memory, or in a storage medium such as a CD-RW or CD-R, and the computer's processor may read and execute the program code stored in the storage means or storage medium.

[0244] The functions realized by the components described herein may be implemented in a circuit or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), CPUs (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to realize the functions described herein. A processor is considered to be a circuit or processing circuitry, including transistors and other circuits. A processor may be a programmed processor that executes a program stored in memory.

[0245] In this specification, circuitry, unit, and means are hardware programmed to perform or execute the functions described herein. Such hardware may be any hardware disclosed herein, or any hardware known to be programmed to perform or execute the functions described herein.

[0246] If the hardware is a processor that is considered to be a type of circuitry, then the circuitry, means, or unit is a combination of hardware and software used to constitute the hardware and / or processor.

[0247] While several embodiments of this disclosure have been described above, these embodiments can be implemented in a variety of other forms, and various omissions, substitutions, and modifications are permitted without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents.

[0248] (Note) The details described in each of the above embodiments are noted below.

[0249] (Appendix 1) A program for operating a computer including one or more computer processors, configured to store various information such as knowledge and systems related to support for work in a storage unit, the program comprising steps of causing one or more computer processors to: receive an input of consultation content from a consulter; determine whether to generate a first prompt including a search result obtained by searching various information stored in the storage unit according to the type of the received consultation content, or to generate a second prompt including a predetermined instruction text without including the search result, and generate either the first or the second prompt according to the determination result; obtain an answer generated by a large language model according to an instruction indicated in the prompt by providing the generated first or second prompt to the large language model; and present the answer generated by the large language model to the consulter.

[0250] (Appendix 2) In the receiving step, an input of consultation content of a type such as wanting to hear a complaint or wanting to hear a story is received, and in the generating step, when the type of the consultation content is consultation content of a type such as wanting to hear a complaint or wanting to hear a story, it is determined to generate a second prompt, and the second prompt is generated. The program according to Appendix 1.

[0251] (Appendix 3) In the receiving step, presenting an object corresponding to the type of content the consulter wants to consult to the consulter, including at least a first object corresponding to consultation on knowledge, systems, etc., and a second object corresponding to a type such as wanting to hear a complaint or wanting to hear a story, receiving a designation of an object from the consulter, and in the generating step, generating a first prompt when the first object is designated according to the type corresponding to the designated object, and generating a second prompt when the second object is designated. The program according to Appendix 2.

[0252] (Note 4) The program described in Appendix 3 generates a second prompt in the generation step, including the information of the person seeking advice received in the receiving step and instructions to offer words of encouragement regarding the person's concerns.

[0253] (Note 5) The program described in Appendix 4 accepts, in the receiving step, information about the consultant's attributes such as age and gender, and information about the attributes of people involved with the consultant as part of the consultant's consultation content, and in the generating step, generates a second prompt, which includes the received information about the consultant's attributes and the attributes of people involved with the consultant to generate the second prompt.

[0254] (Note 6) In the receiving step, the program stores the input content of the consultation, such as the request to have someone listen to a complaint or to talk to someone, in a memory unit, and the program further causes one or more computer processors to perform the steps of summarizing the input content of the consultation, such as the request to have someone listen to a complaint or to talk to someone, stored in the memory unit, and presenting the summarized results to the person making the request, as described in any of the appendices 3 to 5.

[0255] (Note 7) A program as described in any of Appendix 2 to 6, which, in the receiving step, accepts input of the content of a consultation regarding support systems provided by local governments, the national government, private companies, etc., and in the generating step, if the content of the consultation is about support systems, generates a first prompt that includes the search results obtained by retrieving information on support systems from the memory unit and an instruction to summarize the search results.

[0256] (Note 8) The computer provides care-related information to the person seeking advice, and the memory unit stores information on support systems, specifically information on the long-term care insurance system, career counseling, and mental health; the memory unit stores information on care service provision status and counseling techniques; the memory unit stores information on laws, ordinances, and guidelines, specifically information on the Labor Standards Act; and in the receiving step, if the computer receives input from the person seeking advice regarding information on support systems, care-related information, laws, ordinances, and guidelines, it generates a first prompt in the generation step; and in the receiving step, if the computer receives input from the person seeking advice regarding wanting someone to listen to their complaints or to talk to them, it generates a second prompt in the generation step, as described in any of the appendices 3 to 7.

[0257] (Note 9) The computer provides childcare-related information to the person seeking advice, and stores in its memory information information at least one of the following: childcare support systems at companies, support systems provided by government agencies, career counseling, and mental health; and stores in its memory information information the availability of childcare support services; and in the receiving step, if the computer receives input from the person seeking advice regarding at least one of the following: information on support systems or information on childcare, it generates a first prompt in the generation step; and in the receiving step, if the computer receives input from the person seeking advice such as "I want someone to listen to my complaints" or "I want someone to listen to me," it generates a second prompt in the generation step, as described in any of the appendices 3 to 8.

[0258] (Note 10) The computer provides career-related information to the person seeking advice, and stores in its memory information information on support systems, including at least one of the following: company personnel systems, career counseling (including introduction of career role models), and mental health. In the receiving step, if the computer receives input from the person seeking advice regarding support system information, it generates a first prompt in the generation step. In the receiving step, if the computer receives input from the person seeking advice regarding wanting someone to listen to their complaints or to talk to them, it generates a second prompt in the generation step. This is a program as described in any of the appendices 3 to 9.

[0259] (Note 11) A method executed by a computer having one or more computer processors, wherein a memory unit is configured to store various information such as knowledge and systems related to support for work, and the program performs the following steps: one or more computer processors receive input of consultation content from a client; depending on the type of consultation content received, generate a first prompt including search results from various information stored in the memory unit, or generate a second prompt including a default instruction without including search results, and generate either the first or second prompt according to the determination result; provide the generated first or second prompt to a large-scale language model to obtain an answer generated by the large-scale language model according to the instructions shown in the prompts; and present the answer generated by the large-scale language model to the client.

[0260] (Note 12) Information processing device, configured in a memory unit to store various information such as knowledge and systems related to support for work, wherein the control unit of the information processing device performs the following steps: receiving input of consultation content from a consultant; determining whether to generate a first prompt including search results from various information stored in the memory unit, or a second prompt including a predetermined instruction without including search results, depending on the type of consultation content received; generating either the first or second prompt according to the determination result; obtaining an answer generated by the large-scale language model in accordance with the instructions shown in the prompt by providing the generated first or second prompt to a large-scale language model; and presenting the answer generated by the large-scale language model to the consultant.

Claims

1. A program for operating a computer having one or more computer processors, The memory unit is configured to store various information such as knowledge and systems related to support for working. The program is configured on one or more computer processors. Steps include receiving input from the person seeking advice, Depending on the type of consultation content received as input, the system determines whether to generate a first prompt including search results obtained by searching the various information stored in the storage unit, or to generate a second prompt including a default instruction without including the search results, and generates either the first or second prompt according to the determination result. The steps include: providing the generated first or second prompt to the large language model to obtain a response generated by the large language model in response to the instructions indicated in the prompt; A program that performs the steps of: presenting the answer generated by the large-scale language model to the person seeking advice.

2. In the aforementioned acceptance step, we accept input of consultation content of the type that is to ask for someone to listen to your complaints or to talk to you. The program according to claim 1, wherein in the generation step, if the type of consultation content is the type of consultation content of wanting someone to listen to a complaint or wanting someone to talk, the program determines to generate the second prompt and generates the second prompt.

3. In the aforementioned acceptance step, This involves presenting the person seeking advice with objects corresponding to the type of issue they wish to discuss, including at least two objects: a first object for inquiries about knowledge, systems, etc., and a second object for inquiries about wanting someone to listen to complaints or talk to them. The aforementioned client has provided the specification of the aforementioned object. The program according to claim 2, wherein in the generation step, depending on the type corresponding to the specified object, a first prompt is generated if a first object is specified, and a second prompt is generated if a second object is specified.

4. The program according to claim 3, wherein in the generation step, the second prompt is generated, and the second prompt is generated including the information of the person seeking advice received in the receiving step and an instruction to offer words of advice regarding the person seeking advice.

5. In the aforementioned acceptance step, the client's information, including their age, gender, and other attributes, as well as information about the attributes of the person the client is involved with, is accepted as part of the client's consultation details. The program according to claim 4, wherein, in the generation step, the second prompt is generated, and the second prompt is generated including the attribute information of the person who received the consultation and the attribute information of the person the consultation is involved with.

6. In the aforementioned receiving step, the input content of the consultation request, such as wanting someone to listen to a complaint or wanting someone to talk to, is stored in the memory unit. The program further provides the following to one or more computer processors: A step of summarizing the input content of consultations of the type that are stored in the memory unit, such as complaints or requests for someone to listen to one's story, The program according to claim 3, which performs the step of presenting the summarized results of the input content to the person seeking advice.

7. In the aforementioned application step, we accept input regarding the content of inquiries about support programs provided by local governments, the national government, private organizations, etc. The program according to claim 2, wherein in the step of generating, if the content of the consultation concerns the support system, it generates a first prompt that includes search results obtained by searching for information on the support system from the storage unit and an instruction to summarize the search results.

8. The aforementioned computer provides information about caregiving to those seeking advice. The aforementioned memory unit stores information on support systems, specifically information on at least one of the following: the long-term care insurance system, career counseling, or mental health. The memory unit is configured to store information related to caregiving, including at least one of the following: the status of caregiving services provided or counseling techniques. The aforementioned memory unit stores information on the Labor Standards Act as information on laws, ordinances, and guidelines. In the receiving step, if the consultation content received from the consultant concerns at least one of the following: information on the support system, information on long-term care, or information on laws, ordinances, or guidelines, then in the generation step, the first prompt is generated. The program according to claim 3, wherein, in the receiving step, the program receives input from the person seeking advice that the person seeking advice wants someone to listen to their complaints or to talk to them, and in the generating step, the program generates the second prompt.

9. The aforementioned computer provides information about childcare to those seeking advice. The aforementioned memory unit stores information on support systems, including information on at least one of the following: company childcare support systems, support systems provided by government agencies, career counseling, and mental health. The aforementioned memory unit stores information related to childcare, including the status of the provision of childcare support services. In the receiving step, if the consultation content received from the consultant concerns at least one of the information on the support system or the information on childcare, the first prompt is generated in the generation step. The program according to claim 3, wherein, in the receiving step, the program receives input from the person seeking advice that the person seeking advice wants someone to listen to their complaints or to talk to them, and in the generating step, the program generates the second prompt.

10. The aforementioned computer provides the client with information regarding their career. The aforementioned memory unit stores information on support systems, including at least one of the following: the company's personnel system, career counseling (including the introduction of career role models), and mental health. In the receiving step, if the content of the consultation from the person seeking advice is an inquiry about information on the support system, then in the generation step, the first prompt is generated. The program according to claim 3, wherein, in the receiving step, the program receives input from the person seeking advice that the person seeking advice wants someone to listen to their complaints or to talk to them, and in the generating step, the program generates the second prompt.

11. A method performed by a computer having one or more computer processors, The memory unit is configured to store various information such as knowledge and systems related to support for working. The program is performed by the one or more computer processors, Steps include receiving input from the person seeking advice, Depending on the type of consultation content received as input, the system determines whether to generate a first prompt including search results obtained by searching the various information stored in the storage unit, or to generate a second prompt including a default instruction without including the search results, and generates either the first or second prompt according to the determination result. The steps include: providing the generated first or second prompt to the large language model to obtain a response generated by the large language model in response to the instructions indicated in the prompt; A method for performing the steps of: presenting the answer generated by the large-scale language model to the person seeking advice.

12. An information processing device, The memory unit is configured to store various information such as knowledge and systems related to support for working. The control unit of the information processing device, Steps include receiving input from the person seeking advice, Depending on the type of consultation content received as input, the system determines whether to generate a first prompt including search results obtained by searching the various information stored in the storage unit, or to generate a second prompt including a default instruction without including the search results, and generates either the first or second prompt according to the determination result. The steps include: providing the generated first or second prompt to the large language model to obtain a response generated by the large language model in response to the instructions indicated in the prompt; An information processing device that performs the steps of: presenting the answer generated by the large-scale language model to the person seeking advice.

Images